|UFDC Home||myUFDC Home | Help|
This item has the following downloads:
1 SOCIO ECONOMIC CHANGE IN THE TRANSITION FROM PATRON CLIENT TO SOCIAL MOVEMENT NETWORKS IN BRAZILIAN AMAZONIA By MASON CLAY MATHEWS A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL F ULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2011
2 2011 Mason Clay Mathews
3 To my Mom, who has supported me in every way possible over the years, through all the difficulties that graduate school and life present.
4 ACKNOWLEDGMENTS This dissertation was a collaborative effort in every phase. My parents have supported me in every way possible during all the stages of developing this research. I would not have been able to achieve my goal of f inishing this dissertation without the support of my dissertation chair, co chair, and committee. My advisor Marianne Schmink went above and beyond the call of duty, reading proposals, providing emotional support, and guiding me through the difficult proc ess of writing this dissertation. All of my committee members, Chris McCarty, Steve Perz, and Eric Keys coached and shepherded me through the difficult process of developing my research ideas, writing proposals, collecting and analyzing data, and writing the final document. Without their guidance and assistance I could not have accomplished this project. In so many ways this dissertation is also the result of the generosity of dozens of Brazilians who gave me information, told me their personal stories, and trusted me to retell their stories in this dissertation. My host country agency in Brazil, the Nova Cartografia Social da Amaznia also helped me immeasurably during the entire research process. Alfredo Wagner Berno de Almeida and all of the research staff at the Nova Cartografia played a role in bringing this research to fruition. I must also thank the many wonderful Brazilians who helped me get through the difficulties of conducting research and the incredible hospitality and friendship they provid ed. The National Science Foundation, Inter American Foundation, and the Working Forests in the Tropics IGERT fellowship provided crucial financial support without which I would have been unable to complete this dissertation.
5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ .......... 10 LIST OF FIGURES ................................ ................................ ................................ ........ 12 LIST OF ABBREVIATIONS ................................ ................................ ........................... 13 ABSTRACT ................................ ................................ ................................ ................... 16 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 18 2 PATRON CLIENT RELATIONSHIPS ................................ ................................ ..... 22 Theory and Patron Client Relationships ................................ ................................ 22 Research Question and Patron Cl ient Theory ................................ ........................ 24 What are Patron Client Relationships? ................................ ................................ ... 26 Conditions for Patron Client Relationships ................................ ............................. 26 Resource Scarcity and Inequality ................................ ................................ ........... 26 Strong State / Weak State ................................ ................................ ...................... 27 Economics and Isolation ................................ ................................ ......................... 28 Patron Client Profiles ................................ ................................ .............................. 30 Patron Client Processes ................................ ................................ ......................... 31 Resource Control ................................ ................................ ................................ .... 32 Generalized Exchange ................................ ................................ ............................ 32 Expectations and Thresholds ................................ ................................ .................. 34 Alternatives and Change ................................ ................................ ......................... 36 3 "TRADITIONAL" PATRONAGE IN THE SOCIOECONOMIC EVOLUTION OF LBREA ................................ ................................ ................................ .................. 38 Exchange, Expectations, and P atron Client Ties ................................ .................... 39 Exploration and Extractivism on the Purus River ................................ .................... 40 Early Exploration Descimentos and Drogas do Serto ................................ ........ 41 Aviamento : Credit, Debt, and Exchange ................................ ................................ 42 Expansion up the Purus ................................ ................................ .......................... 44 The Rubber Boom and the Settling of the Purus ................................ .................... 45 The Seringal Rubber Production System ................................ ................................ 47 The Culture of the Barraco ................................ ................................ .................... 49 The Regime do Toco ................................ ................................ .............................. 51 Patron to Client Exchanges in the Regime do Toco ................................ ......... 54 Client to Patron Exchanges in the Regime do Toco ................................ ......... 55
6 Resource Control in the Regime do Toco ................................ ......................... 57 Boom Bust Cycles in Lbrea ................................ ................................ ................... 61 From the Regime do Toco to Arrendamento ................................ ........................... 62 Patron to Client Exchanges in the Rent System ................................ ............... 63 Client to Patron Exchanges in the Rent System ................................ ............... 63 Resource Control in the Rent System ................................ .............................. 65 Subsidized Rubber Production ................................ ................................ ............... 68 The End of Subsidies and the End of the Seringal ................................ ................. 69 From Arrendamento to Seringal Abandonment ................................ ...................... 70 Patron to Client Exchanges in Seringal Abandonment ................................ ..... 71 Client to Patron Exchange in Seringal Abandonment ................................ ....... 71 From Abandonment to State Control ................................ ................................ ...... 75 State Control and Patron Client Ties ................................ ................................ ...... 78 Chapter Conclusio ns ................................ ................................ ............................... 80 4 CONDITIONS FOR PATRON CLIENT RELATIONSHIPS ON THE PURUS RIVER ................................ ................................ ................................ ..................... 89 Exclusion and Patron Client Ties: Aviamento on the Puru s Today ......................... 89 Production Systems, Seasonality, and Aviamento ................................ .................. 90 Production Calendars, Seasonality, and Credit ................................ ...................... 91 Fishing in the Vrzea ................................ ................................ .............................. 97 Logging Season in the Vrzea ................................ ................................ ................ 98 Resource Distribution, Qu antity, Quality, and Credit ................................ ............... 99 Rubber ................................ ................................ ................................ .............. 99 Brazil Nuts ................................ ................................ ................................ ...... 101 Andirob a and Copaiba Oils ................................ ................................ ............. 103 Variation in Resource Access by Household ................................ ........................ 104 Floods of Fortune and Famine ................................ ................................ .............. 104 Flooded Forests, Livestock, and Economic Stability ................................ ............. 105 Health, Vulnerability, and Credit ................................ ................................ ........... 108 Hygiene ................................ ................................ ................................ .......... 108 Malaria ................................ ................................ ................................ ............ 109 Filariasis ................................ ................................ ................................ ......... 111 Hepatitis ................................ ................................ ................................ ......... 112 Health and Credit ................................ ................................ ................................ .. 113 Education, Literacy, and Patronage ................................ ................................ ...... 115 Exclusion f rom "Formal" Credit and Dependence on Aviamento .......................... 116 Geographic Distance ................................ ................................ ...................... 117 Citizenship Documents ................................ ................................ ................... 118 Chapter Conclusions ................................ ................................ ............................. 120
7 5 SOCIAL CHANGE, COMMUNITY ORGANIZATION, AND "NEW" SOCIAL NETWORKS: FROM TWO VILLAGES TO TWO EXTRACTIVE RESERVES, NETWORKING AND ORGANIZING OUT OF CRISIS ................................ .......... 122 Community Level Currents ................................ ................................ ................... 125 Vila Limeira ................................ ................................ ................................ ........... 125 Negative Social Currents Resource Conflict ................................ ...................... 135 Fishing Conflicts ................................ ................................ ............................. 135 Traditional Patrons and Resource Use ................................ ........................... 137 Negative Economic Currents ................................ ................................ ................ 140 Land Tenure and Indigenous Lands ................................ ................................ ..... 141 N etworking, Social Capital, and Positive Social Currents ................................ ..... 147 Patron Conflicts and the Catholic Church ................................ ....................... 148 Christian Base Communit ies ................................ ................................ .......... 149 Rural Workers' Union ................................ ................................ ..................... 151 Pastoral Land Commission ................................ ................................ ............. 155 Fi shing Conflicts and Social Movements ................................ ........................ 156 Indigenous Missionary Council (CIMI) ................................ ............................ 158 Children's Pastoral Pastoral da Crian a ................................ ...................... 162 Networking with State Agencies to Resolve Agricultural Problems ....................... 163 Community Leaders ................................ ................................ ....................... 164 Municipal Politicians ................................ ................................ ....................... 166 Social Movement and State Agency Networks ................................ ............... 168 Evangelical Church Networks ................................ ................................ ......... 171 Accessing Resources through Community Associations ................................ 172 Vila Vitria ................................ ................................ ................................ ...... 174 Problems and Challenges ................................ ................................ .............. 182 Extractive Reserves ................................ ................................ .............................. 184 Vila Limeira and the M dio Purus Extractive Reserve ................................ .... 18 5 Network Interactions ................................ ................................ ....................... 194 Social Movement Networks ................................ ................................ ............ 194 Traditional Patron C lient Networks vs. Community Leader Networks ............ 195 Chapter Summary ................................ ................................ ................................ 196 6 NEW SOCIAL NETWORKS AND THE DISTRIBUTION OF SOCIAL CAPITA L ... 200 Social Capital and Patronage ................................ ................................ ............... 201 Tie Strength and Extensity ................................ ................................ ............. 202 Embeddedness ................................ ................................ .............................. 204 Position Generators ................................ ................................ ........................ 206 Position Generator Measurement ................................ ................................ ... 207 Using the Position Generator ................................ ................................ ......... 208 Producing the Position Generator Alter List ................................ .................... 209 Sampling ................................ ................................ ................................ ............... 211 Sample of Community Inhabitant Social Network Rating Interviews .............. 215 Conducting the Position Generator Interviews ................................ ............... 216
8 Respondent Characteristics ................................ ................................ ........... 216 Position Generator Questions ................................ ................................ ........ 217 Social Capital in Riverine Communities ................................ ................................ 222 Embedded Resources ................................ ................................ .................... 222 Embedded Resource Results ................................ ................................ ......... 224 Assistance and Tie Strength ................................ ................................ ........... 226 Tie Strength Summary ................................ ................................ .................... 231 Network Variability, Network Extensity, and Social Capital ................................ ... 232 Research Question #1 ................................ ................................ .......................... 232 Contact and Community Contact Scores ................................ ................. 232 W eighted Contact Scores ................................ ................................ ........ 233 Dependent Variables ................................ ................................ ...................... 233 Extractive Reserve Knowledge Score ................................ ............................ 233 Wealth Indicator Scores ................................ ................................ ................. 234 Statistical Analysis ................................ ................................ .......................... 236 Ordered Logit Models ................................ ................................ .............. 236 Chi Squares ................................ ................................ ............................. 236 Hypotheses ................................ ................................ ................................ .... 237 Statistical Results Research Question #1 ................................ ............................. 238 Extractive Reserve Knowledge ................................ ................................ ....... 238 Chi Square Results ................................ ................................ ........................ 240 Resex Information an d Network Variability ................................ .............. 240 Wealth and Network Extensity ................................ ................................ 241 Research Question #2 ................................ ................................ .......................... 242 Dependent Variables ................................ ................................ ...................... 242 Independent Variables ................................ ................................ ................... 243 Correlations ................................ ................................ ................................ .... 243 Linear Regression ................................ ................................ .......................... 244 Recognition ................................ ................................ ................................ ........... 244 Hypotheses for Recognition ................................ ................................ ........... 244 Recognition Results and Discussion ................................ .............................. 245 Recognition Summary ................................ ................................ .................... 248 Membership ................................ ................................ ................................ .......... 249 Hypotheses for Membership ................................ ................................ ........... 249 Membership Results and Discussion ................................ ............................. 250 Membership Summary ................................ ................................ ............. 253 Contact ................................ ................................ ................................ ................. 255 Hypotheses for Contact ................................ ................................ .................. 255 Contact Results and Discussion ................................ ................................ ..... 255 Contact Summary ................................ ................................ .................... 260 Community Contact ................................ ................................ .............................. 262 Hypothe ses for Community Contact ................................ ............................... 262 Community Contact Results and Discussion ................................ .................. 262 Community Contact Summary ................................ ................................ 267 Who are the Patrons? ................................ ................................ ........................... 269 The Patro in the Eyes of Community Inhabitants ................................ ................ 271
9 Understand ing the Term Patro in Lbrea ................................ ..................... 271 Power ................................ ................................ ................................ ....... 272 Contact ................................ ................................ ................................ .... 273 Communit y Contact ................................ ................................ ................. 274 Resource Generator Interview Results ................................ ........................... 275 Chapter Summary ................................ ................................ ................................ 278 7 DISCUSSION: POST SERINGAL PATRONAGE ................................ ................. 280 APPENDIX: ORGANIZATION INFORMATION ................................ ........................... 288 LIST OF REFERENCES ................................ ................................ ............................. 302 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 311
10 LIST OF TABLES Table page 3 1 Control of Key Resources in the Regime do T oco ................................ .............. 60 3 2 Control of Key Resources in the Rent System ................................ .................... 67 3 3 Rural Urban Migration in Brazil, Amazonas, and L brea 1970 to 2000 .............. 71 3 4 Control of Key Resources in Seringal Abandonment ................................ .......... 74 3 5 Control of Key Resources under State Control ................................ ................... 80 3 6 Control/Provision of Key Resources in the Regime do Toco .............................. 82 6 1 Likelihood of Providing Assistance ................................ ................................ ... 228 6 2 Network Variability and Contact ................................ ................................ ........ 238 6 3 Network Variability and Community Contact ................................ .................... 239 6 4 Net work Variability and Contact Weighted by Assistance ................................ 239 6 5 Network and Variability and Community Contact Weighted by Assistance ...... 240 6 6 Recognition Scores ................................ ................................ .......................... 248 6 7 Membership Scores ................................ ................................ .......................... 253 6 8 Contact Scores ................................ ................................ ................................ 260 6 9 Community Contact Scores ................................ ................................ .............. 267 6 10 Who are the patrons? (Power) ................................ ................................ .......... 272 6 11 Who are the patrons? (Contac t) ................................ ................................ ....... 273 6 12 Who are the patrons? (Community Contact) ................................ .................... 274 A 1 Organization Information ................................ ................................ ................. 289 A 2 Organization Assistance ................................ ................................ .................. 291 A 3 Resources Provided by Groups and Organizations in Lbrea .......................... 293 A 4 Who are the Patrons? ................................ ................................ ....................... 295 A 5 Network Variability and Contact ................................ ................................ ........ 299
11 A 6 Network Variability and Community Contact ................................ .................... 299 A 7 Network Variability and Contact Weighted by Assistance ................................ 300 A 8 Network Variability and Community Contact Weighted by Assistance ............. 300 A 9 Village Distances (Research Sites) ................................ ................................ .. 301
12 LIST OF FIGURES Figure page 4 1 Prod uction Calendar, Male Respondent, Santa Rosa, Purus River .................... 92 4 2 Production Calendar, Female Respondent, Santa Rosa, Purus River ............... 93 4 3 Production Calendar, Male Respondent, Vila Limeira, Purus River .................... 95 5 1 Seringal Patronage Triangle ................................ ................................ ............. 126 5 2 Post Seringa l Networks ................................ ................................ .................... 165 5 3 Seringal Patronage Triangle ................................ ................................ ............. 197 5 4 Post Seringal Multi Hub Network ................................ ................................ ...... 198 7 1 Seringal Patronage Triangle ................................ ................................ ............. 280 7 2 Post Seringal Multi Hub Network ................................ ................................ ...... 281 A 1 Map of Lbrea ................................ ................................ ................................ .. 296 A 2 Map of Deforestation ................................ ................................ ........................ 297 A 3 Map of Conservation Units Lbrea ................................ ................................ ... 298
13 LIST OF ABBREVIAT IONS ACAR Association of Credit and Rural Assistance (Associao de Crdito e Assistncia Rural) ADS Sustainable Development Agency of Amazonas ( Agencia de Desenvolvimento Sustentvel do Amazonas) AMIMP Association of Indigenous Women of the Middle Purus ( Associao de Mulheres Indgenas do Mdio Purus) APACJG Association of the Agroextractive Producers Jos Gonalves Community (Associao dos Produtores Agroextrativistas Comunidade Jos Gonalves) APADRIT Association of the Agroextractive Producers of the Assembly of God of the Ituxi River ( Associao dos Produtores Agroextrativistas da Assemblia de Deus do Rio Ituxi ) APAVIL Association of the Agroextractive Producers of Vila Limeira ( Associao dos Produtores Agroextrativistas de Vila Limeira) APREMP Ass ociation of the Producers of the Extractive Reserve of Lbrea (Associao dos Produtores da Reserva Extrativista de Lbrea) ASPACS Association of the Agroextractive Producers of Sardinha Colony ( Associao dos Produtores Agroextrativistas da Colnia do Sar dinha) ATAMP Association of the Agroextractive Producers of the Middle Purus ( Associao dos Produtores Agroextrativistas do Mdio Purus) CEB Christian Base Communit i es ( Comunidades Ecleisias de Base) CDH Human Rights Commission (Commiso de Direitos Human os) CIMI Indigenist Missionary Council ( Conselho Indigenista Missionrio) CNBB National Conference of the Bishops of Brazil (Conferncia Nacional dos Bispos de Brasil) CNS National Council of Rubber Tappers (Conselho Nacional dos Seringueiros) CNPT Nationa l Center of Sustained Development for Tradicional Populations ( Centro Nacional de Desenvolvimento Sustentado das Populaes Tradicionais)
14 CONTAG National Confederation of Agricultural Workers ( Confederao Nacional dos Trabalhadores na Agricultura) CPF Reg ister of Personal Identity (Cadastro de Pessoa Fsica) CPT Pastoral Land Com m ission (Comisso Pastoral da Terra) EMATER Technical Assistance and Rural Extension Enterprise of the State of Amazonas (Empresa de Asistencia Tcnica e Extenso Rural do Estado d o Amazonas) FETAGRE Federation of Agricultural Workers ( Federao dos Trabalhadores da Agricultura) FUNAI National Indian Foundation ( Fundao Nacional do ndio) FUNASA National Health Foundation ( Fundao Nacional da Sade) GTA Amazon Working Group (Grupo de Trabalho Amaznico) IBAMA Brazilian Institute of Environment and Renewable Natural Resources ( Instituto Brasileiro do Meio Ambiente e dos Recursos Naturais Renovveis) IBGE Brazilian Institute of Geography and Statistics (Instituto Brasileiro de Geogra fia e Estadstica) ICMBio Chico Mendes Institute for Biodiversity Conservation ( Instituto Chico Mendes de Conservao da Biodiversidade) IDAM Institute of Agricultural and Forestry Development of Amazonas ( Instituto de Desenvolvimento Agropecurio e Flore stal do Amazonas) IEADLAM Assembly of God Evangelical Church of Lbrea Amazonas ( Igreja Evanglica Assemblia de Deus Lbrea Amazonas) IEB International Institute of Education of Brazil ( Instituto Internacional de Educao do Brasil) INCRA National Instit ute of Colonization and Agrarian Reform ( Instituto Nacional de Colonizao e Reforma Agrria) INSS National Institute of Social Security (Instituto Nacional de Seguro Social) ISEI International Socioeconomic Index ITEAM Land Institute of Amazonas ( Institut o de Terras do Amazonas)
15 MANASA Madereira Nacional S.A. MMA Ministry of the Environ ment ( Ministrio do Meio Ambiente) MMC Peasant Women's Movement ( Movimento de Mulheres Camponesa) MMTRL Movement of Rural Women Workers of Lbrea (Movimento de Mulheres Trab alhadores Rurais de Lbrea) OPAN Oper a tion Native Amazonia (Operao Amaznia Nativa) OPIMP Organization of the Indigenous Peoples of the Middle Purus ( Organizao dos Povos Indgenas do Mdio Purus) PROBOR Program of Incentives for Natural Rubber Producti on ( Programa de Incentivo Produo de Borracha Vegetal) PRONAF National Program of Strengthening Family Agriculture ( Programa Nacional de Fortalecimento da Agricultura Familiar) RESEX Extractive Reserve SEARP State Secretary for the Articulation of Publi c Policies and Social and Popular Movements ( Secretaria de Estado de Articulao de Polticas Pblicas aos Movimentos Sociais e Populares) SEPROR State Secretary of Production (Secretaria Estadual de Produo) SESP Special Service of Public Health (Servio Especial de Sade Pblica) SPI Indian Protection Service (Servio de Proteo aos ndios) SUDHEVEA Superintendence of Rubber ( Superintendncia da Borracha) UEA University of the State of Amazonas (Universidade do Estado do Amazonas)
16 Abstract of Disse rtation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy SOCIO ECONOMIC CHANGE IN THE TRANSITION FROM PATRON CLIENT TO SOCIAL MOVEMENT NETWORKS IN BRAZILIAN AMAZONIA By Mason Clay Mathews May 201 1 Chair: Marianne Schmink Co chair: Chris McCarty Major: Interdisciplinary Ecology In Brazil approximately 12,063,068 hectares of forests and other landscapes have been designated as federal extractive reserves designed for self sustaining exploitation and the conservation of renewable natural resources. Researchers have used a variety of social science methods to study the dramatic social change that typically accompanies the creation of extractive re serves as communities bond into local social movements, build alliances with national and international organizations, and overturn longstanding patron client relations through land reform processes. This research project employed the theoretical concepts of social capital combined with social network analysis and ethnographic interview techniques to conduct a study of how isolated residents of two extractive reserves in the southern part of Amazonas State use their personal social networks to access resou rces. Network analysis combined with ethnographic data collection techniques generated quantitative and qualitative data regarding the level of interaction and cooperation between government institutions, local organizations (such as trade unions), and res erve residents. In addition to providing information on what types of
17 social networks have replaced the patron client relations that have long been associated with rural Amazonia, social network analysis provided a new approach for measuring the access th at extractive reserve residents have to markets, loans, and other types of capital. Natural rubber production was a key component of the economy of the municipality of Lbrea, in the southern part of Brazil's Amazonas state, from the 1870s until the earl y 1980s. The production unit established to produce this commodity, the seringal was characterized by rigid patron client ties between rubber tappers and landowners. This social system has changed with the collapse of extractivism and the rise of social movements pursuing social justice and environmental conservation in the Amazon. In this dissertation I explain how the networks of inhabitants of the extractive reserves in the municipality of Lbrea transitioned from social systems dominated by a single patron to systems in which individuals used relationships with social movements, government agencies, and other actors to access resources. A better understanding of the social networks of reserve residents may improve the success rate of conservation and development projects in the reserves as well as provide useful information for policy makers as they make decisions regarding the decentralization of resource management in the region.
18 CHAPTER 1 INTRODUCTION "A good patron never refuses to help us whe n we need something. When we ask he helps," Celina explained as she smoked a cigarette she had hand rolled from a scrap of old notebook paper. "A bad patron wants to be repaid immediately and does not give much time to repay debts," she added as smoke sl owly curled upward into the thatched roof from the lumpy cigarette hanging precariously from her lower lip. Celina, like many of the inhabitants of the margins of the Purus River, had grown up in a social system in which the exchange of forest products, p articularly rubber, for essential supplies dominated much of daily existence. A series of expectations regarding these exchanges had developed over the years and helped people like Celina to judge the validity of their exchange based relationships. In th is dissertation I present some findings, gathered during interviews with people like Celina, about how these exchanges, the relationships associated with them, and their perceived validity have changed over the past fifty years in the municipality of Lbre a in the southern part of the Brazilian state of Amazonas. Patron client ties were the foundation of an economy based on the extraction of forest products in much of the Brazilian Amazon for nearly 200 years. Over the last fifty years development projects inspired by modernization and other socioeconomic theories, together with global market changes have led to a series of new land uses that have replaced extractivism in some areas of the Amazon. In some areas the changes from extractivism have occurred m ore slowly while the increasing presence of government agencies and social movements has provided alternative social relationships to the long standing patron client ties that typified the production of rubber
19 and other extractive products. The purpose of this dissertation research was to determine what kinds of social structures, if any, have replaced the patronage long associated with extractive economies as a result of the socioeconomic changes in the municipality of Lbrea in southern Amazonas state. Socioeconomic changes over time can lead to the replacement of "traditional" dyadic patron client relationships with complex social networks in which resources are accessed through several patrons (Boissevain, 1977; C. Escobar, 1994; J. C. Scott & Kerkvliet, 1973; Theobald, 1983, 1992) Some people have more success in developing networks under the "new" socioeconomic conditions than others, and therefore have more needs met. What socioeconomic factors explain the changes from "traditional" dyadic patron client ties to "new" complex social networks and what characteristics make some individuals more successful at developing these networks than others? Does having a more extensive network make you better off or is it the quality of the specific ties? This dissertation is laid out in seven chapters, beginning with this introductory chapter. In Chapter 2 I discuss how the literature on patron client ties explains: a) what patron client relationships are and how the y function; b) under what conditions they are thought to exist; c) how they are viewed in different ways (positive or negative) and; d) the explanations for why and how they change over time. Patron client relationships have long characterized extractive economies in the Brazilian Amazon and in Chapter 3 I will explain the socioeconomic conditions that led to the rise of production systems based on patron client ties in the municipality of Lbrea, where I conducted field research for this dissertation. T he final collapse of
20 natural rubber production combined with the hyperinflation and economic difficulties that impacted Brazil in the late 1980s and early 1990s had profound impacts on patron client relationships in the area, as did the arrival of a host o f government agencies, social movements, religious groups, and other actors that offered alternatives to the old patronage system. As I analyze the socioeconomic changes that have occurred over the past fifty years in Lbrea I will show how "traditional" patron client arrangements have in some cases been replaced by "new" patron client arrangements due to changes in the control of resources. I will show how the local understanding of the word patron has changed and how it differs among different segments of the population. In spite of the changes in these relationships, many of the conditions that inspire people to engage in patron client relationships still exist in Lbrea. I will show that in spite of the socioeconomic changes of the last fifty years many inhabitants still lack vital resources and services that made it necessary to form relationships with "new" patrons and other actors, who controlled access to these resources. In addition to people's lack of access to many resources and services the socioeconomic changes over the past fifty years have caused profound changes in the use and control of natural resources in this area, which in many cases has led to conflict between user groups. The formation of social networks with employees of governm ent agencies and social movements is one solution that many communities have employed to resolve natural resource conflicts. In Chapter 5 I describe how three communities formed a series of relationships with newly arrived government agencies and social m ovements to address natural resource conflicts, consolidate their claims to land and natural resources, and to access credit and other financial resources. In order to
21 analyze the formation of these relationships I use as a point of departure the process o f the creation of the Mdio Purus and Rio Ituxi Extractive Reserves. This chapter reveals how the "new" relationships were formed and how they differ from the "traditional" patron client arrangements. In Chapter 6 I explore how the "new" networks differ among individuals in a sample of riverine communities in Lbrea. Using social capital theory and the position generator social network analysis method, I measured the networks of individuals from this sample of riverine inhabitants in order to show which groups (leaders vs. non leaders, men vs. women, etc.) are more likely to have better developed "new" networks and hence be better off. In order to determine if some subgroups of riverine populations in Lbrea are better at forming relationships after the collapse of the "traditional" patron client relationships I developed a modified version of the position generator social network analysis technique to measure: 1) which of the new groups and organizations riverine peoples recognized; 2) with which of thes e groups they have contact; 3) how they rate the power of these groups and organizations; and 4) which of these groups provide them access to resources and assistance. This information was collected to determine the relationship between an individual's nu mber of contacts (network extensity); 1) the amount of information they have related to the creation of the two extractive reserves in 2008, and 2) the amount of wealth they have acquired. I hypothesized that there would be a positive correlation between the extensity of an individual's networks and the amount of information and wealth they have acquired. In Chapter 7 I summarize the results of this analysis and my findings about how patron client ties have changed in Lbrea and what these results mean on a wider scale.
22 CHAPTER 2 PATRON CLIENT RELATIONSHIPS Theory and Patron Client Relationships Patron client relations have been dissected and analyzed by researchers and academics coming from different theoretical backgrounds and with objectives that are of ten quite different. Anthropology, Sociology, and Political Science have all made contributions to the study of patron client ties. Each of these disciplines approaches the concept in their own way, which has led to some confusion regarding what the term s mean and for whom. The development of patron client theory came at a time when many academics were rejecting structural functional theoretical bases in sociology and anthropology and looking instead to personal and interpersonal relations, power relati ons, and networks to explain behavior (Eisenstadt & Roniger, 1980) E arly research on patron client ties focused on the relatively limited, dyadic, interpersonal, semi institutionalized relations between a single patron and one or several clients (Eisenstadt & Roniger, 1980) The anthropologists who conducted this work tended to focus on personal relations such as the copa renthood relations of compadrazgo in small, rural communities (Foster, 1961; 1963; Mintz & Wolf, 1950) Much of the patron client research was conducted within a wider body of literature known as peasant studie s. Anthropologists, with a tradition of conducting detailed ethnographic studies with small sample populations, often focused on the micro level characteristics of these relationships. According to Weingrod (1968) anthropologists used the term to describe a particular kind of interpersonal relationship in which there are marked inequalities between the two parties (i.e. landowners and peasants), a reciprocal bond, and personal, face t o face
23 interactions. For Weingrod (1968) "the study of patronage as phrased by anthropologists is the analysis of how persons of unequal authority, yet linked through ties of interest and friendship, manipulate their relationships in order to attain their ends." Political scientists also adopted and expanded on the theory of patron client ties to help explain the relationships that determine how favors, employment, and other resourc es are exchanged for votes and other items within the realm of politics. According to Weingrod (1968) political scientists understood patronage as, "the study of how political party l eaders seek to turn public institutions and public resources to their own ends, and how favors of various kinds are exchanged for votes." His principal distinctions between the two are that anthropologists understood patron client ties as dyadic, social r elationships while political scientists viewed them as part of an election cycle and as part of formal organizations (Weingrod, 1968) Political scientists often used Frank So uraf's (1960) definition in which patronage is "an incentive system a political currency with which to purchase' political activity and political responses." A series of studies and publications in the 1960s a nd 1970s further illustrated how patronage is used in politics and many further limited the analysis to an exchange of votes for jobs, resources, and services distributed and exchanged within political parties and governments (Bearfie ld, 2009) This narrow conceptualization led some researchers to believe that important relationships and exchanges outside the realm of political parties and governments were being left out and that political scientists should return to a more anthropol ogical approach that focuses on the relationships between patron and client and the reciprocal nature of the exchanges (Bearfield, 2009)
24 Sociologists have approached the study of patron client ties in a similar way to anthropologi sts, but with perhaps a more systemic focus. In addition to analyzing the relationship between patron and client, sociologists may also focus on the social norms associated with these relationships and how they fit into larger social systems. For sociolog ists patron client ties may simply be one component of larger "social structures" (Martin, 2009) For others, like Brazilian sociologist Roberto Da Matta, patron client ties, as well as other social phenomena like the Brazilian jeitinho are strategies for overcoming systemic social imbalances, s uch as the disparities in wealth between different segments of the Brazilian population (DaMatta, 1990) The distinct approaches of the different disciplines and the foci of their analyses have led to different understandings of w hat patron client ties are and how the terms should be used. In spite of the disagreement over the meaning of certain terms and how they should be used, anthropologists, sociologists, and political scientists have constantly borrowed from each other as th ey advance the understanding of how these relationships work and what they mean. In this dissertation I will also draw from the three disciplines, particularly those works that analyze patron client relationships in rural, agrarian settings, like those I encountered among rural inhabitants in the Brazilian Amazon. Research Question and Patron Client Theory In this dissertation I will apply select components of the large body of literature that focuses on patron client relationships, patronage, and cliente lism to address the following research question: Socioeconomic changes over time can lead to the replacement of multiplex, dyadic patron client relationships for those in which resources are accessed through
25 several patrons (Martin, 2009) Some people have more success in developing networks un der the "new" socioeconomic conditions than others, and therefore have more needs met. What socioeconomic factors explain the changes from "traditional" dyadic patron client ties to "new" complex social networks and what characteristics make some individu als more successful at developing these networks than others? How well do these "new" networks provide people with the resources they need and want compared to the networks they have replaced? What do the changes in Lbrea say about similar changes in ot her places in which people face the socioeconomic changes associated with globalization, democratization, and the changes in social structures associated with them? In this chapter I will discuss the aspects of patron client relationships most relevant to my research question, namely: What are patron client relationships? What are the socioeconomic conditions associated with patron client relationships? How do patron client ties function? How do patron client relationships change over time? In an alyzing the changes in patron client relationships in Lbrea I believe that changes in the control of resources and services, changes in the perceptions regarding "traditional" patron client arrangements, and the arrival of alternative sources of resources and services have all been important components leading to the transition from the "traditional" patron client relationships associated with extractive industries in the Amazon, particularly rubber production, to a "new" kind of patronage involving a wide variety of actors. Consequently, in this chapter I will focus on the theoretical
26 elements associated with understanding how patron client relationships change over time. What are Patron Client Relationships? According to the Blackwell Dictionary of Soc iology patron client relationships are: those in which a client depends upon a special relationship with a patron who is in a position of influence or POWER. Typically, the patron provides protection or the means to earn a living (such as loans or acces s to land for cultivation) in exchange for loyalty, support, obedience, and services from the client. It is a relationship that can exist between individuals, GROUPS, or SOCIETIES. Although patron client relations are often based on tradition, RELIGION, o r contracts of one kind or another, the INEQUALITY inherent in the relationship makes it ripe for EXPLOITATION and abuse. This was the case, for example, in the relationship between nobles and PEASANTS under FEUDALISM in Europe, and in the SHARECROPPING s ystem found throughout the American South, especially after the Civil War ended slavery ("Patron client relations", 2000 Original Capitalization) Condit ions for Patron Client Relationships Understanding the socioeconomic context in which patron client relationships emerge is a key element for understanding how they change over time. These kinds of relationships date back to antiquity and the conditions t hat foment them have Resource Scarcity and Inequality The scarcity of and control over important resources is thought to be an important condition for the existence of patron client relationships. The degree and extent of local power, personified by the patrons and local mediators, depends on the one hand on the nature and degree of appropriation or monopolization of goods or services, and on the other hand on the rarity or precariousness of the resources (such as employment or land) and their importance for the survival of the clients. The fewer alternatives to an asymmetrical arrangement, the greater the probability of the dependent client submitting 'passively' to the dominating power of the master or patron (Silva, 1994)
27 In Lbrea, like much of the Brazilian Amazon, resources like land were certainly not scarce, as vast expanses of forest were in many instances available. However, the ability to control these lands and the natural resources found on them generally depended on an individual's access to credit, supplies, transportation, and a source of labor. Access to credit, supplies, and transpo rtation were scarce for many inhabitants of the area and the few individuals able to control access to them were in a position to offer them to less fortunate individuals in exchange for another scarce resource, labor. Inequality is thought to be a key el ement of patron client relationships, which are formed to ameliorate inequalities by allowing people with fewer resources to access them from those who have more resources (Martin, 2009) Inequality characterized the social setting in the Brazilian Amazon for most of its recorded history. Some h ad access to credit and other resources, while many did not. These conditions led to fertile setting for the formation of patron client relationships. Strong State / Weak State In early studies of patronage one of the key criteria for explaining the existence of patron client ties was the incomplete interaction between citizens and the state. In areas where the state was weak and had not gained complete control, it was thought that conditions for patronage and clientelism were better. In this way of thinking the focus was on the pre modern, pre capitalist, traditional nature of these socioeconomic conditions. It was thought that once these conditions gave way to those in which a strong state with "modern" bureaucratic systems took over, clientelism and patronage would tend to decline. One may posit that resort to patronage mechanisms will be the more pronounced where the weak are disproportionately weak, the strong disproportionately strong, and formal, alternative mechanisms for protecting
28 citizens laws, court systems, police, procedural rules of the game, etc. remain embryonic, manipulable or perhaps imbued with little or no legitimacy (Waterbury, 1977) However, newer research conducted in places that have witnessed the dramatic affects of democratization, return to civilian rule, increased mobilization, decentralization, neol iberalism, and other socio economic change has shown that patronage and clientelism, instead of disappearing, often simply change form and are adapted to the new conditions (Roniger !"#$%&' Ayata, 1994) Martin (2009) makes the important distinction that it may not be the presence of the state, but its ability to reach all of the citizens. As I will show in Chapter 4, an increased physical presence of government agencies does not necessarily mean that every one will be able to access resources and services through them. Due to its remoteness, the municipality of Lbrea was long characterized by the limited presence of government agencies and civil society organizations. In many ways it could be characteriz ed as a "pre modern" government system linked to a merchant capital economy based on extractive industries. However, over the past twenty years the Brazilian government has expanded its presence in the area and a host of civil society organizations have e stablished a presence there, making Lbrea an interesting place to analyze the way patron client relationships change over time in response to the increasing presence of the state and civil society. Economics and Isolation Economic conditions are a key e lement of any understanding of patron client relationships and how they change over time. The following socioeconomic characteristics are thought to increase the likelihood of the existence of patron client relationships: 1) Extensive, extractive economie s aimed at expanding control over large
29 areas to produce raw materials in exchange for finished goods from urban centers; 2) Oligarchic landholding strata with an export oriented economic focus, absentee landholders, and vast gaps between urban and rural s ectors; 3) Ruling and hegemonic classes economically, politically, and ideologically linked to imperialist powers; 4) Regional imbalances, with the accumulation of capital in urban centers (Roniger, 1990) Other soc ioeconomic structural characteristics that may encourage patron client relationships include: "(1) the persistence of market inequalities in wealth, status, and power which are accorded some legitimacy; (2) the relative absence (or collapse) of effective, impersonal guarantees such as public law for physical security, property, and position often accompanied by the growth of semi autonomous local centers of personal power; and (3) the inability of either kinship units or the traditional village to serve a s effective vehicles of personal security or advancement (J. C. Scott & Kerkvliet, 1973) ." As I will show in Chapter 3, all of the above characteristics were in evidence in Lbrea from the time of its foundation as a remote colonial outpost established to produce rubber and other forest products in exchange for finished products from urban centers. The following quote by Luis Roniger describes a set of conditions in which "traditional" patron client relationships are likely to form. The more 'traditional,' localized patron client networ ks existed where the peripheries (particularly agrarian sectors) were tenuously connected to capitalistic markets as simple commodity producers. In such settings, there was a 'seigneurial' appropriation of basic resources, of means of livelihood and, abov e all, of lands, which were leased to peasants in the form of fragmented and small strips in return for labor, cash, or rents in kind. Moreover, precapitalist forms of work organization, in the framework of rent capitalism, developed, and landowners and m erchants had sole access to peasant labor and much of their production. This in turn was usually associated with a low level of mechanization and of technological and
30 capital investment; a scarcity of credit facilities; and poor communications. The low c apacity of landless workers and minifundist peasants to draw a livelihood in a relatively independent way, the precariousness of agricultural contracts, and the effects of multiple part time occupations, especially among peasants, were also conducive to th e emergence of the 'captive' clienteles that are characteristic of such localized clientelistic networks (Roniger, 1990) Interestingly enough, this quote in many ways describes the conditions that existed for more than one hundred years in the municipality of L brea, as we will see in chapter three. Patron Client Profiles Having analyzed the types of socioeconomic conditions in which patron client relationships tend to form, it is important to begin to unpack the relationships themselves. What are the differences between patrons and clients? What kind of interaction occurs between patrons and clients? The following profile of patron client relationships provides a snapshot of some of the characteristics of thos e who engage in patron client relationships. For the purposes of this paper, the profiles developed over the years by Luis Roniger and Samuel Eisenstadt are useful for illustrating some of the key components of patron client relationships (Eisenstadt & Lemarchand, 1981; Eisenstadt & Roniger, 1980; Roniger, 1990) For these authors patron client relationships are vertical (hierarchical) bonds based on inequality and power differences that are usually found in stratif ied societies in which benefits and resources are distributed selectively to certain elements of the population. Generally the patron is from a more well off social position and controls or monopolizes the means of production, market access, distribution of public goods and/or community services that the client needs The relationships are "particularistic and diffuse" (goods and services are exchanged according to the
31 personal traits of the actors) and are highly selective. Patron client ties involve si multaneous exchange of two different types of resources and services (" instrumental" economic or political resources and expressive resources consisting of things like promises of loyalty and solidarity) in a package deal in which both sides offer the ot her instrumental and expressive resources. Both patron and client expect their attachment to be a relatively unconditional, long term relationship. Credit and obligations are also long term and the relationships are formed between individuals of high and low status, not between organized corporate groups. The ties are often neither fully contractual nor legal and may be illegal but are based on informal, though tightly binding understandings that are periodically renegotiated. Most of these ties are voluntary, despite their implied long range commitment, and can, at least in theory, be broken off by either side (Eisenstadt & Lemarchand, 1981; Eisenstadt & Roniger, 1980; Roniger, 1990) Patron Client Processes Profiles like the one above are useful to define certain key characteristics of patron client relationships and those individuals who engage in them. However, researchers and academics have long understood that the processual elements of how these relatio nships are formed and maintained better reveal their true nature than the typologies, which may be "little more than lists of parts (Lemarchand, 1981) ." I believe that the most salient processual components for explaining these relationships in remote areas of the Brazilian Amazon, like Lbrea, are: 1) the cont rol and monopoly of access to certain resources and services; 2) the arrangements of "generalized exchange" of "instrumental" and "expressive" resources between patrons and clients; 3) the thresholds and expectations associated with these exchange
32 arrangem ents; and 4) the possibility (or lack thereof) of alternatives to patron client relationships. Resource Control According to Waterbury (1977) one way to identify a potential patron is to determine who has monopolistic control over a necessary resource. One of the characteristics of the "classic" or "traditional" patrons who dominated the extractive industries in the Amazon in the colonial period and during the rubber boom was their ability to monopolize control over a series of important resources. As I will explain in greater detail in Chapters 3 and 4, control of the following six resources (market access, land/resources, health services, education, religion, political representation, and information) was fundamental for a patron to maintain his leverage and enabled him to provide key resources and services in exchange for forest product commodities in Lbrea, both then and now. Generalized Exchange With vital resources at his (or her) disposal a patron or potential patron is able to engage in relationships of "generalized exchange" in which a series of extra social obl igations serve as insurance against risks and uncertainties found in routine' market exchanges (Eisenstadt & Roniger, 1984) An understanding of the resources being exchanged between patrons and clients has long been thought to be one of the keys to illumin ating the inner workings of these relationships (Eisen stadt & Roniger, 1980) In short the question here is what are the resources that patrons control and can offer to potential clients, and what do the clients have to offer in return for these resources? In a study particularly relevant to the analysis p resented in this dissertation, Scott and Kerkvliet (1973) provided an example of this kind of analysis based on their study of
33 how traditional patrons in an agrarian setting in Asia had lost legitimacy. They observed that in the patron client ties they were studying patrons were expected to provide the following to their clients: 1) basic means of subsistence (access to land for cultivation, the provision of seeds, equipment, marketing services, technical advice, and so forth); 2) subsistence crisis insurance (loans in time of economic dis tress, help in case of sickness or accident, or insurance in case of a poor harvest); 3) protection (shield the client from private and public dangers); 4) brokerage and influence (extract rewards from the outside for the benefit of his clients); and 5) co llective patron service (services for the community as a whole sponsor village festivals and celebrations, host visiting officials, mediate disputes, etc.) (J. C. Scott & Kerkvliet, 1973) Scott and Kerkvliet (1973) pointed out that measuring the flows from client to patron may be more difficult, but no less important for understandin g how the relationship functions. In the context of Southeast Asia where they were conducting research, clients were expected to provide patrons: 1) basic labor service (client contributed his labor and other specialized skills to the farm, office, or ent erprise); 2) supplementary labor and goods (some substantial, some symbolic); 3) promotion of the patron's interests (J. C. Scott & Kerkvliet, 1973) The ability to demand that his or her clients engage in exclusive exchange with him or her is thought to be a key element of patron client relationships (Martin, 2009) When patrons are able to e nforce this "one patron per client" rule, social structures, called patronage pyramids, form (Martin, 2009) Patrons generally use their ability to control scarce resources to invoke, or threaten to invoke, sanctions against their clients in the event that they break this "one patron" rule and ex change with other patrons
34 (Martin, 2009) A patron threatening to curtail credit or expel a client from his property is an example of how control of scarce resources provides patrons with the ability to invoke, or threaten to invoke, sanctions (Martin, 2009) In Lbrea, and much of the Brazilian Amazon, "classic" or "traditional" pat rons, as I will call them here, used their contacts with providers of credit and other key resources to provide necessary resources (food, medicine, household items) to poorer, less well connected laborers in exchange for a variety of natural resource comm odities. The ability of patrons in Lbrea to control resources, provide services, and invoke sanctions to maintain exclusive exchange arrangements changed over time, based on a series of socioeconomic changes. Four types of "generalized exchange" arrange ments between patrons and clients existed at different times in Lbrea : 1) regime do toco ; 2) arrendamento or rent arrangements; 3) abandonment; and 4) state control. In Chapter 3 I will discuss how these exchanges functioned and what types of resources p atrons and clients offered one another in these exchanges. Expectations and Thresholds Between patrons and clients a series of expectations are formed regarding their arrangement of "generalized exchange", which is usually in a constant state of negotiati on. Clients often expect that their patron will provide certain resources (credit, food, medicine, etc.) and will not leave them wanting in times of need. In exchange patrons may expect deference, political support, or some form of exclusive commercial i nteraction. Thresholds generally exist on both sides of these relationships that when crossed can lead one or both sides to question the legitimacy of the arrangement and to seek alternatives (J. Scott, 1977)
35 In Lbrea, as in much of the Brazilian Amazon, gatherers of forest products expected their patrons to provide them with supplies on credit and not let them fall below a certain poverty threshold. Local patr ons usually expected their clients to work exclusively for them and refrain from selling precious forest products to others, even if they offered better prices. A patron that could not provide the necessary supplies may find himself abandoned by his worke rs or run the risk that they would sell their products to someone else. A patron that was seen as being too unfair in the terms of exchange might also run the risk of being abandoned, having products sold clandestinely to other buyers, or even receive adu lterated or inferior forest products in exchange for the supplies he provided. The difficulty patrons had in monitoring vast areas of forest made the subtle forms of daily resistance described in James Scott's (1985) "Weapons of the Weak" a common response for many clients when they felt they were being treated unfairly. However, patrons also responded to actions that they felt were contrary to their "agreement" with their clients. A client that was caught selling forest products to others or adulterating the product might have his access to the patron's suppli es curtailed or might be expelled from the patron's land. In Chapters 3 and 5 I analyze how the ability of "traditional" patrons in Lbrea to provide the expected resources to their clients ebbed and flowed over time and manifested itself in the changes back and forth from regime do toco to arrendamento or rent agreements, until the point when "traditional" patrons provided virtually none of the benefits their clients expected. Once this threshold was crossed the legitimacy of these "traditional" relatio nships was called into question and clients began to search for alternatives.
36 Alternatives and Change In the event that thresholds are crossed and patron client relationships dissolved, the obvious question becomes what alternatives exist for both parti es. If the patron has complete monopolistic control over necessary resources and services, options may be limited for a client trying to replace him or her. Even in the event that monopolistic control is not complete, the alternative for many clients may simply be to find another patron who can provide the necessary resources in exchange for their deference, labor, etc. Once patrons lose the ability to maintain monopolistic control over key resources, particularly land, their ability to maintain exclusi ve relationships with their clients diminishes (Martin, 2009) In these instances, when patrons can no longer provide key resources or invoke sanctions based on their control of key resources (such as threatening to curtail credit or evict clients from their property) they no longer have the abil ity to maintain the "one patron per client" rule and the patronage pyramid begins to collapse (Martin, 2009) One possible outcome when a system based on highly multiplex patron client relationships begins to change is that clients will have to develop a more diffuse network of relationships wit h a greater number of patrons to obtain the resources previously provided by a single patron (Boissevain, 1977; C. Escobar, 1994; J. C. Scott & Kerkvliet, 1973; Theobald, 1983, 1992) When this occurs clients begin to seek out other sources of resources and services. In some cases, when they have to form relationships with several actors to obtain these resources, new social structures begin to take shape. Often the arrival of the state is associated with patrons' loss of control over land and the ability to provide key resources. In these cases, state agencies and
37 other actors may begin to provide resources in a new system in which former clients must form relationships with a series of brokers and intermediaries with connections to other actors who can provide access to resources and services (Martin, 2009) Until recently in Lbrea, as in many parts of the Brazilian Amazon, in the event that a patron client relationship was dissolved the only viable alternative for accessing the vital resources provided by the old patron was to find a new one. Rubber estate owners and merchants controlled virtually all economic activity over vast areas, and alternative avenues to access credit and necessary supplies were extremely limited. However, over the past fifty y ears tremendous socioeconomic changes have occurred in the area that may have made available a series of "new" patron client arrangements with government officials, social movement organizations, religious entities, and NGOs to replace the "traditional" pa trons that dominated the area for so long. In Chapter 5 I present case studies from three communities in Lbrea to show how people responded to collapse of the patronage pyramids that had formed during the rubber boom and how they were beginning to form s ocial networks in which they accessed resources and services through a series of brokers and intermediaries. In summary, over the past fifty years a series of circumstances have led to socioeconomic changes in Lbrea that have potentially changed; 1) who controls key resources, 2) the nature of the exchanges between potential patrons and clients, 3) the thresholds and expectations associated with these arrangements, and 4) the alternatives that clients have to access the vital resources that were once sup plied by a limited number of potential patrons. These dramatic changes made Lbrea an interesting place to investigate how patron client relationships change over time.
38 CHAPTER 3 "TRADITIONAL" PATRON AGE IN THE SOCIOECON OMIC EVOLUTION OF L BREA Patron c lient relationships have been a key component of economies based on forest products in many parts of the Brazilian Amazon since the period of Portuguese colonization. In order to extract dispersed forest products and ship them to national and internationa l markets, a credit system was developed in which essential supplies (and luxury items) were purchased on credit in exchange for forest products. This chain of credit was also a chain of patron client relationships that began with the collectors working i n the forests and linked them through a series of intermediaries to the export houses that sold the commodities on international markets and imported the trade goods for which they were exchanged. During the rubber boom of the late 19 th and early 20 th cen turies this economic system, known as aviamento became institutionalized throughout the Brazilian Amazon and continues to exist in many areas (Weinstein, 1983) The municipality of Lbrea in the southern part of the of the state of Amazonas was founded as a direct result of the rubber boom, and forest product extraction continued to be the basis of the municipality's economy until the mid 1990s. In this chapter I describe how patron client relationships, and the aviamento credit system, have changed in riverine communities in the interior of Lbrea over the past fifty years. I describe: 1) the historical socioeconomic setting in which patron client relationships developed; 2) the resources exchanged between people living in riverine communities and their "patrons"; 3) the expectations upon which these exchanges were (and are) based and the ability of both sides to fulfill these expectations; and 4) the vulnerabilities and alternative relationships (or lack th ereof) that helped determine
39 whether or not people in riverine communities engage in patron client relationships, and may ultimately decide their continued participation in these arrangements. Exchange, Expectations, and Patron Client Ties For the client the basic purpose of the patron client contract, and therefore the cornerstone of its legitimacy, is the provision of basic social guarantees of subsistence and security. If and when the terms of trade deteriorate sufficiently to threaten these social r ights, which were the original basis for attachment and deference, one can anticipate that the bond will quickly lose its legitimacy. The patron may still be able to extract services from the client but clients will increasingly consider the relationship unjust and exploitative (J. C. Sc ott & Kerkvliet, 1973) One of the processual elements considered essential for understanding patron client relationships is the notion of "generalized exchange" in which a series of extra social obligations, based on norms of reciprocity, serve as insura nce against risks and uncertainties found in routine' market exchanges (Eisenstadt & Roniger, 1984) The perceived legitimacy of a patron client relationship for those involved often rests on this notion of reciprocity and is manifested in a series of expec tations regarding the interactions and exchange between patron and client (J. Scott, 1977) In this chapter I provide information about different exchange arrange ments (and their perceived legitimacy) that patrons and clients in Lbrea engaged in during distinct periods of time. These include the: 1) regime do toco periods, 2) arrendamento or "rent" periods, 3) periods of rubber estate abandonment, and finally 4) the state control period. The regime do toco exchange arrangements are generally associated with the rubber boom periods (1870 1910) and (1943 1945) when rubber prices were high and credit was more available to seringal owners. The arrendamento exchange arrangements were more common in the "bust" periods, when rubber prices were lower and credit was less available (1920 1930s) and (1960s 1970s). However, these arrangements often
40 existed at the same time, depending on local conditions. The abandonment ph ase, characterized by the collapse of the seringal production system and the dissolution of the exchange between seringal owners and rubber tappers began in Lbrea in the 1970s and ended in the late 1990s, when only one or two old time seringal owners rema ined. What I am calling the state control period began in the 1970s with the creation of Indigenous Lands in the municipality, intensified in the 1990s with the creation of more federal conservation units, and became consolidated in the 2000s, by which ti me almost the entire municipality was covered by conservation units. I do not mean to imply that the state has absolute control of all aspects of resource use, but the transition from privately owned seringais to government controlled conservation units c ertainly implied potentially dramatic changes in the way resources were used. The interactions between patrons and clients are shaped to a large degree by the socio economic conditions in which they are embedded (Roniger, 1994a) In addition to describing the details of the exchange between riverine peoples and their patrons, I illustrate the broader historical changes that have occurred in Lbrea over the past fifty years that have created and shaped these relationships. Understanding changes in the socio economic context can help reveal: 1) the nature of the resources available for exchange in patron client relations, 2) who co ntrols or monopolizes access to these resources, 2) and the changing nature of the formal and informal social norms that govern exchanges between patrons and clients. Exploration and Extractivism on the Purus River Every afternoon we observed, as we steame d along, wisps of smoke rising from the banks at frequent intervals. These indicated the places where men were engaged in the process of manufacturing india rubber the Purus being quite the head quarters of this business (Brown & Lidstone, 1878)
41 The sediment rich Purus River slowly meanders its way 3,200 kilometers down from the foothills of the Peruvian Andes to its confluence with the main channel of the Amazon (Goulding, Barthem, & Ferreira, 2003) Thousands of oxbow lakes rich in aquatic resources have split off the main channel of the Purus as it periodically changes course during the yearly flood cycle. The rich forests and lakes that surround the river have been the scene of dramatic social and economic change over the past 150 years. The Purus Valley has a long history of human occupation and the richness of its aquatic and forest resources attracted Amerindian peoples long before the arrival of Europ eans. The river and its tributaries were inhabited by a variety of ethnic and linguistic groups, including the Mura, Apurin, Paumari, Zuruah, Jamamadi, and Deni, among others, who were rapidly enveloped in the wave of exploration and settlement that beg an in the 1840s. The exploration of the Purus River and the founding of the city of Lbrea coincided with the expansion of extractive industries in the southwestern Brazilian Amazon and the institutionalization of a series of patron client relationships u pon which economic activities in the region were based for more than one hundred years. Early Exploration Descimentos and Drogas do Serto The lower sections of the Purus River near its confluence with the main channel of the Amazon saw an initial wave o f Brazilian explorers and traders during the colonial period. A significant part of the Amazonian economy in the 17 th to 18 th centuries was based on a series of forest products (cacau, sarsaparilla, and cinnamon, among others) known generally as drogas do serto (Reis, 1998) Indigenous slaves did much of the collecting of forest products and rowed the canoes that transported them to markets (Ribeiro, 1996) In addition to being bonded into slavery, indigenous groups were
42 incorporated into an incipient economy based on merchant capitalism in which forest products were exchanged for bea ds, metal tools, alcohol, and other items. During the initial phase of expansion up the Purus, Brazilians collecting the drogas do serto brought in a supplies for an extended period, which were purchased on credit in Belm or Manaus, and then returned t o these urban centers to pay their debts after extended periods with the forest products they had collected (de Melo, 1994) During this period initial contact was made with many indigenous groups, systems of trade and barter were established to facilitate the extraction of forest products, and Catholic Church missionaries relocated some of the indigenous groups from the Lower Purus as part of the descimentos in which indigenous peoples were concentrated to form missions (Kroemer, 1985) While the Lower Purus was depopulated to form mission villages, the middle and upper sections of the river were l ittle explored by Europeans until the middle of the 19 th century (Ribeiro, 1996) The trade and barter system established during the initial contact between Brazilian explo rers and merchants and indigenous peoples developed into an extended debt based system of exchange known as aviamento that became the foundation of the region's economy. Aviamento : Credit, Debt, and Exchange Understanding the aviamento credit system, which began during the colonial period, is key to understanding the economy of the Amazon during the rubber boom period and beyond. The dispersed geographic distribution of the forest products that were so important to the colonial economy of the Amazon, the di stance to their intended markets in Europe, and the limited availability of banks and currency in many places may begin to explain the establishment of the aviamento system (R. Santos, 1980; Weinstein, 1983) Forest products were often located in distant areas that presupposed
43 the formation of expeditions to go and harvest them. Transportation and food supplies would be needed to get merchants and collectors out to these areas, while trade goods were often needed to exchange with the indigenous peoples who collected the forest products. Theref ore, a line of credit would be essential for someone unwilling or unable to bankroll his or her own expedition. A creditor willing to advance trade goods for extended periods of time was often necessary for someone hoping to collect forest products. The bandeirante explorers working in southern Brazil depended on similar credit arrangements to fund their slave raids and expeditions in search of precious metals and stones (Hemming, 1978) In the A mazon, aviamento was based on a chain of credit as well as a chain of patron client relationships that began with the collectors working in the forests and linked them through a series of intermediaries to the export houses that sold the commodities on international markets and imported the trade goods for which they were exchanged. The verb aviar in Portuguese literally means "to send" or "to dispatch". The term aviamento refers to a process in which supplies are advanced to producers of a commodity o n credit with the understanding that the producer will pay back the line of credit, often with interest, at the end of the harvest season, crop cycle, etc. In this process the person furnishing the credit and/or supplies is known as the aviador and the re ceiver is sometimes referred to as the aviado The person supplying the credit was (and is) also referred to as the patro (patron) and person receiving the credit was (and is) often referred to as the fregus (client). Both terms were still common in L brea when this research was conducted in 2008 2009. Excellent scholarly works have chronicled how this extensive system of long term credit, often involving a series of
44 intermediaries, was employed to bring the dispersed forest products of the Amazon to n ational and international markets (Barham & Coomes, 1996; Dean, 1987; Frank & Musacchio, 2006; R. Santos, 1980; Weinstein, 1983) Expansion up the Purus In the 1840s and 1850s early explorers like Joo Cameta, Seraf im Salgado, and Manoel Urbano da Encarnao explored extensive areas of the Purus Basin (Bittencourt, 1918) These early explorers were interested in forest commodities and began to bring indigenous groups under their control to establish the labor force required to extract these products from the forest and row them to distant p orts (Kroemer, 1985) From the time of these earliest expeditions there was a keen interest in finding tributaries that would link the larger tributaries of the Amazon (such as the Purus and Madeira Rivers) and fu rther facilitate shipping and trade linkages. In 1864 65 the British explorer William Chandless journeyed yet further into the upper reaches of the Purus; his maps and descriptions are some of the best from this period of exploration. At the time of Cha ndless' trip the Purus was inhabited primarily by indigenous groups, the traders of the drogas do serto were not establishing year round residences, although some of them had established feitorias which were supply sheds that were precursors to the tradi ng posts that would become ubiquitous during the rubber boom (Chandless, 1866) Iron axes, fishhooks, arrowheads, beads, and other trade articles that dazzled the indigenous people were used to establish trade relationships that often deteriorated into debt peonage a nd slavery, as occurred in other parts of the Amazon (Hemming, 1978; 1 987) The early traders and explorers of the Purus were interested in many of the same forest products salsaparilla, cacau, copaiba oil, turtles and turtle eggs that had been the mainstay of the region's economy
45 during much of the 18 th century. Howev er, the early exploration and settlement of the Purus paled in comparison to the rapid expansion that coincided with the economic boom that would shape the socioeconomic system of the Amazon for the next seventy years. The Rubber Boom and the Settling of the Purus The invention of the process of vulcanization by Charles Goodyear in 1839, industrialization in Europe and the United States, and a series of inventions (bicycles, automobiles, electricity) led to an increasing international demand for rubber in the second half of the 19 th century (Dean, 1987; R. Santos, 1980; Weinstein, 1983) Aside from some lower grade natural rubbers produced in Africa, the Amazon was the only source of high grade natural rubber until plantations were developed in Asia at the end of the 19 th century. The economic incentives generated by international demand for rubber combined with the desire to expand territorial claims and discover new trade routes led to the rapid settling of the m iddle and upper sections of the Purus River in the second half of the 19 th century. The establishment of steamship lines on the Amazon in 1853 greatly facilitated this phase of expansion by making it possible to bring in supplies and labor from far away c ities and return with rubber and other forest products (L oureiro, 2007; Reis, 1998) Steamships began regular service on the Purus in 1869 and were the deciding factor in the rapid settlement of the area and expansion of the rubber trade (da Cunha, 2003c) The news of th e natural riches on the Purus spread by early explorers and the traders of the drogas do serto generated a wave of settlers interested primarily in the ever expanding rubber business.
46 The expansion of rubber production often led to tragic results for th e indigenous peoples of the Purus, who were either massacred, died of diseases for which they had no immunity, fled to areas that were not coveted for rubber production, or were assimilated into the rubber based extractivist society that was forming (Ehre nreich, 1929; Steere, 1903) The ever increasing demand for rubber on international markets combined with a tendency to expand production areas (instead of improving production techniques) led to a demand for labor that could not be met by the dwindling A merindian population of the Purus. However, in the late 1870s devastating droughts in Northeastern Brazil created an available labor force that migrated to the area by the thousands to work in the rapidly expanding rubber economy and began to replace and intermix with the indigenous labor force (H. H. Smith, 1879) The middle and upper Purus were settled with astonishing rapidity after Chandless' trip; accounts from the 1870s speak of several thousand rubber tappers setting up operations on the banks of the river (Brown & Lidstone, 1878) Coronel Antonio Pereira Labre, who brought a group of workers from Maranho to establish rubber production areas in 1871, said tha t there were 5,000 civilizados living in the area with thousands more on the way (Labre, 1872) As in other parts of the Amazon, early settlers and explorers commented on the lack of agricultural production on the Purus all the available labor w as tapping rubber which led to a dependence on expensive foodstuffs imported from Manaus and Belm (Chandless, 1866; Labre, 1872) The settlement that Coronel Labre had established with migrants from Maranho in 1871 quickly developed into the town named Lbrea in his honor. By 1880 Labre est imated that there were 25,000 people in the vicinity of Lbrea and the area was
47 given the status of municipality by the government of Amazonas in 1881 in response to its growing population (Bittencourt, 1918) The Purus became a gateway to the rich rubber fields of Acre and served as a major transportation route to Manaus and Belm. By 1890 the P urus was estimated to have 50,000 inhabitants and by 1902 this number was estimated to have risen to 120,000 (da Gama e Abreu, 1895; Jaramillo, 1902) The total rubber exports of the Purus for 1892 doubled those of 1885 and t his number was doubled again between 1892 and 1902 (da Cunha, 2003c) Lbrea, a key stopping point for steamships bringing in supplies and taking out rubber and other forest products, quickly developed into one of th e river's major boom towns. By 1911 nine different newspapers had been in publication in Lbrea and the town became one of the few to have a wireless telegraph service (Ferrarini, 2009) The rapid growth of Lbrea and the settling of the Purus were due to the booming rubber economy, and according to Brazilian author Euclides da Cunha (2003c) Lbrea quickly developed into an important commercial cente r; its revenues and expenses were the highest of any city in Amazonas, including Manaus. This boom economy was based on a system of rubber production, linked to the outside world through the aviamento credit system, known as the seringal This production system and its entrenched patron client relationships shaped social interactions in much of Lbrea until the late 1980s. The Seringal Rubber Production System As the rubber economy began to develop, the old drogas do serto collection system in whic h boats and canoes with supplies were sent on expeditions for extended periods before returning to urban centers, was replaced by a more sedentary system based on rubber production estates (de Melo, 1994) Because the Hevea brasiliensis rubber trees could be tapped every other day or so during the dry season over a period
48 of years, it was possible to establish permanent rubber production ar eas in the forests of the Amazon. In the initial phase of opening a rubber estate mateiros workers with extensive knowledge of the forest, laid out a series of trails connecting the rubber trees from which the precious latex would be extracted (da Cunha, 2003b) In native forests rubber trees have a dispersed distribution, which meant that the seringal often covered an extensive area of forest. Land titles were sometimes established by posse or possession, or were granted by municipal and state government organizations (Woodroffe, 1914) Force was often necessary to maintain a claim in these expanding frontier regions that were effectively out of the reach of the government (Akers, 1914) Trade activity on the rubber estates was conducted at the owner's trading post, known as a barraco The barraco was typically a one or two story wooden building, sometimes with a tile roof, in which the owner of the seringal known as a seringalista stored supplies and conducted his business (Tastevin, 1943) The barraco was built near the margin of whichever waterway provided access to the area and facilitated the shipment of supplies and forest products. The seringalista often called patro or patron, sometimes lived in the upper floor or in a wooden building nearby. This central cluster of buildings often included storehouses, known as armazens and deposits and fuel du mps known as paioles A series of trails radiated out from this central point into the forest to the areas where rubber was produced. The rubber tappers who collected the latex lived in simple wooden structures, usually built of trunks from the paxiba p alm ( Socratea exorrhiza ), and covered with thatched roofs. Their isolated production areas and living quarters were known as a colocao
49 During the tapping season the rubber tapper would rise early in the morning and walk the rubber trails near his hom e, making incisions in the boles of the rubber trees located along a loop trail that would bring him back to a starting point. Trails typically had 100 to 200 rubber trees on them, but the dispersed distribution of the trees meant that a trail might be se veral kilometers in length and might take several hours to walk. Having cut the incisions in the trees the tapper would then walk the trail again to collect the latex from the small cups that he had placed under each incision. The liquid was taken in a b ucket back to a point near his home for the process of defumao or smoking. During this process the tapper would slowly pour the liquid onto a stick positioned over the rising smoke from a fire of palm nuts. The smoke would coagulate the latex into a so lid ball of rubber. These rubber balls, known as peles were then taken to the barraco every 15 days or so to exchange for supplies and pay outstanding debts to the seringal owner. Rubber tappers often worked up to fourteen hour days from the time they began walking the trails until they finished the process of defumao (Akers, 1914) On some of the larger rubber estates mules or canoes were used to transport supplies to the tappers and bri ng back rubber. By 1906, when the Brazilian author Euclides da Cunha led a Brazilian / Peruvian boundary commission expedition to the upper reaches of the Purus there were 400 seringais on the Purus River, not counting those on tributaries like the Ituxi, Pauini, Iaco, and Acre (da Cunha, 2003c) The Culture of the Barraco In order to understand the patron client ties that existed on the seringais and that have shaped the social interactions in ri verine communities along the Purus up until the present, it is critical to understand the interactions associated with the barraco trading posts. Power on the rubber estates was concentrated at the barraco trading post
50 where rubber, Brazil nuts, and oth er forest products were brought from the rubber tappers' homes in the interior of the property and exchanged for food, tools, medicine, clothing and other products that were essential for survival. On most of the seringais money was a rarity and forest pro ducts were exchanged directly for merchandise. This exchange was based on the familiar aviamento system in which a rubber tapper would receive supplies, known locally as mercadoria on credit and then pay back the debt with whatever rubber and other fores t products he produced over an extended period of time. The exchange of rubber and other forest products for needed supplies were at the heart of patron client relationships between seringal owners and rubber tappers during the roughly seventy year perio d that represented the rubber boom. Some believe that the patron client interactions associated with the aviamento credit system became so ingrained that they led to the formation of a cultura do barraco or trading post culture, in riverine communities along the Purus (F. S. D. d. Santos, 2000) In spite of the entrenched nature of this mode of exchange, it is important to recognize that these trading arrangements changed over time and were conditioned to a large degree by the often extreme fluctuations that characterized the extractive economy. In the next section I will describe the types of exchange that occurred between patrons and clients in rubber producing areas during several d istinct economic phases. These include the following exchange arrangements: 1) regime do toco ; 2) arrendamento or rent arrangements; 3) seringal abandonment; and 4) extractive reserves. In this kind of analysis it is tempting, but incorrect, to imply a perfect lineal trajectory to these relationship changes. There were certainly ebbs and flows in the
51 nature of the relationships between patrons and clients that often mirrored the ebbs and flows in the extractive economy and there were periods of time wh en different exchange arrangements existed at the same time. It was clear during timeline interviews with older inhabitants of the Purus in 2008 2009 that they associated certain patron client exchange arrangements with particular time periods. Many resp ondents mentioned the names of these exchange arrangements, such as regime do toco during interviews without being prompted. The information in this section was also taken from the accounts of travelers who passed through Lbrea during different time peri ods. Carlos Teixeira's (2009) excellent analysis of the relational elements of aviamento including his in depth explanations of the regime do toco and arrendamento systems, were also important for developing this analysis. The Regime do Toco The type of exchange between patron and client most often associated with boom periods, when rubber and other forest products had high prices, is known as the regime do toco (Teixeira, 2009) In this system there was a rigid, regimented nature to the exchanges between seringal owners and the rubber tappers who worked for them. Tappers were required to buy all of their supplies in the owner's barraco and sell their rubber only to him. The patron and his staff controlled the tappers' debt ledgers and in the most remote areas could also control information about the price of rubber and supplies. In the regime do toco a p atron could adjust the amount of merchandise that tappers would receive based on their credit standing and use this leverage to further control them. Complete control of a rubber tapper's access to necessary supplies was a key control mechanism for most p atrons because the dispersed nature of the seringal production system made it difficult to monitor the tappers' work and police the
52 clandestine sale of rubber (Barham & Coomes, 1996) With the patrons setting the prices for the rubber that the tappers' exchanged and the supplies that they bought in the barraco there was tremendous opportunity for abuse, and many rubber tappers were unable to cancel their debts (da Cunha, 2003a; Mathews, 1879; Woodroffe, 1914; Yungjohann, 1989) Many tappers who migrated from the Northeast of Brazil were already in debt when they arrived at the seringal owing the seringalista for the cost of transportat ion and food as well as the tools, clothing and other items that were needed to produce the first season's rubber production (Reis, 1953) Keeping tappers in a continuous state of debt was a strategy patrons used to hold on to their labor force, which was in chronically short supply (R. Santos, 1980; Weinstein, 1983) The extremely isolated nature of the area and limited communication with the outside world further strengthened the position of the pa tron vis vis the tapper (da Cunha, 2003c) Rubber tappers who traded rubber to other patrons or to itinerant river traders, known as regates were threatened with expulsion from the seringal or other forms of puni shment. Monitors, known as fiscais periodically visited the interior of the seringal to ensure that tappers were tapping the trees properly and not selling rubber to river traders. Many patrons also had enforcers on their seringal staff, known as capang as who ensured that the rules were followed. Hair raising tales of capangas chaining errant tappers to logs, applying lashes, and inflicting other forms of corporal punishment are still told in riverine communities along the Purus. The coercive nature of the regime do toco is part of the lore and legend of the rubber boom and there are many accounts from travelers
53 deploring this debt system as purely exploitative and unjust (da Cunha, 2003a; Labre, 1872; Mathews, 1879; H. H. Smith, 1879; Woodroffe, 1914; Yungjohann, 1989) Although tales of coercive behavior from patrons dominate much of the writing about these relationships, some writers point out that rubber estate owners were also trapped in the debt based ex change system of aviamento (Barham & Coomes, 1996; Benchimol, 1999; C. Costa, 1940; Frank & Musacchio, 2006; Weinstein, 1983) At the beginning of the rubber safra or tapping season, patrons would buy the supplies they needed for their tappers on credit in Manaus and Belm and then repay their debts in rubber at the end of the tapping season (C. Costa, 1940) Tappers who sold their rubber to outside parties or ran away after borrowing supplies could make it difficult for a seringalista to pay his debts. When tappers borrowed supplies and were then unable to pay due to illness or death from disease, which was common, the seringalista was left with the debt. Finally, the extended nature of the credit system meant that rubber prices, currency exchange values, and the cost of supplies might fluctuate wildly from the time a seringa lista purchased his supplies on credit to the end of the season when he paid in rubber (Lange, 1911; Pearson, 1911) Rubber estate owners often charged very high prices on the supplies in their barraco to offset these risks. In summary, there were a series of risks that rubber estate owners assumed that may explain the high cost of supplies and the poor reputation of the debt system (Barham & Coomes, 1996) Therefore, in spite of the way these relationships are often presented, because patrons needed the tappers' labor to repay their debts, power was probably not completely unidirectional on the seringais even during the boom periods. Pa trons had
54 a limited capacity to monitor rubber tappers and if they were too harsh in their treatment of the tappers or too stingy with credit and supplies they might find themselves victims of some of the "Weapons of the Weak" that peasants in many places have used to try to level the playing field with their more powerful patrons (J. C. Scott, 1985) In the rubber forests of the Amazon this often amounted to adulterating rubber with other latexes or with heavy objects to increase its weight, selling the rubber to river traders or other parties, or even abandoning the seringal w ithout paying their debts. Even if the tapper did not resort to these tactics, harsh treatment might compel him to seek out another patron, thereby decreasing the amount of labor that the original patron had on his seringal which was generally already in short supply. For these reasons, conflict with his tappers due to poor treatment would probably have put many seringal owners at risk of not being able to pay back their own debts. It seems more likely that while the system was by no means egalitarian, the relationships between patrons and clients were more nuanced than they are often portrayed to have been, with both sides giving and taking in a relationship in which expectations were periodically reevaluated and renegotiated. To unpack how patron clie nt relationships in Lbrea have changed over time, it is instructive to analyze and compare the exchange of resources between patrons and clients in the regime do toco and compare it to the exchange arrangements that replaced it. Patron to Client Exchang es in the Regime do Toco In the regime do toco system patrons were generally expected to: 1) set up the tapper's house in the forest; 2) provide the tools necessary to tap rubber (on credit); 3) open rubber trails and clean them at the beginning of the se ason; 4) provide transportation (mule trains or canoe) to take the tapper's rubber (and other forest
55 products) to the barraco and bring in supplies; 5) provide long term credit and access to supplies; 6) provide medicine or transportation in times of illn ess (on credit); 7) organize and pay for religious celebrations and election parties, 8) carry out other social services, such as raising orphaned children, often engaging in fictitious kinship relationships as godfather to tappers' children, and 9) provid e protection for the tappers in an often wild, frontier region with virtually no police force (Benchimol, 1999; F. S. D. d. Santos, 2000; F. S. D. d. Santos & Muaze, 2002; Teixeira, 2009) Client to Patron Exchange s in the Regime do Toco In the regime do toco arrangement the tapper was expected to: 1) provide the labor that was necessary (and often in short supply) for tapping rubber and collecting forest products; 2) provide labor for other important off season ac tivities (e.g. cutting firewood for steam ships); 3) give the patron gifts or agradas to show deference from time to time (fish, game, etc.); 4) buy supplies from the patron's barraco ; 5) give their patron the first opportunity to buy any item he was goin g to sell, and 6) in some areas they might be expected to vote (if they voted at all) the way the patron said they should (Benchimol, 1999; Teixeira, 2009) These arrangements varied during different time periods and even from seringal to seringal as some patrons were more benevolent than others. In spite of its often coercive nature, there were reciprocal aspects to the exchange between patrons and clients in the regime do toco The patron provided a constant source of credit and supplies, even in the off season when the tapper had no rubber to exchange, or in other times of crisis. Although on the one hand many respond ents in interviews along the Purus decried the lack of freedom they had in these arrangements, they also lamented
56 the abandonment of the regime do toco because of the security it provided in times of crisis. It seems obvious from the interviews I conduct ed, and another study conducted on the Purus in the late 1990s (F. S. D. d. Santos & Muaze, 2002) that patrons in the regime do toco used the "carrot" as much as the "stick" in their relationships with tappers. For every hair raising story of the cruelty of capanga enforcers there were two mo re that invariably included the refrain no faltava nada nothing was lacking and o meu patro era bom demais my patron was exceedingly good to me revealing the complex nature of these arrangements. The remoteness of Lbrea, the dispersed natu re of the rubber production areas, and the difficulties in monitoring the actions of dispersed tappers meant that purely coercive relations between patrons and tappers were probably only feasible, if at all, in periods when high rubber prices made it possi ble to hire extra staff to monitor tappers' actions and enforce the rules (Weinstein, 1983) Otherwise maintaining the trust and deference of the tappers through benevolent treatment may have been just as, if not more, common than coercive behavior. By providing year round credit and access to supplies even during the months when a tapper was not producing rubber or other forest products, medicines in times of illness, etc. the patron filled a crucial role. F or much of this time the government was non existent, visits from religious groups and politicians were sporadic at best, and there were no civil society organizations operating in the area. In short the patron was the only source of many of the items tha t were essential for survival, hence his position of power and the seemingly endless opportunities for exploitation. However, the patron also depended on the tappers to
57 produce the rubber that he would use to pay his own debts, which might explain the "ca rrot and stick" nature of these relationships. Resource Control in the Regime do Toco Social relationships in which roles, exchanges, or affiliations overlap are said to exhibit multiplexity (Verbrugge, 1979) In the regime do toco patrons and clients had relations with h igh levels of multiplexity. In these multiplex relationships patrons often filled several social roles as: boss, protector, godfather to their children, etc. Their ability to control access to their rubber estate and the fact that there simply were not m any other groups willing or able to provide key resources certainly strengthened their position. An analysis of the control of and access to six important resources can help us better understand the arrangements between patrons and tappers: 1) Market acc ess / credit, 2) Land / resource control / ownership, 3) Health / Education services, 4) Religion, 5) Political access, and 6) Information. In the regime do toco period the patron had almost complete control over all six of these resources and services, p utting him in a position of nearly complete control of those who lived on his seringal Market access and credit: In the regime do toco the patron controlled access to credit, supplies and markets, which was the basis of much of his control over the tapper s. For the most part people living on the seringal had to sell any resources collected to the patron, or at least give him the first option to buy before trying to sell them to someone else. Patrons had relationships with banks, traders, and other source s of supplies that the ordinary tapper did not have. At times patrons would allow residents to sell animal pelts, fish, etc. to someone else, but usually only if they had no use for them. Clandestine exchanges with itinerant river traders, known as regat es also existed, but could lead to harsh reprisals from the patron if discovered.
58 Land / resource control / ownership: During the rubber boom period in Lbrea seringal owners took the time and money to obtain official land titles that recognized their claims. During this period there was very little questioning about the validity of a patron's claim to the seringal (unless it was in a conflict with another seringal owner over boundaries). During interviews in 2008 2009 older inhabitants said that they had no way of knowing that many of these land claims were false or inflated. They said that all of the resources on a particular seringal belonged exclusively to the owner and tappers from one seringal would not hunt or fish on another seringal without e xplicit permission from the owner. This was a widely recognized norm. In spite of the large distances that made monitoring difficult, the leverage that the patron had in being able to expel a tapper that broke the rules, or limit his access to credit and supplies, were apparently quite effective in maintaining order. Patrons normally had close relationships with the few police officers in town and could call on them to evict a tapper that broke the rules (F. P. Costa, 2005) Patrons also made sure that outsiders did not take resources from the seringal without permission. Health and education: In the regime do toco the patron also had control over a tappers' access to health and education. Medicines were generally only available, if at all, in the barraco and patrons could also arrange for transportation to the city in case of more serious illness. The municipal government had extr emely limited capacity to provide health care in urban areas and no capacity to provide services in rural areas. Most of the seringais did not have even rudimentary education, but those that did only had schools if the patron allowed it.
59 Religion: Since patrons controlled access to the seringal they also controlled access to religious services. When priests periodically visited the interior to conduct religious services they arranged these visits with the seringal owners. Usually they stayed in the pat ron's house and often conducted services in and around the barraco Until the ideology of liberation theology became popular in the late 1960s, the Catholic Church was not known for questioning social hierarchies and in Lbrea the church did not openly q uestion the hierarchies of the seringal system until the 1970s (personal communication Bishop of Lbrea). Political representation: Politicians who were campaigning in the interior also coordinated their visits with the seringal owners. Typically these visits were limited to campaign cycles if they occurred at all. Patrons maintained strong relationships with politicians, unless they were political enemies, and had much easier access than tappers to politicians (F. P. Costa, 2005) As landowners and businessmen they were often sought out for financial support during campaigns or for their ability to deliver the votes of the tappers on their seringal making their friendship important for politicians to acquire. As in the case of the priests, politicians visiting during campaigns would stay in and around the barraco or patron's house during their visits. Patrons often organized par ties at election time in which food and drink would be provided for the tappers who came in from the interior of the seringal to vote. Information: In days before portable transistor radios, much of the information that came from the outside world, like everything else that came and went from the seringal would pass first through the barraco or patron's house. There is no doubt that information passed between tappers, but it was certainly hindered by the dispersed
60 nature of the seringal layout. In som e parts of the Brazilian Amazon tappers would leave the seringais in the rainy season and stay in the nearest town until the next tapping season (Lange, 1912) in these ca ses tappers must have had more access to information about the price of rubber and supplies than those who lived full time on the seringal Table 3 1. Control of Key Resources in the Regime do Toco Resource Controlled / Provided by Credit / supplies / market access* Seringal owner Land / resource control Seringal owner Health and education Seringal owner Religious services Seringal owner Political representation Seringal owner Information Seringal owner *The main exception to the patrons' contro l during the regime do toco may have been interactions with the itinerant river traders, who provided a clandestine source of exchange for goods, information, etc. However, as previously mentioned there was an element of risk in engaging in these relation s. Until relatively recently the tremendous physical distances from the seringais to the small interior towns like Lbrea strengthened the position of the patrons and river traders vis vis their clients. Tappers were often separated from urban center s by several days, if not weeks, of hard rowing in open canoes. During boom periods there were many more ships going up and down the rivers that tappers could travel on, if they had the resources to do so. During interviews conducted in 2008 2009, I was told that there were many fewer boats on the rivers than before; however, this lack of boats was offset to some degree by the proliferation of small outboard rabeta motors that had greatly reduced the amount of time it took inhabitants to get to town. Du ring the boom periods, credit and supplies were considered the most important resources over which patrons maintained control and used as leverage in their relationships with their clients. When rubber prices were high, patrons could buy
61 supplies on exten ded credit and pass it on to the tappers; however during economic busts credit restrictions and low prices often dramatically changed the ability of the patrons to provide credit to the tappers, which often placed the legitimacy of the rigid regime do toco exchange arrangements in question. Boom Bust Cycles in Lbrea The rubber boom that had led to the founding of Lbrea and the institutionalization of the aviamento system in the area was greatly impacted by the development of rubber plantations in Asia fro m seeds spirited out of the Amazon by the British in the 1870s (Dean, 1987; Jackson, 2008; Weinstein, 1983) Expanding plantation production in Asia drastically lowered the price of rubber on international markets, sending a shock wave down the aviamento credit system. Rubber producers could no longer repay their debts and the defaults led to bankruptcies all the way up the aviamento chain. The rubber tappers at the bottom of this chain of debt were often in a per ilous position as their source of foodstuffs, medicine, clothing, and other essential items began to disappear. The situation became so desperate that in the years following the bust of 1910 11 there was a mass exodus of workers from many of the seringais to urban areas like Manaus and Belm (Benchimol, 1999) Many rubber tappers eventually decided to return to their places of origin in the Northeast of Brazil (Benchimol, 1999) Labor, which had always been scarce, became insufficient on many of the old rubber estates to keep them operating (Schurz, Hargis, Marbut, & Manifold, 1925) Howeve r, rubber production continued and even increased in some years of the 1920s. Although the 1930s were a decade of extremely low rubber production in the Brazilian Amazon, production never ceased entirely, nor did the patron client relationships that had b ecome entrenched in the boom years.
62 Along the Purus those that stayed in the rubber estates often had to dramatically shift their production activities in order to survive. Many inhabitants began to do agriculture on the fertile, sediment rich beaches o f the Purus that appear in the low water season (Gondim, 2001) Others did their best to survive by fishing, collecting turtles, or wor king as Brazil nut collectors (Gondim, 2001) Many of the essential supplies that had once been provided by rubber estate owners were n ow furnished by river traders, who continued the practice of exchanging them for forest and agricultural products on credit (Gondim, 2001 ) Seringal owners who were able to diversify their production to incorporate Brazil nuts and other products were sometimes able to maintain a barraco with supplies and continue with the regime do toco Many of those who could not abandoned or sold the ir seringais and others decided to dramatically renegotiate the exchange arrangement they had with the tappers who worked for them. These renegotiated relationships implied a whole new series of obligations and expectations between patrons and clients. The regime do toco appears to have dominated during the times when rubber prices were high (such as the pre 1910 boom years) and in the periods when government subsidies made credit available to patrons (such as during World War II and in the years followi ng the conflict). When rubber prices dropped significantly or when subsidies were not available, new exchange arrangements between patrons and tappers often developed. From the Regime do Toco to Arrendamento When rubber prices fell, the aviamento system o ften faltered, and in some cases collapsed altogether, which led some rubber estate owners to modify their demands vis vis the production of rubber and their obligations to provide supplies on credit. In these arrangements, often called arrendamento or rent, tappers were no longer
63 obligated to sell all of their production to the patron and they were not obligated to buy supplies in his barraco (Teixeira, 2009) Instead, tap pers paid a flat rate "rent" in rubber for each of the trails they tapped to their patron and could sell the rest to whomever would buy it (Teixeira, 2009) Antonio Loureiro's (1981) book, based on newspaper archives from the town of Sena Madureira, further up the Purus, provides fascinating accounts of how patrons went as far as putting advertise ments in the newspaper in the attempt to rent their rubber trails after the big crash. Although patrons were demanding less in this system, they also significantly reduced the services they provided to the tappers. Once again we can analyze these patron client relationships by unpacking the exchange that took place between patron and client. Patron to Client Exchanges in the Rent System In the "rent" system the patron: 1) allowed the tapper to use resources on his land; 2) may or may not have made supp lies available in the barraco ; 3) may or may not have provided transportation for rubber and supplies; 4) probably would not maintain the rubber trails; 5) may or may not have provided credit but typically would reduce credit or reduce the amount of time given to repay the debt; 6) would probably not organize or hold parties / festivals; 7) may or may not have provided medicine and /or transportation in case of emergency (Schurz et al., 1925; Teixeira, 2009) In the "rent" system, land and resource control became the patron's main leverage over the client, who often traded directly with river traders but still paid rent. Client to Patron Exchanges in the Rent System The main obligation of the tapper / client in this system was t o pay the "rent" for each trail he was tapping. However, the option to buy supplies on credit in the barraco was often greatly reduced, or eliminated altogether. On many seringais in Lbrea
64 patrons began to physically move from the seringais to town and may or may not continue to operate a barraco In this system the tapper often had to find another source of the supplies, which were vital for survival in these remote areas. In Lbrea river traders increasingly became the source of these supplies. H owever, this meant developing a new line of credit with the river trader for periods when the tapper had no rubber or other forest products to trade for supplies. Depending on the quantity of "rent" expected for each trail from the patron, the ability of the tapper to buy supplies was often reduced. I was told that some patrons demanded the renda be paid with the first rubber produced during the tapping season, which is often when the trees produce best. Chagrined tappers said that they had to try to make a living with the rest of the season's production, often the period when the trees were less productive. In the regime do toco the demands of the patron were offset to some degree by the constant availability of credit and supplies. In the rent system t his safety net was often largely withdrawn, which often put the tapper's subsistence in doubt. When a patron begins to make demands and no longer provides the resources that enable their client to maintain even a subsistence existence, the client often be gins to question the legitimacy of the arrangement (J. Scott, 1977; J. C. Scott & Kerkvliet, 1973) In the regime do toco credit and supplies were the patrons' main source of control and leverage in their relationships with their clients. However, during rent periods when tappers were free to sell surpluses to whomever would buy them, this was not the case. During interviews along the Purus in 2008 and 2009 older residents explained that they
65 paid rent because the patron said he owned the land and everything on it and he would get the police to expel them if they did not pay. Resource Control in the Rent System At this point it is instructive to once again analyze the six resources that were controlled by seringalistas in the regime do toco and co mpare them to the "rent" system. Credit / supplies / market access: Increasingly tappers in the "rent" system had to look for other sources of credit and supplies, although some patrons still maintained barraces and were able to provide supplies on credit In Lbrea the river traders increasingly became the source of supplies for many inhabitants of the interior. They entered into arrangements with the river traders similar to those they had had with the seringal owners, receiving supplies often on cre dit in exchange for forest products. Inhabitants often expanded their agricultural production to offset the shortage of foodstuffs. Land / resource control / ownership: In the "rent" system the seringalistas often no longer maintained a physical presenc e on the seringal ; many moved to urban areas, and left a gerente or manager in charge of the seringal The infrastructure on the rubber estate was often abandoned and left to slowly deteriorate. In some cases the manager did not live on the seringal eith er and only passed sporadically to collect the rent. In some cases instead of hiring a manager the seringal owner would lease the area to an arrendatario who would pay a fee to the owner and charge the "rent" himself. By this time the capangas that had o nce been so feared were long gone but seringal owners could still threaten eviction to those who did not pay rent, through their relationships with local police. Generally, however, the tappers had much more freedom with the way they used resources than in the regime do toco system. Land tenure
66 arrangements, which had always been uncertain, became even more so when the seringal owner no longer lived on the estate. Health and education: The closing of the barraces in this period often removed the limited a ccess that inhabitants had had to medicines and many people started to turn to "traditional" medicines to offset this problem (F. S. D. d. Santos & Muaze, 2002) Municipal agencies in some cases provided some basic health services by periodically sending doctors and dentists to the communities Religion: The period of the final shift from the regime do toco to the rent system in Lbrea coincided with a shift in the policy of the Catholic Church regarding its work with the poor. In the 1960s the church hierarchies in many parts of Latin Americ a adopted the tenets of Liberation Theology, and began to associate the fate of the poor with social inequalities (Berryman, 1987) Priests in Lbrea were no longer willing to ignore the exploitation that they felt was occurring between patrons and rubber tappe rs. In the 1970s the Prelacy of Lbrea began to form Christian Base Communities in many of the communities that were forming on the former seringais They began to form direct lines of communication with the inhabitants, no longer deferring to the hierar chies of the seringal system, in which they would arrange everything through the patrons. They began to carry out literacy campaigns and programs to elect community leaders. Priests in Lbrea were instrumental in forming the Rural Workers' Union of Lbre a in 1979. In this period they open a Pastoral Land Commission office in the Prelacy to address land tenure issues, and began to get involved in indigenous rights issues in the municipality.
67 Political representation: Contact with politicians was still g enerally limited to political campaigns, but in the absence of the patrons politicians increasingly began to communicate with community leaders and inhabitants. Information: People had access to radios in some cases, interactions with other community memb ers, and increasingly with the priests who were forming the Christian Base Communities. The budding social movement organizations also began to provide crucial information. For example, the Rural Workers' Union and Pastoral Land Commission agents began t o inform inhabitants of their rights and of the false nature of the land claims of many patrons, who often charged "rent" from families when they had no valid title to the lands upon which the families lived. Table 3 2. Control of Key Resources in the Ren t System Resource Controlled / Provided by Credit / supplies / market access Seringal owner (or no one) Land / resource control River traders (some seringal owners) Health and education None (or limited government services) Religious services Direct c ontact with priests Political representation Direct contact with politicians (limited) Information River traders, priests, etc. The rent system was common during periods when rubber prices were low and patrons could no longer provide supplies on credi t, including the period after the first rubber crash, most of the 1920s and 1930s, and then again in periods after World War II when government subsidies were withdrawn. As previously mentioned, the regime do toco and rent systems often existed side by si de, depending on several factors particular to the situation of each seringal Under the rent system, control changed drastically; patrons no longer demanded exclusive exchange arrangements but they were also less willing (or able) to provide necessary su pplies and credit in times of need. Patrons often moved to nearby towns and were even less likely to be able to
68 monitor resource use, although they sometimes hired gerentes or managers to do so. Tappers began to expand their interactions with river trade rs, who often became their primary source of credit and supplies. The patron client relationships in these cases became less multiplex in many cases as tappers could no longer obtain everything from a single patron. Subsidized Rubber Production During W orld War II the governments of Brazil and the United States attempted to revive natural rubber production in the Amazon to offset the capture of the Asian rubber plantations by the Japanese (Martinello 2004; Secreto, 2007) This government program pumped money back into the rubber economy and led to a new wave of migration from the Northeast of Brazil. So called soldados da borracha or rubber soldiers were sent by the thousands up the Purus and into Acre to tap rubber as part of the war effort. Although this program lasted only a few years, it led to the establishment of a series of the rubber subsidies in the Amazon that enabled many of the old rubber production estates to keep functioning. With the national demand for rubber expanding in Brazil due to its own industrialization period, beginning in the 1950s, there was an incentive to maintain natural production in the Amazon to decrease Brazil's dependence on foreign rubber. The PROBOR rubber pr oduction programs were periodically renewed and included monies for: seringal improvements (infrastructure), loans for operating costs (supplies), and funds for opening and maintaining rubber processing facilities (Lbrea had one of these) ("PROBOR II Segundo Programa de Incentivo Produo de Borracha Natural: Fundamentos Objetivos Normas Operativas", 1978; "Programa de Incentivo Produo de Borracha Vegetal Normas Operativas
69 Aplicveis ao Financiamento de Projetos PROBOR ", 1972) In an ominous twist for Amazonian production these subsidy programs also included funding to expand the rubber plantations in So Paulo state and in Bahia that could produce greater quantities per hectare at lower costs (Hoelz & Martinez, 1972; "PROBOR II Segundo Programa de Incentivo Produo de Borracha Natural: Fundamentos Objetivos Normas Operativas", 1978; "Programa de Incentivo Produo de Borracha Vegetal Normas Operativas Aplicveis ao Finan ciamento de Projetos PROBOR ", 1972) The End of Subsidies and the End of the Seringal Rubber subsidies kept production alive in Lbrea into the 1980s; as late as 1982 the Beneficiamento de Borracha Ltda. plant in Lbrea sold approximately 2 million t ons of processed rubber and there were 120,000 hectares of planted rubber producing latex in the municipality (Peret, 1985) However, a series of financial shocks in the late 1 980s and early 1990s would finally bring an end to the seringal system and dramatically alter the surviving patron client relationships between tappers and seringal owners. In the 1980s and early 1990s Brazil suffered from rampant inflation and a debt cr isis that led to dramatic changes in economic policy. Brazil's currency was changed several times in a few short years and economic austerity measures were applied. The rubber industry in Brazil, which had enjoyed protected status for years in the form o f high import tariffs on Asian rubber and the PROBOR subsidies, soon found itself without this support (Cardoso, 2009) Without these subsidies natural rubber production in the Amazon was not able to compete with rubber produced on plantations in Asia or those that had been planted in Bahia and So Paulo (Brooke, 1995) I was informed in an interview that the rubber production plant in Lbrea closed down for good in 1992. Beginning in 2003, state and federal subsidy programs were implemented and a few
70 hundred tappers in the interior of the municipality began tapping rubber again as a complement to other activities. Although some former owners and arrendatrios still attempted to charge rent, the prices of rubber without the subsidy programs became so low that it was no longer worthwhi le to take the time to try to charge rent. Patrons whose seringais had extensive Brazil nut groves still had an incentive to charge rent from the occupants, and some continued with this practice in 2008 2009. Once timber and other resources had been extra cted, many former seringal owners completely abandoned their claims to these areas, leaving the former rubber tappers to fend for themselves. I learned that the sharp fluctuations in currency rates and inflation were also devastating for many of the river traders who had continued to furnish supplies in exchange for forest products. These shocks greatly impacted the aviamento credit system that people in the interior of Lbrea, and throughout rural Amazonia, still depended upon (Aramburu, 1994) From Arrendamento to Seringal Abandonment With the end of government subsidies in the l ate 1980s and early 1990s, the price of rubber fell so low that most people in Lbrea stopped tapping it altogether. A mass exodus from the rural villages along the river to urban areas that had begun years earlier intensified. Table 3 3 shows the dramat ic rural urban migration trend in Lbrea from 1970 to 2000 compared to Amazonas state and Brazil. With these dramatic socioeconomic changes, even the arrendamento arrangements began to break down. In other cases landowners continued to try to extract rent payments from the inhabitants of the former rubber production areas, many of whom were now involved in the timber industry, fishing, or collecting Brazil nuts. Those who remained and did not move to urban areas began to question even more the
71 exchanges w ith patrons that now had virtually no benefits for them. Once again it is instructive to analyze the exchange between patrons and clients in this period. Table 3 3. Rural Urban Migration in Brazil, Amazonas, and L brea 1970 to 2000 1970 1980 1991 2000 Brazil Total 93,134,846 119,011,052 146,825,475 169,799,170 %Urban 55.9% 67.6% 75.6% 81.2% %Rural 44.1% 32.4% 24.4% 18.8% Amazonas Total 955,203 1,430,528 2,103,243 2,812,557 %Urban 42.5% 59.9% 71.4% 74.9% %Rural 57.5% 40.1% 28.6% 25.1% Lbrea Total 16,737 22,026 33,052 28,956 %Urban 18% 37.5% 46.7% 66.6% %Rural 82% 62.5% 53.3% 33.4% (IBGE Census Data from Oct.2009) Patron to Client Exchanges in Seringal Abandonment At this point many of the old patrons had long abandoned the area and in m any cases had even stopped attempting to charge rent. The barraces had mostly disintegrated back into the tropical forest and inhabitants had turned to river traders to exchange their products and acquire supplies. In places that had once been dominated by a single patron who had been the tappers' link to the outside world, people were now beginning to form relationships with different groups that could help them with different needs. In some cases new patron client ties began to form to replace those t hat had existed on the rubber estates. Client to Patron Exchange in Seringal Abandonment With the fall in prices in rubber and other extractive products, the refusal (or inability) of former seringal owners to provide supplies on credit, and the knowledg e that they had been charged "rent" on lands to which patrons often had no legal title, the inhabitants began to refuse to pay rent and to rebel against the patron client ties that no longer appeared to provide even minimal benefits. Inhabitants of the vi llages that had formed in these areas were increasingly interacting with social movements, religious
72 organizations, and government agencies that were encouraging them to organize and fight for their rights. In this abandonment phase the services that had once been provided or controlled almost exclusively by the seringalistas were further divided among a new set of actors. Credit / supplies / market access: For most inhabitants of the interior the river traders became their main access to markets and su pplies on credit. Most of the inhabitants of the interior greatly reduced, and in some cases stopped altogether, the collection of forest products and instead began to engage in fishing and the production of agricultural products, particularly manioc and beans. Improved access to outboard motors made access to markets in town easier and inhabitants living closer to Lbrea began to purchase supplies directly from shop owners in town. Land / resource control / ownership: By this time most of the barraces t hat had been at the center of the patron's power and proof of his ownership of the area were long gone. Even the periodic visits by patrons, managers, and arrendatrios had ceased in many communities. The land tenure situation became even murkier as inha bitants discovered they had some nominal land rights based on their occupation of the land, but these rights were generally weak and inhabitants were constantly unsure if they would be evicted at some point. The federal government began to create extensiv e indigenous territories that implied the eviction of any non Indians living on them, including many riverine communities comprised of former rubber tappers. Government agencies began to have an increased presence in the municipality and became more invol ved in land issues. In some cases this led to a situation of "open access" in which outside fishermen, loggers, and other groups began to use resources
73 that the people living in the villages felt belonged to them. Resource conflict was inevitable in these situations. The government agency charged with enforcing Brazilian environmental laws, IBAMA, began to monitor resource use in Lbrea more strictly in the 1990s. The Pastoral Land Commission and other social movements began to try to help resolve resour ce use conflicts in these areas. Health and education: In the 2000s the municipal and state governments began to invest in education in the interior and inhabitants of the villages on the former seringais began to gain access to primary education service s. Health care continued to be limited to occasional visits by a medical barge, usually during campaign season, and to the malaria control efforts carried out by municipal and federal agencies. Religion: The policy of the Catholic Church regarding the C hristian Base Communities began to change in the 1980s and 1990s as Liberation Theology became less popular with the church hierarchy. However, the pro poor work of the Catholic Church in Lbrea continued and new organizations and projects were establishe d to help the inhabitants of the interior; in urban areas they also helped the flood of people arriving from the rural areas. The Pastoral Land Commission (CPT) continued to organize community leaders and help them with land tenure issues. In the 1980s an d 1990s there was an explosion of Evangelical churches in Lbrea and great number of people began to convert from Catholicism. This mirrors a larger trend in Brazil, which has witnessed a tremendous expansion in the number of Evangelical churches (Stoll, 1990, 1993) Political representation: Interactions with politicians continu ed to be limited to campaign periods for the inhabitants of the villages in the interior. In many cases these
74 sporadic interactions were characterized by the exchange of votes for supplies, diesel motors, and other resources. For inhabitants able to trav el to the town of Lbrea, the mayor's office and city council often became places to ask for food, medicine, transportation, and other items. Information: Social movement organizations increasingly supplied information to inhabitants. In the 1990s gover nment agencies began to help communities form community associations in an attempt to help them market and sell their products and also to access government assistance programs. One of the main reasons for the formation of these organizations was the inju stice and exploitation of the old patron client system that had existed between seringal owners and tappers, and the need to replace them with more just interactions. Government agencies also brought Table 3 4. Control of Key Resources in Seringal Abando nment Resource Controlled / Provided by Credit / supplies / market access Federal gov. agency (or no one) Land / resource control River traders (some shop owners) Health and education Municipal and fed gov. agencies Religious services Direct contact w ith priests Political representation Direct contact with politicians (limited) Information Soc. mov., river traders, priests, etc. information regarding potential solutions to the unclear land tenure situation of people living in the villages, particu larly the possibility of creating extractive reserves and agricultural settlements. Inhabitants of the riverine communities were increasingly worried about being expelled from the lands they occupied, as a result of the creation of indigenous lands, and their leaders were actively looking for solutions to their land tenure uncertainty.
75 From Abandonment to State Control The collapse of the rubber industry, and others based on forest products, in the Amazon coincided with a rapid increase in the presence of social movements, church organizations, NGOs, and government agencies that began providing services and assistance in a way that often directly challenged or replaced the old patron client ties. The legitimacy of the patron client exchange arrangements had been called into question by clients whose patrons provided fewer and fewer services in exchange for forest products and/or "rent" payments. Information about the dubious nature of many patrons' land claims further weakened the legitimacy of the old arrangements for many inhabitants. In Lbrea the role of the Catholic Church was fundamental in the emergence of social movements that would challenge the patron client system that dominated the area. In the early 1970s the priests in Lbrea decided that they would begin working directly with the communities on the seringais to form Christian Base Communities. They held meetings directly with the workers on the seringais and no longer coordinated their work with the patrons. The decision to organize the CEBs was made by the local priests, but was inspired in part by the tenants of Liberation Theology that emerged from the Vatican II decree of Pope John XXIII and the ideas generated at the meeting of Latin American bishops in Medelln in August 1968 (Berryman, 1 987) Part of this work included the formation of community leaders that would help organize and teach other people in the communities (Boff, 1985) In addition to spreading religious teachings and literacy to their neighbors, these leaders began to act as "brokers" between people in their communities and the rest of the world. People from government agen cies, politicians, etc. typically sought out community leaders when they arrived in the
76 communities or when they wanted to spread information. The community leaders would often travel to Lbrea to meet with people in social movement and government agencie s to resolve issues on behalf of their communities or individuals in the communities. In the late 1970s one of the priests in Lbrea began to help organize community leaders and they formed the Sindicato de Trabalhadores Rurais de Lbrea (Rural Workers' Union of Lbrea), which established links with national rural workers unions like Federao dos Trabalhadores da Agricultura (FETAGRE) and Confederao Nacional dos Trabalhadores na Agricultura (CONTAG). The union directly challenged the injustices that we re occurring on the seringais Rubber tappers brought their debt ledgers to the union with complaints of being cheated. On several occasions when patrons threatened to throw tappers off their seringais for one infraction or another the union intervened, even providing legal counsel on several occasions. In the 1970s indigenous people in the municipality of Lbrea began to demand the creation of Indigenous Lands as a way of reestablishing their rights to the lands and resources that they had lost during t he expansion of the rubber economy in the Amazon in the late 19 th century. The Catholic Church of Lbrea opened a Comisso Pastoral da Terra (CPT) office in the early 1980s to address land conflicts in the area and a Conselho Missionario Indigena (CIMI) o ffice to address human rights issues involving the indigenous peoples in the region. The Catholic Church worked with these newly created social movement organizations to set up community associations that were designed to provide an alternative for people to buy merchandise and sell their production. These associations
77 were created to replace the patron client relationships that had become institutionalized in the seringal production system. After the year 2000, regional social movement organizations like the Grupo de Trabalho Amaznico (GTA) and Conselho Nacional de Seringueiros (CNS) also established a presence in Lbrea. These groups saw the creation of Extractive Reserves as a way of maintaining extractivist livelihoods, breaking unjust patron client ties, and conducting land reform. Since the 1980s these goals had been joined with the goals of national and international environmental organizations, which saw the Extractive Reserves as a way to maintain forest cover and conserve biodiversity. In ad dition to the rise of social movements in Lbrea, there was a shift in the way many state and even some national government organizations conducted their work. State government agencies in Lbrea like the Instituto de Desenvolvimento Agropecurio e Florest al Sustentvel do Estado Amazonas ( IDAM) and Agencia de Desenvolvimento Sustentvel do Amazonas (ADS) now coordinate a large part of their activities through community associations, which they often help to create and legalize, and often form partnerships with local social movement organizations. A wide array of rural credit programs designed to benefit small producers are accessed through government agencies like IDAM. Over the last ten years, social programs like Bolsa Familia became available to people in Lbrea and became a vital source of income for many families in the wake of the collapse of extractive industries. Bolsa Familia was a cash transfer program that benefited approximately eleven million Brazilians (Soares, Ribas, & Osrio, 2010)
78 Local government agencies b egan to increase their presence in the communities at about the same time that the seringal system began to collapse. This was about the same time, 1985, in which Brazil returned to democratic rule. However, the contact was (and is) often limited to the election season when candidates travel to the communities to campaign. Vote buying was (and is) common and communities often feel abandoned once the politicians are elected. For the most part, politicians only visit the communities during campaign activi ties. State Control and Patron Client Ties In May 2008 the Mdio Purus Extractive Reserve, which covers 604,290 hectares and containing more than ninety communities, was created. In June 2008 the Rio Ituxi Extractive Reserve, covering 776,940 hectares and eleven communities, was created on the Ituxi River. The inhabitants of these reserves, many of whom worked on the old rubber estates and had patron client ties with estate owners, will see yet more changes in their social interactions. Credit / supplie s / market access: The extractive reserve model also typically favors one or more community associations to be the direct link between the inhabitants and the government agencies that provide assistance. Project monies and resources are typically channele d through this "mother association" or associao me on to the inhabitants. The community leaders that head these associations are a new power within the reserves and may often become the new link that people have to go through to access government progr ams and project funds. Many government agencies also hope to replace the old river traders and landowners, who are perceived to be exploitative in their relations with reserve inhabitants, with community associations. However, at the time this research w as conducted it was obvious that most of the newly
79 created community associations had neither the operating capital nor the business contacts to buy and sell with the riverine communities in the reserves. River traders and storeowners in Lbrea were still the main market access that riverine communities depended on to obtain supplies in exchange for forest and agricultural products. Land / resource control: The creation of an extractive reserve implies almost complete government control over the natural re sources found within the reserve. Communities have usufruct rights to resources in the reserve but they do not own the land. With this arrangement the old patron client ties with seringal owners should be a thing of the past. However, the communities wi ll now have to answer to the government officials in charge of co managing the extractive reserve. Theoretically they are required to have management plans in order to utilize the resources found within the reserve, implying the elaboration of expensive a nd time consuming official documents. Health and education: Municipal and federal government increasingly provided these services. Since the year 2000 the municipality of Lbrea had built one room school buildings in most of the communities along the ri ver. However, since the schools in the riverine communities were limited to the primary school grades, children had to live in the town of Lbrea if they wanted a higher education. Health services, one of the major worries for riverine community inhabita nts, were still non existent outside of the urban center except for periodic visits from the municipal health boat and the FUNASA indigenous health agency. Religious services: Priests and nuns visited the riverine communities on the Purus approximately t wo to three times per year to conduct religious services. Many inhabitants of these communities also had contact with them when they visit the town of
80 Lbrea. A boom in the number of converts to Evangelical faiths had also changed the religious dynamics in these areas. Pastors lived in some of the communities and in many Evangelical churches were being built alongside Catholic churches, or replacing them all together. Political representation: Politicians often depended on relationships with community l eaders as a way to reach the wider population. However, they did occasionally visit the communities during political campaigns. Information: Social movement groups like the Pastoral Land Commission and the National Council of Rubber Tappers increasingly provided information in riverine communities, including information about the extractive reserves, community associations, government programs, land tenure, and other issues. River traders, politicians, and other groups often provided their own informati on, which often contradicted the information provided by social movement groups, particularly regarding the creation of extractive reserves, which many of these actors did not want to see created. Table 3 5. Control of Key Resources under State Control Re source Controlled / Provided by Credit / supplies / market access River Traders / Gov. agency Land / resource control Community associations (river traders) Health and education Municipal and fed gov. agencies Religious services Direct contact with pr iests Political representation Direct contact with politicians (limited) Information Soc. mov., river traders, priests, etc. Chapter Conclusions The Brazilian exploration of the Purus River and the establishment of the municipality of Lbrea in the se cond half of the 19 th century coincided with the Amazonian Rubber Boom. The patron client relationships associated with the
81 aviamento credit system and the seringal rubber production unit dominated Lbrea's economy for approximately one hundred years. Pa tronage and aviamento continued even during periods when rubber lost value and people were engaged in the extraction of other forest products and family agriculture. Through the years patron client relationships changed based on: 1) the ability to control and provide key resources, 2) the economic ebbs and flows that altered resource control and exchange expectations between patrons and clients, and 3) the alternatives (or lack thereof) to the patron client relationships that were established during the ru bber boom. Table 3 6 below summarizes the changes in the control of resources and provision of services that occurred over time with socioeconomic changes that included: the booms and busts of an economy based on the extraction of forest products (particul arly rubber) and the arrival of government agencies and social movements that challenged the position of the seringal owner patrons as the sole provider of necessary resources and services. In the rigid regime do toco exchange system, rubber tappers were expected to maintain exclusive resource exchange arrangements with a single patron (forest products for supplies). Although they often demanded exclusivity in their exchanges, patrons often provided a safety net in times of need in remote areas where no other help was available. Patrons and clients had highly multiplex relationships, which included the exchange of a range of resources and services. In the more relaxed arrendamento or rent arrangements, patrons charged rent for the right to use resource s, but no longer demanded exclusive resource exchange. Once they paid their rent, tappers could sell their products to whomever would buy
82 them. Although they no longer demanded exclusivity, in the rent arrangements patrons were less likely to provide cre dit or serve as a safety net in times of crisis, which led many tappers to begin to question the legitimacy of these exchanges. Instead of dealing with a single patron who controlled most of the resources and services available to them, tappers began to f orm (out of necessity) relationships with whomever could or would provide vital resources and services. Table 3 6. Control/Provision of Key Resources in the Regime do Toco Resource Controlled / Provided by Credit / supplies / market access* Seringal o wner Land / resource control Seringal owner Health and education Seringal owner Religious services Seringal owner Political representation Seringal owner Information Seringal owner Control/Provision of Key Resources in the Rent System Resource Cont rolled / Provided by Credit / supplies / market access Seringal owner (or no one) Land / resource control River traders (some seringal owners) Health and education None (or limited government services) Religious services Direct contact with priests Po litical representation Direct contact with politicians (limited) Information River traders, priests, etc. Control/Provision of Key Resources in Seringal Abandonment Resource Controlled / Provided by Credit / supplies / market access Federal gov. agenc y (or no one) Land / resource control River traders (some shop owners) Health and education Municipal and fed gov. agencies Religious services Direct contact with priests Political representation Direct contact with politicians (limited) Information Soc. mov., river traders, priests, etc. Control/Provision of Key Resources under State Control Resource Controlled / Provided by Credit / supplies / market access River Traders / Gov. agency Land / resource control Community associations (river trader s) Health and education Municipal and fed gov. agencies Religious services Direct contact with priests Political representation Direct contact with politicians (limited) Information Soc. mov., river traders, priests, etc.
83 The final collapse of the s eringal system due to the removal of rubber subsidy programs and tariffs on rubber imports in the late 1980s and early 1990s caused yet another shift, as most of the old seringal owner patrons completely abandoned the seringal properties, and many of the r iver traders also went out of business. The Catholic Church began organizing Christian Base Communities and no longer deferred to the hierarchies of the old seringal system. During this phase government agencies increased their presence in Lbrea and beg an to provide credit through government programs, an alternative to the seringal patrons that had not been available in the past. Social movement groups also began to arrive at this point and also became potential alternatives for information and assistan ce with community organization. As this research was being conducted many of the riverine communities were incorporated into federal extractive reserves, which could once again change the balance of resource control and the provision of services to one i n which the government would play an even greater role. In this scenario, instead of deferring to the resource use rules imposed by a seringal owner, community inhabitants may need permission from government officials to conduct certain resource use activ ities. While this research was being conducted, government agencies and social movement groups were trying to help communities form associations and replace traditional market actors, like river traders. However, these associations were still struggling to organize their business operations and had not yet been able to replace the river traders as a source of credit and supplies. Over the past fifty years Lbrea's socioeconomic systems underwent dramatic changes. The patron client relationships that be came entrenched in the region's social
84 fabric, on the one hand seem to have dissipated along with the seringal rubber production system on which they were based. As patrons lost the ability to provide access to credit and control resources, they lost thei r ability to enforce the exclusive nature of their exchanges with their clients. As Martin (2009) points out, this often leads to a situation in which the patron can no longer enforce the "one patron" rule. This led to a breakdown in the patronage pyramids that had become establi shed during the rubber era. In areas where rubber estate owners used to control almost all of the resources necessary for survival, inhabitants began forming relationships with new actors (including government agencies, religious organizations, and socia l movements) that increasingly provided the resources once controlled by single patrons. However, in many cases the direct exchange of forest products and crops from family agricultural plots continued to follow the patterns established during the rubber boom period, and for many inhabitants, river traders and local merchants continued to provide credit and supplies as part of patron client relationships based on informal exchange agreements. Early studies of patronage predicted that once weak state condi tions gave way to those in which a strong state with "modern" bureaucratic systems took over, clientelism and patronage would tend to decline (Waterbury, 1977) Later, researchers found that patronage and clientelism did not necessarily disappear with modernization, but simply changed form and were adapted to the new socioeconomic reality ()*%+,&-"!"#$%&' Ayata, 1994) Some scholars have speculated that the activities of "new social movements" combined with the rapid proliferation of access to print and news media will undermine the foundations of clientelism (Gay, 1998) Other authors have predicted the
85 decline of patron client ties when developing countries integrate into modern capitalist economies (Gough, 1981; Vercruijsse, 1984) T he presence of the state and of civil society certainly increased in Lbrea over the past fifty years. In some cases people simply traded in "traditional" patron client arrangements, in which landowners who controlled key resources (land, markets, etc.) we re the patrons, for "new" patron client arrangements in which a new set of actors (government agencies, NGOs, etc.) controlled the key resources and hence were in a position to become the "new" patrons (Roniger, 1994b) As highlighted in Tab le 3 6 above, highly multiplex relationships in which one patron provided access to markets, credit, health care, and other resources and services were replaced by less multiplex relationships in which inhabitants had to seek out a variety of different act ors to obtain these necessities. Areas that had once been the privately owned fiefdoms of an oligarchy of seringal owners, by 2009 had been converted into federally controlled conservation units that implied an entirely new resource control hierarchy. Se rvices, like health care, which seringal owners had once provided (if they were provided at all) were increasingly being provided by state and federal agencies (albeit in an extremely limited fashion). These changes from dyadic, highly multiplex patron cl ient relationships to a multitude of relationships, each with fewer obligations but also providing fewer resources, is not unique to Lbrea. Similar changes have occurred in other parts of Brazil, in other Latin America countries, and in other regions of the world (C. Escobar, 1994; Roniger, 1990, 1994b; Silva, 1994) On the one hand the "new" system appeared to be more equitable in many ways, and isolated riverine peoples had more freedom to engage in relationshi ps with a variety
86 of actors. On the other hand, in many cases they no longer had the safety net that the dyadic patron client relationships often provided, as each "new" relationship generally provided a more limited set of resources and services. Client s might no longer face the abuses of the old seringal owners, but at the same time they were now forced to expend more energy to develop and cultivate a multitude of relationships with several actors instead of one highly multiplex relationship with a sing le patron able to provide most of the resources they needed. These new relationships, in spite of being more equitable, often provided fewer resources per contact than previous arrangements. This may help explain why on the one hand, inhabitants decried the exploitation of the old system, and on the other they lamented the loss of the security that it often provided. The changes in economic activities, resource control and the arrival of alternatives to traditional patrons (like government agencies and social movements) that I outlined in Chapter 2 as important components of understanding patron client relationships were certainly important in the evolution of these ties in Lbrea. However, as I will show in the following chapters, patron client relatio nships (albeit modified ones) continued to persist in the area. One question that needs to be addressed is why do these relationships persist in some cases even when these dramatic changes have occurred? For one thing, dealing with some "new" actors, pa rticularly government agencies, may imply negotiating a complex, impersonal, bureaucratic structure with which inhabitants who are used to personalized face to face interactions may be unfamiliar or uncomfortable. In a similar case in Peru, the transition from face to face clientelistic arrangements to impersonal bureaucracies actually made it more difficult for isolated peasant communities to access resources and improve their situation in the short and
87 medium terms (Guasti, 1981) Engaging in familiar, albeit imbalanced, patron client inte ractions may still be more comfortable for some inhabitants than the official structures that are slowly replacing them (#$%&' Ayata, 1994) In addition, the inability of the "new" actors to reach all of the inhabitants who may need the resources and services they provide leaves opportunities for more "traditional" patron client relationships to thrive and survive. Even constitutional, democratic "modern" governments are unable to provide for everyone's needs all of the time (Land, 1983) which may lead those who are excluded from the resources and ser vices they provide to seek whatever other options may be available to them, including patron client relationships. Some of the "new" actors with the potential to replace the old patrons, like government agencies, often made demands that were difficult for inhabitants familiar with the old patron client relationships to satisfy. For example, in order to participate in their programs, government agencies in Brazil typically demanded official documentation, such as personal identification documents, that som e riverine inhabitants had not yet acquired. In short, just because there were "new" actors that could be alternatives to the old patrons did not mean that everyone was able to access the resources they offered. A lack of official documentation is just on e of a series of factors that continued to exclude many riverine inhabitants in Lbrea from engaging in relationships with the "new" actors that were replacing the old seringal owners as a source of key resources and services. In the next chapter I will show how a series of factors continued to exclude many of the inhabitants of riverine communities from engaging in relationships
88 with these "new" actors, which may explain the persistence of patron client relationships in spite of the dramatic socioeconomi c changes outlined above.
89 CHAPTER 4 CONDITIONS FOR PATRO N CLIENT RELATIONSHIPS ON THE PURUS RIVER Exclusion and Patron Client Ties: Aviamento on the Purus Today In spite of the social, economic, and political changes that have occurred in Lbrea over th e past fifty years, many of the inhabitants of the interior were still dogged by a series of conditions that compelled them to depend on patron client ties and the aviamento credit system. An understanding of the conditions that put people in positions of vulnerability may be one of the keys to patron client ties and how these relationships change over time (Weingrod, 1977) What needs or wants would inspire a person to f orm a relationship with a patron? I argue in the following pages that inhabitants of riverine villages faced a set of circumstances that, when combined, put them in a position of having to resort to patron client relationships to meet basic needs. The se asonality of production activities that riverine inhabitants of Lbrea engaged in required them to obtain some form of credit to get from one harvest to the next. In addition to problems of seasonality, inhabitants had to confront the tremendous physical distances between their homes and markets in urban centers, which greatly increased transportation costs. Isolation and a lack of infrastructure made crop storage and transport difficult and precarious. Uncertain harvest outcomes, poor infrastructure, an d low prices for the commodities they sold, coupled with the high prices of needed supplies that had to be shipped from distant ports through a chain of intermediaries, also made finding a source of credit necessary for many inhabitants. The patchy and di spersed nature of many of the forest products they depended upon often increased their production costs and augmented their need for credit. Inhabitants of these remote villages also faced a series of health problems
90 that often limited their ability to en gage in production activities and increase their expenses. Why would people engage in patron client relationships to access credit instead of going through more "formal" channels, such as taking out a bank loan? In this section I discuss a series of con ditions that tended to exclude many of the inhabitants from "formal" sources of credit, leaving them little option but to engage in patron client relations that enabled them to access "informal" sources of credit. These include: lack of education, lack of citizenship documents, and lack of collateral (land title). Production Systems, Seasonality, and Aviamento The people, patron and all, were from Cear, and now, as there was too much rain for rubber working, were busily engaged, some making canoes, othe rs handsawing planks under a shed near the beach, and still others cutting wood for the steamers (Steere, 1903) Chambers, et al. (1981) argued that for many inhabitants of rural areas in the tropics there is a direct correlation between seasonality and poverty. The seasonality of productive activities based on dry and wet climatic cycles can lead to periods of feast and famine throughout the year (Chambers et al., 1981) Seasonal climatic changes are also associated in many areas with outbreaks of illnesses that can greatly impact production activities (Chambers et al., 1981) These factors often place people in situations of extreme vulnerability and may accentuate the effects of poverty. Janet Townsend (1985) showed how the challenges of seasonal resource use, health problems, transport difficulties, and other factors inspired Amazonian peasants to borrow from river trade rs as a survival strategy in times of scarcity. In this section I explore how the seasonal nature of production activities in riverine communities along the Purus and Ituxi Rivers, combined with impacts of the illnesses
91 with which inhabitants contend, we re directly linked to their need to access some form of credit in order to obtain needed supplies. Although the inhabitants of these communities obtained much of their food from rivers, forests, and family agricultural plots, there were a series of items t hat could not be produced locally. These items included: fuel, salt, cooking oil, medicines, clothing, tools, cooking gear, and other items. People living in these remote areas relied on the natural resources at their disposal as a source of wealth that they could exchange for these vital supplies. By analyzing peoples' production activities we can see that the seasonal nature of their production activities implied that during some points of the year they may not have had a product in hand that they coul d directly exchange for supplies, necessitating some form of credit until they once again harvested crops or collected forest products. Production Calendars, Seasonality, and Credit Below I have included diagrams of production calendars from male and fema le inhabitants living in two of the communities inside the newly created Mdio Purus Extractive Reserve along the margins of the Purus River in Lbrea. These calendars are the result of interviews in which I asked people a series of questions about the pr oduction activities they engaged in during each month of the year. Analyzing these production systems is one point of departure for understanding why people engaged in aviamento and patron client relationships. Above the calendar months in Figures 4 1, 4 2, and 4 3 there are blocks that indicate the production activity that the person is engaged in during that month of the year. I have also included a line that represents the tremendous fluctuation in the water level (as much as twenty meters) of the Pur us River near Lbrea, which profoundly affected the production activities in which people c ould
92 engage. Below the calendar are a series of arrows that represent the constant need for mercadoria or supplies that people had throughout the year and whether o r not they were able to buy through direct exchange or on credit. Figure 4 1. Production Calendar, Male Respondent, Santa Rosa, Purus River In the production calendar in Figure 4 1 there were three time periods when the male inhabitant may not have had a product in hand that could be exchanged directly for supplies: February to March when subsistence fishing was the main activity; May to July when he was engaged in the laborious task of preparing the beaches for planting but did not yet have a crop to e xchange for supplies.
93 Figure 4 2. Production Calendar, Female Respondent, Santa Rosa, Purus River It is important to point out that women, in addition to household duties and child care, were usually just as involved in other production activities as the men. Figure 4 2 is an example of the norm for many women in the communities along the Purus River, who along with the children, contributed to virtually all of the household's production activities. In some cases the exception was the work that took pla ce in the centro areas that were far from the household, like Brazil nut gathering. However, I conducted several interviews with women who had tapped rubber and sorva, gathered Brazil nuts, and accompanied their husbands in virtually all of the household production activities. Some of these women even did these activities by themselves when their husbands were ill or when they had died. These interviews revealed situations similar to those described in Cristina Wolff's (1999) book "Mulheres da Floresta," which refutes many of
94 the claims of historians that downplay or ignore the presence of women in rubber production areas and also ignore their contributions to agricultural and extractive activiti es. In spite of the fact that the entire family was usually involved in virtually all of the production activities, the seasonal nature of these activities and uncertain outcomes of agriculture and extractivism made it difficult to combine production acti vities in a way that eliminated periods in which households did not have an agricultural or extractive product to trade directly for supplies. In some cases an exceptional harvest may have enabled a household to get through the months when they had no pro ducts to exchange for supplies. However, the reverse was also true; a poor harvest due to insects, river fluctuations, or other unforeseen problems could actually extend the period in which a household had to buy on credit. Changes in the price of inputs like fuel, could also have huge impacts on households trying to produce enough to get them through the year. With low prices for extractive and agricultural products, and high prices for supplies that had to be shipped from distance ports through a chain of intermediaries, in many cases even a good harvest was insufficient to get a family through the year without buying on credit. A low market price for the beans they produce, for example, may have meant that a household would need access to some form of credit to buy supplies even if they had had a relatively successful harvest in terms of production quantities. In Figure 4 3 we see a calendar in which the respondent was able to couple activities together in such a way that the period when he may not hav e had a product in hand to trade for supplies was greatly reduced. However, even in the best set of
95 circumstances when production activities could be combined in such a way that a household had products to exchange throughout the year, unforeseen circumst ances, such as a serious illness, could make access to some form of credit a necessity. This was also true if a household wanted to purchase expensive tools (chainsaws, outboard motors, etc.), household items (gas stoves, television sets, etc.), or costly building materials (roofing materials, wood planks, etc.). In these instances it was even less likely that a family would have the surplus on hand to buy these products outright without a establishing a line of credit. Figure 4 3. Production Calendar, Male Respondent, Vila Limeira, Purus River A comparison of these three calendars reveals production systems in which inhabitants attempted to couple together seasonal production activities throughout the
96 year in order to have products they could exchange f or supplies or even earn some kind of revenue. At the time this research was conducted, some families along the Purus were once again tapping rubber as a secondary or tertiary production activity due to federal and municipal subsidies that significantly r aised the price above market levels. However, rubber tapping was only feasible in the dry season, because during the rainy season the mid day rain showers would fill the rubber tappers' tigelas (collection cups) before they could collect the latex. The s easonal flooding in the extensive areas of vrzea on the Purus also meant that during six months of the year the tappers' rubber trails were underwater. Brazil nut harvesting, another important extractive activity, was also seasonal; the seedpods began fa lling in November and December, and collection then took place from January through May. The production of andiroba oil was also a seasonal endeavor; the trees dropped their seeds in March, collection took place in April, and the seeds were cooked and the oil extracted in the following months. Agricultural activities on the fertile beaches of the Purus, the main production activity in most riverine communities on this river, were also subject to the rise and fall of water levels. In May water levels typ ically began to drop rapidly, exposing beaches newly covered with an extremely fertile layer of sediment, which were subsequently weeded and prepared for planting. Manioc, corn, beans, watermelon, jerimum, and other crops were planted at this time. Perio dic weeding was required during the rest of the low water months as the crops matured. September and October were the months in which beans and other perishable crops were harvested off the beaches. Manioc was harvested last as the water levels began to rise again in November and December, when the beaches were once again covered with water. January and December were
97 the months when the manioc was soaked, grated, and toasted into manioc flour. Along the Purus beach agriculture became the most important production activity after the collapse in rubber prices in the 1980s and 1990s. In a survey of community characteristics and production data collected by Jos Maria Carneiro de Oliveira, President of the ATAMP Association, and himself an inhabitant of the Mdio Purus Extractive Reserve, 78 of 95 the communities where interviews were conducted listed beans as their number one productive activity. Manioc flour was generally the second most important, followed by other activities. Fishing in the Vrzea The thousands of oxbow lakes along the course of the Purus made it extremely rich in fish resources; however, fishing activities also had seasonal limitations. The lower water levels of the river in the dry season made fishing easier as fish were concentrated and easier to catch. However, spawning and migration patterns often determined the location of certain species at different times of the year. In some months individuals were spawning in the headwaters and in others they formed schools and feed in the m ain channel. I learned that in 1998 environmental laws were established in Lbrea to prevent the extirpation or extinction of important commercial species. These laws were in response to the rapid deterioration of the fishery around Lbrea due to increa sed pressure from commercial fishing operations from Manaus and Boca do Acre. With the collapse of extractive activities in the area in the early 1990s, fishing became an increasingly important activity for many riverine communities also, further increasi ng the pressure on this resource. When this research was conducted in 2008 and 2009, six important commercial species Matrinx ( Brycon amazonicus ), Pacu ( Mylossoma sp. ),
98 Sardinha ( Triportheus sp. ), Aruan ( Osteoglossum bicirrhosum ), Mapara ( Hypophthalmu s sp. ), and Curimat ( Prochilodus nigricans ) were protected during the defeso period (Nov.15 to March 15) when they spawn and produce eggs. There were also limitations placed on commercial fishing of Tambaqui ( Colossoma macropomum ) and Pirarucu ( Arapaim a gigas ). The species that were to be protected during the defeso were negotiated yearly by IBAMA and then communicated to the public. Members of the Colnia de Pescadores de Lbrea received a government stipend during this period to offset wages lost du e to the fishing restrictions. Non members were ineligible to receive the benefits. The defeso laws implied further seasonal limitations for fishing activities because they placed restrictions on key commercial species. Logging Season in the Vrzea The timber boom along the Purus from the 1970s to the mid 1990s also had seasonal characteristics. Large buoyant tree species in the vrzea like the Ceiba pentandra were the target of logging companies who used them to make plywood. These huge trees were c ut by men with chainsaws in difficult and dangerous working conditions right before the coming of the annual floods so they could be floated to the main waterways, bound together into jangadas or log rafts, and floated downriver to sawmills in Itacoatiara. Depending on the species, the trees could not be felled too far in advance of the flood season because insects, fungus, and rot could lower the value of the timber. Since there were no roads in the area, the loggers had to wait for the floods in order t o be able to float the logs out to the main river channel. A great deal of timber was lost when water levels did not reach high enough for logs to be floated out. The transport and sale of madeira em tora or unprocessed logs, was prohibited in the
99 late 1 990s, and the large jangadas that floated down the Purus ceased. People engaged in logging in 2008 2009 along the Purus and Ituxi Rivers often cut the wood into planks while in the forest and transported them in canoes to the small sawmills and furniture makers in Lbrea. Resource Distribution, Quantity, Quality, and Credit Rubber Variations in the production capacities of key resources were also important to understanding the inhabitants' need for credit. Although rubber tapping was no longer a primary a ctivity for most households along the Purus and Ituxi Rivers, they still engaged in it and it provides an interesting example of how resource distribution affects the need for credit. Rubber production quantities and qualities were far from homogenous thr oughout the Amazon basin and often varied greatly from one river valley to the next, within a municipality, or even within a seringal production unit. The rubber trees in the vrzea forests along the main stretch of the middle section of the Purus around Lbrea produced lower quantities of latex per season than the rubber trees on many of the tributaries of the Purus. Annual production for a rubber tapper on the Purus might average from 200 to 600 kilos of rubber while a tapper on the Ituxi or Pauini migh t produce as much as 800 to 1,200 kilos. The rubber trees on these tributaries were larger and produced much higher quantities of latex per day than those on the middle stretch of the Purus. Tappers who had worked on the Pauini informed me that they often tapped enough to earn a saldo or surplus at the end of the season. These same tappers complained about the poor production capacity of the rubber groves along the Purus, where they moved when the rubber industry on the Pauini collapsed in the 1980s and ea rly 1990s.
100 In addition to the production quantities of rubber groves in different locations, there was the question of latex quality. As many as twenty four different subspecies in the genus Hevea have been identified, and the taxonomy of the species ha s always been a challenge for botanists (La Rue, 1926) Hevea brasiliensis was considered to produce the best quality rubber, but there are several other species that also produced latex of secondary qua lity. Hevea benthamiana H. discolor H. rigidifolia H. foxii H. pauciflora H. minor and H. cuneata were all species that were considered producers of "fair to good" quality rubber (La Rue, 1926) T he rubber tappers and former patrons I interviewed for this research made a simple distinction between trees that produced "forte" (strong) and "fraca" (weak) rubber. The higher quality "forte" rubber was generally worth more than the "fraca." Testing th e elasticity of the rubber that had dried on a tigela (collecting cup) was the way they made this distinction: the "forte" would snap back with more elasticity when pulled than the "fraca." Tappers and patrons informed me that the rubber groves along the middle Purus had a mix of trees producing "fraca" and "forte" rubber, while the tributaries like the Ituxi and the Pauini had much higher concentrations of "forte." So in addition to the production quantities that varied greatly from one geographic locat ion to the next, the quality of the rubber, and hence the money that could be made from it, also varies. The differences in rubber tree production potential based on geographic location were noted in research done in the area in the 1920s: the lower stret ches of the Purus had very low rubber production capacity, while the further upstream one went, the better the production numbers and quality of the rubber (Schurz et al., 1925) The point is that tappers in areas with high production, like the Pauini
101 River, would often ear n enough during the tapping season to offset the months when they were not tapping, while tappers on the middle Purus often engaged in other activities, like Brazil nut collecting, to offset their lower production of rubber. Obtaining a source of credit f rom a patron was often a necessity to offset the limited production capacity of some rubber forests. Tapping over a period of years or cutting too deep into the bark could also lower the production of a rubber grove. In addition to the variation from riv er to river there were also differences in rubber production within micro regions. Some of the tappers had centros or colocaes that produced more than others on the same seringal During the early 1990s rubber tapping was generally abandoned along the Purus and Ituxi Rivers as a production activity, due to the collapse in prices when government subsidies were cancelled and protective tariffs on imported rubber were lifted. In recent years subsidies at the municipal and state levels inspired some inhabi tants to begin tapping rubber again, as evidenced in the production calendars I have presented here. Brazil Nuts Combining production activities throughout the year to offset low production from one seasonal activity had long been the norm for inhabitants in this region. Brazil nut collection was an important activity in the wet season for many inhabitants along the Purus when rain made rubber tapping unfeasible. However, Brazil nut harvests also faced problems of seasonality. Brazil nut trees along thi s section of the Purus and its tributaries typically started to drop their seedpods in November and December. It was extremely dangerous to be in the groves when the cannon ball sized pods were falling from as high as 100 feet to the ground, so the collec ting season typically did not begin until January or February when most of the seedpods had fallen. The Brazil nut harvest
102 typically came to an end in April or early May. Production quantities for Brazil nuts often varied tremendously from year to year. Some years the production was high and other years a grove might produce virtually no Brazil nuts. In recent years lower than average precipitation and caterpillar infestations had also lowered the production of many groves. This variation in production could also determine whether or not a household will need to buy on credit or seek yet another alternative production activity. The geographic distribution of Brazil nut groves presented another problem related to the need for credit. The Bertolletia e xcelsa tree is an upland species that did not grow in the extensive vrzea forests along the Purus around Lbrea. It was found on the few patches of upland near the banks of the Purus but generally the Brazil nut groves were found on the tributaries, like the Ituxi River, with higher ratios of terra firme upland to vrzea and in the forests that are some distance from the banks of the Purus. The distance from the banks of the river, where most households were located, meant that during the Brazil nut har vest gatherers had to travel to the groves and set up temporary living arrangements until the end of the harvest. In addition to the fuel needed to reach the groves, gatherers needed other supplies to get through the harvest (food, ammunition, cooking sup plies, etc.). These items were usually bought on credit from a patron and then paid back at the end of the harvest with the Brazil nuts they had collected. In some cases the price of Brazil nuts fluctuated during the season to the point where it was diff icult to pay back the loan for supplies. The uncertain land tenure of the Brazil nut groves was also related to the problem of credit and debt. When gatherers were collecting the nuts that had fallen they could already see the new seedpods on the tree, w hich gave them an idea of what the harvest
103 production for the following year would be. Even so, I spoke to several families that went ahead with a Brazil nut harvest even though they knew that it was not going to be a good year and that they would probabl y end up not making any money (or even be in debt) once they paid back the patron who sold them supplies on credit. When I asked why they would borrow the supplies and do the harvest knowing that the trees were not going to produce much they answered that their "ownership" of the grove was based on use. If they did not collect Brazil nuts in their grove for a year or two another family could take it over. They decided to buy the supplies on credit and go into debt because they did not want to lose their "rights" to their Brazil nut grove. In other cases people from families that once lived in the interior but had moved to Lbrea had managed to maintain their "ownership" of certain groves that they no longer worked directly. Other needy families then ar ranged with them to pay "rent" in order to collect Brazil nuts in their grove. Having to pay rent to another family on top of buying supplies on credit often made it even harder for a family to pay back loans and make a profit. In addition, when this res earch was conducted it was getting more and more difficult to obtain supplies on credit to finance the production of Brazil nuts. Banks were not interested in loaning money for these activities and the storeowners were becoming more and more leery of loan ing supplies and money that gatherers would be hard pressed to pay back. Andiroba and Copaiba Oils The uncertain nature of the markets for many forest products also made surviving between seasonal activities a challenge. The oil derived from the seeds o f the andiroba tree ( Carapa guianensis ) and the sap that was tapped from the trunk of the copaiba ( Copaifera sp. ) tree were known to have medicinal properties and were also used in the
104 production of cosmetics. Inhabitants of the Purus had a long history o f using andiroba oil to make soap and it had become a commodity sought after by companies from southern Brazil that manufactured cosmetics. Copaiba oil was also used for cosmetics. The problem with these products was that prices could fluctuate wildly fr om year to year and even during a production season. A household might find that this activity provided an income one year and had almost no value the next, forcing them to find another activity to replace it, or once again buy on credit. Variation in R esource Access by Household The distribution of resources by household also played an important role in determining which production activities households engaged in and how they combined activities throughout the year. Some households had extensive beac h areas for agriculture, while others did not. Some households had Brazil nut groves, while others did not. Some households had productive lakes for fishing, while others did not. The need to string together a series of production activities to get thro ugh the year helps us to understand the conflicts that ensued when resources, like Brazil nut groves, became contested. It also helps explain why extractive reserves were so appealing for communities who were desperately trying to consolidate their land t enure situation and felt like they had few alternatives to do so. Floods of Fortune and Famine The often fickle nature of the flood cycles of the Purus could greatly impact a household's ability to successfully string together production activities and av oid having to purchase on credit. River levels had to drop soon enough to allow crops to be planted and reach maturity before the next flood cycle. A flood cycle that came too soon could wash a crop away before it had been harvested. For households that
105 depended to a great extent on their bean crop as a source of wealth to buy needed supplies, a reduced low water season might mean that their crops would not have time to mature before the floods come again, putting them in an even more precarious financial situation. In 2009 the rubber tapping season was cut short on the Purus because the water levels dropped much later than usual. When I went out to the communities to tap rubber in July 2009, many tappers, who often began tapping in late May and early J une, were just beginning to tap because their trails were just then becoming dry enough to walk. This meant that their yearly yields were greatly reduced due to the shortened tapping season caused by the unusual flood cycle. The flood cycles of the river c ould quickly wipe out crops or delay extractive activities in ways that forced people to search for credit to buy needed supplies. Flooded Forests, Livestock, and Economic Stability One of the strategies increasing employed in other parts of the Amazon i s the acquisition of livestock as a buffer to offset these seasonal variations. Livestock can be walked to markets, sold at any time during the year, and there was generally a constant demand for beef and pork products. However, the flood cycles in the v rzea forests along the middle Purus placed limitations on inhabitants' ability to engage in large scale animal husbandry activities. Of the 90+ villages on the Purus between Lbrea and Pauini only five or so had terra firme uplands close by; the rest wer e inundated for up to six months per year. This meant that during the flood season it would be very difficult to find a place for a large herd of cattle. In other parts of the Amazon riverine communities built marombas (floating corrals) for their cattl e to live on during the flood season and cut grasses for them to eat
106 during this period (N. J. H. Smith, 1999) I saw no marombas on the Purus or any of the tributaries during nearly two years of field research. Even if communities had built them they would be limited in the number o f cows they could put on these rafts during the flood season. Most of the terra firme uplands near the middle section of the Purus between Lbrea and Pauini were incorporated into the extensive Indigenous Lands that have been created in the area, and eve n if riverine communities wanted to establish pastures in these areas, which were often quite some distance from the banks of the river where they lived, they no longer had access to many of these areas. It is important to note that the few communities on the Purus that still had access to terra firme uplands were raising cattle and some hoped to expand their herds and pasture areas. Even some of the vrzea communities grazed a limited number of cows on the exposed beaches during the low water season. Th ese small herds (generally less than 10 head) could cause fierce conflicts within and between villages when they strayed into agricultural plots and destroyed crops. Pigs were common in some of the riverine communities along the Purus, but they were gene rally left to run free and were not corralled or fenced. In one community the pigs had become feral and formed herds that occupied nearby forests. Other communities had no pigs and cited the odor and unclean nature of the pigs as a reason they did not ra ise them. A dependence on beach agriculture may have also been a limiting factor that prevents many communities from increasing their animal herds. Both pigs and cattle can rapidly destroy a plot of beach agriculture and many families may have elected to limit their livestock to avoid this problem.
107 In spite of the extensive terra firme uplands along the Ituxi River I found only one household that had livestock of any kind in the riverine communities I visited there. People explained that they did not h ave the capital to invest in fences for pasture areas for cattle and they had constantly seen their pigs killed by jaguars, alligators and other wild animals. Some people said they did not have the costume (custom) of raising livestock. This included kno wledge regarding: vaccinations, pasture management, and other information necessary to successfully raise livestock in tropical areas where they are plagued by insects, diseases, and poor foraging conditions. The fact that the Ituxi was even more remote a nd isolated than the Purus may also explain this trend. Small canoes were often the only mode of transportation for many households, which also would make transport of livestock more difficult. The contrast between the radical change to a livestock econom y in much of the nearby states of Acre and Rondnia and the continued dependence on faltering extractive activities in municipalities like Lbrea has been noted by other researchers (F. S. D. d. Santos & Muaze, 2002) The extensive vrzea areas and lack of road networks may help explain this, as well as the incentives for agriculture and livestock raising offered at different times by state and municipal governments in Acre and Rondnia that may not have been offered in Amazonas state where Lbrea is located. As some respondents mentioned, the lack of a costume of raising cattle may also have played a role. In summary, patchy resource distribution, varying quantity and quality of production based on location, the seasonal nature of agriculture and extractive activities, and the impacts of sea sonal flooding, combined with uncertain markets,
108 limited the ability of households to purchase costly supplies without accessing some form of credit. Health, Vulnerability, and Credit In respect of ague [malaria] the Ituxy has the worst name of all the a ffluents: the Pamans, a tribe on it, are said to be always suffering, and consequently very indolent and unwilling to work, whatever price offered (Chandless, 1866) Notvel a tradio mrbida de Lbrea, considerada uma das cidades mais doentias do Purus (Cruz, 1913) ." (Notable is the tradition of morbidity in Lbrea, considered one of the most unhealthy cities of the Purus.) In fact, certain localities in the Purus and the Jurua, as in other rivers, have a reputation for salubrity that is borne out by actual observation. However, some places are notorious centers of infection. One such is the village of Labrea, the principal town on the Purus and formerly a center of considerable importance, whose entire population in 1922 was sai d by an inspector of the Public Health Service to be afflicted with malaria (Schurz et al., 1925) For many the town of Lbrea, and much of the Purus River, became synonymous with disease and infection, a reputation that dates back to observations by early explorers and researchers. During the rubber boom the Ituxi River became known as the cemitrio da humanidade (cemetery of humanity) due to the number of fatalities from malaria epidemics (Bittencourt, 1918) A deadly strain of Hepatitis D first identified in the area became known as the Febre Negra de Lbrea (Black Fever of Lbrea) and the area was still associated with this virulent strain of Hepatitis. Health concerns, which can also be seasonal, are directly related to production activities and could also determine the need for credit for many inhabitants of the Purus and its tributaries. Hy giene Hygiene in the communities along the Purus and its tributaries was generally very poor. Inhabitants did not have access to treated water, and human waste was disposed
109 of in pit latrines, which due to the high water table and seasonal flooding consta ntly threatened to contaminate the drinking water supply. Gastrointestinal complaints such as diarrhea and various kinds of stomach parasites were common. Malaria In addition to the maladies associated with lack of hygiene and water treatment, inhabitan ts faced a series of illnesses that were transmitted by the abundant swarms of insects that plagued the banks of the Purus. The sediment rich waters of the Purus and the thousands of oxbow lakes that formed as the river changed course were famous for the quantity of insects they produced at different phases of the flood cycle. In the report of his ascent of the Purus Chandless wrote that the unbelievable quantity of mosquitoes, blackflies, midges, and other biting insects prevented early settlers from est ablishing year long residences along the river (Chandless, 1866) Malaria, transmitted by various species of Anopheles mosquitoes, had been common on the Purus and its tributaries since at least the 1860s when Chandless noted its presence in his report (Chandless, 1866) It is not known exactly how or when mala ria arrived in the Amazon, but it appears that it was prevalent in the region by the 17 th century (Stepan, 2003) There is some speculation that it was brought during early European colonization or arrived as a result of the African slave trade (Stepan, 2003) Malaria was considered to be one of the principal obstacles to developing agriculture, extractivism, or industry of any kind in the Amazon from early on in the colonial period. Th e extremely high toll that malaria took on the rubber workers in the 19 th and 20 th centuries and on the workers of infrastructure projects like the Madeira Mamor Railroad led the Brazilian government to sponsor a series of public health studies from 1905 to the 1920s led by the famous Brazilian physicians Oswaldo Cruz
110 and Carlos Chagas. They traveled thousands of miles on the Amazon and its tributaries collecting data and trying to determine how malaria infection could be controlled. In Lbrea they found extremely high incidents of malaria and observed many patients with enlarged spleens, a symptom of chronic infection (Cruz, 1913) They noted the poor sanitary conditions in the area and the almost complete lack of access to medicines and health care (Cruz, 1913) With the lack of tax revenues and general economic collapse associated with the rubber bust in 1913, the Brazilian government was unable (or unwilli ng) to invest in the malaria control and eradication programs suggested by Cruz and Chagas (Stepan, 2003) Malaria was still a considerable problem for the inhabitants of riverine communities in Lbrea in 2008 and 2009. The vast majority of the villages on the Purus and its tributaries did not have even the most basic health facilities. FUNASA, the government agency in charge of indigenous health, and the Health Secretary of the Municipality of Lbrea formed a partnershi p to try to control malaria epidemics in the area. A boat staffed with municipal employees trained to use microscopes to identify malaria and provide treatment was periodically sent to riverine communities. However, the costs of transportation to access remote communities were high and the visits were often limited to the peak seasons of malaria transmission or when the municipality had been notified of epidemics. The lack of health facilities in the villages and the sporadic nature of visits by municipa l health providers meant that inhabitants had to travel to Lbrea in cases of health emergencies. In some cases the municipality sent small boats with outboard motors to retrieve the sick, but this service was often unavailable. This meant that inhabitan ts had to find a way to make the trip, which often took several days in a canoe
111 with a rabeta motor, and then pay for food in Lbrea while they obtained treatment. Generally the health posts and the hospital, which were run as part of Brazil's national he alth program, provided some medicines and treatment free of charge. However, in some cases the inhabitants had to purchase their own medicine. During interviews, residents of the Ituxi listed the high incidence of malaria, lack of health care, and the co sts of traveling to Lbrea to obtain treatment were cited as major concerns. Inhabitants also said that several women had died during childbirth due to the total lack of health facilities. Women in the later stages of a pregnancy often traveled to town t o give birth, which also implied expenditures that were often difficult for inhabitants to pay. Filariasis Filariasis was another malady transmitted by insects that was common in many communities along the Purus and its tributaries. Insects of the Ceratop ogonidae (midges) and Simuliidae (blackflies) families transmit the nematode Mansonella ozzardi the most common variety of this larger family of blood parasites on the Purus and its tributaries (Medeiros, Py Daniel, Barbosa, & Izzo, 2009; Shelley & Coscarn, 2001 ) This particular filarial parasite can cause headaches, fever, coldness in the legs, aching joints, and inflammation of the lymph nodes and liver. A study conducted in riverine communities along the Purus and Pauini Rivers between 2004 and 2006 reveal ed a high incidence of infection (Medeiros et al., 2009) Infected inhabitants were found in all thirty five of the communities sampled with a 24.86% infection rate in the sample of inhabitants from the Pauini Rive r and a 24.6% infection rate in the inhabitants sampled on the Purus River (Medeiros et al., 2009) Infection rates were higher for inhabitants that engaged in farming and extractive activities than for those engaged in domestic activities due to their increased exposure to blackflies (Medeiros et al., 2009) In
112 several of the villages I visited on the Purus every single inhabitant had tested positive for filariasis the last tim e the health boat had passed to provide medical care. A similar study conducted on the Ituxi River revealed that 30.23% of a sample of 129 individuals from twelve communities were infected with the parasite (Medeiros, Py Daniel, Barbosa, & Ogawa, 2008) Hepatitis While I was receiving treatment for malaria in Manaus in 2008 the first thing that physicians brought up when I told them where I was working was the Febre Negra de Lbrea Several books about the municipality of Lbrea refer to this mysterious strain of Hepatitis that was known to wipe out entire families in short periods of time (Ferrarini, 1981; Luxardo, 1973) For some victims the final stages of the disease included internal hemorrhaging and vomiting of a black liqu id, which is how the disease came to be christened febre negra ." I was told during interviews that the mysterious and deadly nature of the disease often led survivors or neighbors to burn down the houses that it had struck and abandon the area. Research in the 1980s revealed that the Febre Negra de Lbrea was a variant of Hepatitis D. Hepatitis continued to be a serious problem in Lbrea, with the Purus, Juru, and middle Solimes Rivers being areas of high endemism of the disease (Braga et al., 2004) Hepatitis A, B, C, and D are all prevalent in and around Lbrea, with Hepatitis D being the most feared because of its lethality. According to one study the rates of infection had dropped since 1989 when the first vaccination campaigns for Hepatitis B were conducted in Lbrea (Braga et al., 2004) After 1989 the numbers dropped considerably, but people in rural areas were still much more likely to acquire the disease than those living in urban areas, probably due to the sporadic nature of the vaccination campaigns and the lack of medical
113 facilities. The Hepatitis B vaccine is often administered in a series and requires refrigerated storage. If the series is not administered according to the proper cycle (3 injections, each a month apart) the vaccine may not provide protection. Many inhabitants of the villages in the interior may not have had complete vaccinations or the vaccinations may not have been properl y stored or administered. In addition to malaria, filariasis, and hepatitis, inhabitants of rural villages faced infection with leishmaniasis, a disease transmitted by black flies. Chagas disease was another ailment spread by insects that can cause hear t diseases. Lbrea also had a reputation as an area with a large population of people infected with leprosy. It was not uncommon to see people in the interior who had become unable to work due to the debilitating effects of the disease, which can be cure d if treated in the early stages. A diet based on manioc flour and fish that was often lacking in vegetables, lacked key vitamins, and often left inhabitants more susceptible to these ailments. Health and Credit The health challenges faced by inhabitant s of the villages of the Purus and its tributaries could have large impacts on their need for credit. First, the insect infestations that plagued the area often coincided with periods of the year when inhabitants were engaged in important production activ ities. Malaria transmitting mosquitoes often reached their greatest abundance during the periods of the flood cycle when river levels were rising ( enchente ) and falling ( vazante ), exactly the time when inhabitants were planting and harvesting their agricu ltural crops. Filaria transmitting blackflies were most common during the high water seasons of the flood cycle. These illnesses could lead to missed days of work or a reduced labor capacity that could lower the final crop yields. Hepatitis could also g reatly reduce an inhabitant's capacity to do the hard labor
114 associated with non mechanized agriculture and forest product collection. When these illnesses were combined with intestinal parasites, diarrhea, and an incomplete diet, the ability to produce su fficient crops and forest products could be further reduced. In addition to the reduced labor capacity associated with these illnesses that could lower production quantities, the need to travel to the city for health care could lead to expenditures for f uel and food, as well as lost work days due to travel. Sometimes medicines also had to be purchased. All of these could lead to expenses that required the acquisition of credit. I conducted surveys in communities on the Purus and on the Ituxi in 2008 an d 2009. In one question I asked respondents to rate seven services (Health, Transportation, Education, Technical Assistance Agriculture, Technical Assistance with Extractivism, Sanitation, and Security) by degree of urgency. 75% of the respondents answer ed that health care was the most urgently needed service in their community. The point here is not to paint the area as a disease infested inferno, but to point out the real challenges that the almost total lack of health services presented for the inhabi tants and how it impacted their need for credit and their ability to pay off debts. Concerns about the lack of health care were mentioned in almost all of the interviews I conducted with inhabitants of the riverine communities on the Purus and Ituxi River s. In the past twenty five years health facilities in the city of Lbrea had certainly improved In 1970 health facilities consisted of one SESP health post and a one physician. In the rubber production areas the government carried out sporadic ma laria control and patrons often had at least basic medicine available for sale in their barraces In 2008 2009 the city of Lbrea had a fifty bed hospital and five health posts
115 distributed throughout the city. The municipality invested in a health barge that periodically made trips into the interior to provide basic health care. In spite of the improvements in health care in the city of Lbrea, the inhabitants of the riverine communities in the interior in most cases lacked even the most basic health f acilities. Since 1999 programs implemented to train agentes de saude (health agents) had given some inhabitants the ability to do tests for malaria and identify other maladies. However, the lack of any medicines or health facilities meant that for the mo st part agentes were limited to reporting on the conditions in their communities. The lack of effective means of communication was also a great concern in many communities. Not only were health facilities unavailable, but in many communities there was no effective form of communication in case of health emergencies. A network of CB radios existed along the Purus and to a lesser extent on the Ituxi, but most of the radios were inoperable much of the time. The lack of health facilities in the interior, in addition to economic difficulties, were one of the deciding factors for many families in their decision to move from rural areas into the cities. Education, Literacy, and Patronage After health care, education was the service that respondents felt was mo st urgently needed in their communities. Improvements had been made to the education system in rural areas in the last five years before I conducted the study and the municipal, state, and federal governments had made improvements in the school system in rural areas of Lbrea. By 2008 2009 when this study was conducted the municipality had twenty schools in rural communities with seventy professors. The requirements for professors in rural areas had also risen and professors were required to have univers ity education; previously many rural schoolteachers had only a primary
116 school education. After 2000 a program to build rural schools was implemented in which twenty schools were built. By the time I conducted this research most of the children in rural a reas at the very least had the opportunity to attend primary school. During interviews respondents did complain to some degree about the quality of the education and the fact that after completing primary school their children would have to go to the city to continue their education. There is no doubt that inhabitants put an extremely high value on education and they brought it up constantly as a service that they wanted their children to have. Education certainly played a role in the changes in patrona ge that have occurred in Lbrea over the years. One of the principal reasons that people were forced to find a benefactor or patro was to obtain help with government paperwork, explanations about how certain procedures are done, etc. The inability to re ad and write made understanding written contracts or other written agreements extremely difficult. This included bank loans, obtaining citizenship documents, obtaining government retirement and social welfare benefits, etc. Respondents often said that th ey would have to go to a patron or someone else they could trust that would explain these things to them. Increasingly community leaders were turning to social movement organizations like the National Council of Rubber Tappers and the Pastoral Land Commis sion, or state government agencies like IDAM, for explanations and information regarding legal procedures, such as the process of registering their community associations. Exclusion from "Formal" Credit and Dependence on Aviamento At this point it is inst ructive to analyze the opportunities that many families have for accessing credit in order to understand their continued dependence on aviamento or some other form of "informal" credit arrangement. I am defining "formal" arrangements
117 as those based on wr itten contracts and "informal" arrangements to be those based on verbal agreements. Geographic Distance The distance that separated households from credit providers was one of the first things that stood out when traveling to communities in the interior of Lbrea. In an area with no roads, the meandering course of the Purus and its tributaries were the only access route between the town of Lbrea and households in the interior. A journey of several days often separated communities from the nearest bank, grocery store, or medical facility. The average distance from the town of Lbrea and the communities where I conducted research was more than 200 kilometers 1 by river. Communities consisted of small groupings of houses on the banks of the river with the only community infrastructure being a one room schoolhouse, an empty health post, a poorly functioning CB radio, and electric generator. Many communities lacked even this basic infrastructure. Aside from river traders, people in remote communities had f ew if any alternatives for purchasing needed supplies, unless they had the resources to make the long journey to the city, where they had to find a place to stay and food to eat while they were there. In some cases they simply found it easier to sell thei r agriculture and extractive products to a passing river trader in exchange for needed supplies. They were aware that they were probably selling for their products for less and paying more for the supplies, but many recognized that this enabled them to av oid the cost of traveling to the city themselves. The reverse was also true; most of the inhabitants of communities that were relatively close to Lbrea preferred not to buy from the river 1 Please see Appendix Table A 9 for table showing distances from Lbrea to the research communities.
118 traders and did most of their shopping in town, where prices were lower. The proliferation of 5, 10, and 15 horsepower rabeta motors over the past fifteen years had greatly improved the access of many families in the interior to markets in the city. A journey that once took a week or two when paddling could be cut to s everal days with one of these motors. However, fuel was still a limiting factor for many families hoping to make the journey to markets in the city. For this reason the sheer immensity of the physical distances between rural communities and the city coul d impact the options that inhabitants had to access markets and credit. Even if a family was able to get to the city they may have been excluded from formal lines of credit. Banks typically required some kind of collateral when they granted a loan. For families in the interior with no land title and no other property considered valuable enough to mortgage, this avenue was usually closed. In addition, extending credit for a family to engage in basic production activities may have been considered too ris ky or not interesting enough financially for a bank looking to invest its money and earn interest payments. High interest rates and a complex bureaucracy were also impediments for families considering a bank loan. In many cases families in the interior w ere also hampered by their lack of citizenship documents, which were also necessary for obtaining loans, opening bank accounts, and accessing government assistance programs. Citizenship Documents The following is an example of how the lack of proper docu mentation excluded many inhabitants from government programs and formal lines of credit. I learned that in October and November 2008 a team from INCRA traveled to all of the communities on the Ituxi and Purus Rivers that were within the limits of the newl y created Rio Ituxi and
119 Mdio Purus Extractive Reserves to register the inhabitants so they could receive government benefits. When a federal extractive reserve was created the inhabitants became eligible for many of the benefits that inhabitants of the a ssentamentos or government land reform colonies were eligible to receive. The Instituto Chico Mendes was in charge of administering the extractive reserves, but INCRA was in charge of administering these benefits. Registered households became eligible to receive credits for housing, food, and agricultural implements. The INCRA team spent twenty nine days registering families in the two areas. In order to become legally registered, inhabitants had to show one identification document, which can be an iden tity card ( carteira de identidade ), a work card ( carteira de trabalho ), or a military identification ( carteira de reservista ). If they were married they were required to show their marriage license. Inhabitants had to be 18 in order to register and the r egistration was usually done at the household level. So if the male head of household did not have all of the documents but the female head of household did, they would be registered in her name. During their trip the INCRA team registered 91 families on the Ituxi and 890 families on the Purus. There were 100 families on the Ituxi and approximately 200 on the Purus that were unable to register for the credits because they did not have the proper documentation. This meant that a significant number of the inhabitants of these areas were excluded from accessing formal lines of credit and assistance and had to turn to informal sources, often based on patron client relationships, if they needed credit and assistance. A study done among business owners in Lbr ea in the 1990s revealed that even for the inhabitants of the city that operated businesses and had property and capital
120 reserves it could be extremely difficult to obtain loans and credit from banks ( Diagnstico Scio econmico e Cadastro Empresarial de Lbrea 1996) In a survey including six different industries in the municipality of Lbrea (clothing, furniture, foodstuffs, meat packing, and locksmiths) 88% depended on their own capital to run their businesses, while only 12% had access to bank loans ( Diagnstico Scio econmico e Cadastro Empresarial de Lbrea 1 996) 63% responded that they faced difficulty in accessing credit due to their inability to provide sufficient collateral, high interest rates, and excessive bureaucracy ( Diagnstico Scio econmico e Cadastro Empresarial de Lbrea 1996) If business owners in the urban center faced this level of difficulty in accessing formal lines of credit, we can only surmise the difficulties that a rural inhabitant wit h no collateral, no citizenship documents, and no education would face in negotiating the bureaucracy and convincing a bank to loan them money. Chapter Conclusions As described in the previous paragraphs, many riverine households in Lbrea were walking a tightrope from one seasonal production activity to another as they attempted to produce goods that they could exchange for the supplies they needed throughout the year. The patchy distribution of many of the resources and the unstable nature of many of t he markets of these products could further increase the vulnerability of riverine inhabitants and their need for credit. In addition to insects, damage by birds and other animals, and other natural elements that could damage crops, the uncertain nature of the annual flood cycle could wipe out crops or prevent them from being planted in time. Lack of infrastructure for transportation and crop storage could lead to losses even when a harvest is good.
121 Riverine inhabitants in the interior of Lbrea were bes et by a number of illnesses that could greatly lower their capacity to engage in production activities. There were virtually no health facilities in the communities along the Purus, which implied that health care had to be obtained in town, implying trave l costs and lost days of work. All of these elements implied that riverine inhabitants would invariably face times of the year in which they would have no products in hand to exchange for much needed supplies. The fact that they were often excluded from formal sources of credit meant that they had to find some other form of safety net should they have a bad harvest, an unforeseen illness, or the need to make an expensive purchase. In many cases they had to seek out relationships with someone who would h elp them through these difficult periods. That someone was often in a position of great leverage in this relationship due to the vulnerable position of the inhabitant that results in part from the complicated nature of their seasonal production activities and the fact that they were excluded from many other forms of credit.
122 CHAPTER 5 SOCIAL CHANGE, COMMU NITY ORGANIZATION, A ND "NEW" SOCIAL NETWORKS: FROM TWO V ILLAGES TO TWO EXTRA CTIVE RESERVES, NETWORKING AND ORGAN IZING OUT OF CRISIS Just as the river systems that they lived on dictated many of their production activities, a series of social and economic currents also had profound effects on the lives of the inhabitants of riverine communities in Lbrea. These social and economic currents, much like th e flows of the Purus River, were often cyclical; flowing forcefully at times, threatening to envelope everything in their path, at other times ebbing to a point in which they were barely perceptible. Just as the meandering Purus River periodically changed its course, leaving behind oxbow lakes and abandoned channels that eventually dry up and are reclaimed by the forest, the social and economic currents that impacted the lives of the inhabitants of riverine communities also ebbed and flowed before coming t o the end, only to be replaced by new social and economic currents. The rubber industry, which was so important to the formation of the society that developed in Lbrea, was itself a finite current, with strong ebbs and flows that shaped the lives of tho se engaged in it for time and was then replaced by other economic activities. Social currents are also finite, ebbing and flowing for a time before new social currents replace them. For example, the social current that led to the formation of the Rural W orkers' Union of Lbrea flowed forcefully for several years, causing significant change in the lives of those that were swept up in it and then saw its power ebb as it was replaced by a new set of social currents in the form of new governmental and social movement organizations. The strength or weakness of these social currents did not imply that they impacted people in the same way or to the same degree. In some cases individuals and
123 communities were hardly impacted by certain social currents, which pas sed by with little or no impact. Other currents appeared unavoidable and individuals found themselves in a position in which they had to confront them or be swept away. In some cases individuals and communities actively sought out a social or economic cu rrent that they felt would benefit them or help protect them from those that threatened to sweep them away. The inhabitants of these communities were not passive actors, helpless in the face of dramatic social and economic change, and often sought out and became incorporated into these broader social and economic currents as a means of adapting to their changing reality. The localized problems and challenges that inhabitants of riverine communities in Lbrea faced were often caused and shaped by larger soc ial forces, and these localized problems and challenges also inspired individuals in these communities to incorporate themselves into larger social forces in order to resolve their localized problems. Typically this implied forming relationships and socia l networks with individuals who were involved in these social currents and could help them fend off the forces that threatened their existence. In this chapter I describe the social and economic events that shaped the recent history of three communities in the interior of Lbrea: Vila Limeira, Floresta, and Vila Vitria. First, I relate the recent history of these three communities as they adapted to the social and economic changes described in Chapter 2. Next I analyze the broader social and economic currents that inhabitants of the community perceived as threats to their wellbeing and way of life. Then I describe the social and economic currents that they perceived as helpful in resolving their problems and useful to thwart the negative currents that they perceived as threats. I describe how leaders from these three
124 villages used social networks to incorporate themselves into these positive social currents in order to achieve their goal of creating two extractive reserves, which they viewed as a solu tion to many of their problems including: land tenure uncertainty, economic crises, and lack of basic services. The history of these three communities and the actions of their leaders illustrate the changes in patron client ties that have occurred in the area, particularly the shift from dyadic, multiplex patron client ties to networks in which a multitude of brokers and intermediaries are sought out to obtain resources and services. These case studies not only reveal how patronage changes, but also provi de examples of how people in remote areas actively form relationships and build social capital to resolve problems and improve their living conditions. I must point out that this story was developed from extensive timeline interviews and daily interactio ns with individuals from these communities. Although I did interview patrons, river traders, and a host of other individuals with different perspectives and points of view, my understanding of the history of these communities, the threats that they faced, and their manner of responding to these threats is the product of my interactions with them. Events that individuals in these riverine communities considered to be threats were seen as beneficial by some other groups. For example, the creation of the In digenous Lands along the Purus River that threatened to push riverine peoples from their communities was seen as tremendously beneficial from the point of view of the long suffering indigenous peoples who stood to achieve some land tenure and resource secu rity of their own. The deforestation, land speculation, and conversion of forests to pasture that represented such a threat to the livelihoods of communities on
125 the Ituxi River were seen as beneficial "development" and "progress" in the minds of the ranch ers and business people who stood to benefit from these activities. Community Level Currents Between the city of Lbrea and the city of Pauini there were more than 90 riverine villages when this research was conducted in 2008 2009. Vila Limeira was a co mmunity located at the far western end of the municipality of Lbrea, abutting the neighboring municipality of Pauini. It is 507 kilometers from the town of Lbrea by river, and depending on the mode of transportation could be several days distant. Vila Limeira consisted of 19 households, most with direct family ties to one another, located several kilometers off the main course of the Purus on a tributary known as the Seruini River. Vila Limeira The inhabitants of Vila Limeira traced their arrival in t he area to 1942 when Napoleo Oliveira, a migrant from Pombal, Paraba, arrived at the nearby Seringal Ermida to work as a rubber tapper. Napoleo Oliveira came as part of the government migration program that was part of President Getulio Vargas governmen t's "Marcha para o Oeste" or "March to the West" campaign. According to one source, by 1940 the Departamento Nacional de Imigrao (National Immigration Department) had already sent 8,000 Northeasters to the Upper Amazon and Acre as part of this migration program, which was designed to populate "unoccupied" areas in the interior and reverse the rural urban migration that was seen as a threat to the large coastal cities (Secreto, 2007) This program predated the Washington Accords and the Soldado da Borracha program by two years, although it was later incorporated into the Soldado da Borracha program (Secreto, 2007) During interviews, the Oliveira family made it clear that
126 Napoleo had not arrived as a soldado da borracha but instead as a migrant participating in this government program. When Napoleo arrived in the Amazon, the patronage pyramids that had become entrenched on the seringal rubber production system still existed. As we can see in Figure 5 1 1 below, supplies came in to the seringal through the owner's contacts with merchant houses, who in turn often had ties to lar ge banks. This is an example of a patronage pyramid, in which a patron uses his influence to maintain control of several clients (Martin, 2009) Figure 5 1 Seringal Patronage Triangle Seringal owner patrons were able to keep their tapper clients by providing supplies on credit on an exclusive basi s. In the event that the tapper sold rubber to someone else, such as a river trader, the patron threaten to invoke sanctions, such as restricting credit or evicting the tapper and his family from the seringal Priests, government 1 This figure was inspired by the patronage py ramid figures in (Martin, 2009)
127 agents, politicians, and others respected the seringal hierarchy and would seek out the patro upon arriving at the seringal In the 1940s the merchant houses in Figure 5 1 were increasingly replaced by government organizations that began to control elements of the rubber indust ry as part of the Batalha da Borracha program (Dean, 1987; Martinello, 2004; Secreto, 2007) After working for a time at the Seringal Ermida, Napoleo decided to try his luck on the rubber rich Pauini River, a tri butary of the Purus, where he met his wife and began a family. Several years later Napoleo returned to the Purus, moving his family to the Seringal Meteripu. Napoleo's movements from seringal to seringal and even from river to river show that despite their tendency to be bound by debt, there appears to have been a good deal of mobility among rubber tappers, and many were constantly moving in search of better tapping areas, more benevolent patrons, better areas for hunting and fishing, etc. (Cabral, 1984 (1949); Pantoja, 2004) By this time few of the seringais still had the capangas that had terrorized some tappe rs and enforced seringal rules with corporal punishment during the Rubber Boom. The wave of migration from the Northeast in the 1940s included men with more education and more awareness about their rights as citizens than the migrations of the late 19 th c entury (Secreto, 2007) Many of the tappers who came as soldados da borracha were from urban areas, and the patron client institutions that had also dominated the rural back lands of the Northeast were less familiar (and perhaps less acceptable) to them than to earlier waves of immigrants (Secreto, 2007) This wave of immigrants included men who were more likely to stand up for their rights and not accept the harsher treatment that had characterized the seringais in the boom period.
128 In addition, as part of the Batalha da Borracha the Brazilian government promised to reform the seringal system an d put an end to the debt peonage and physical punishments that had made the rubber industry so infamous. Tappers were to have written contracts with their patrons that would stipulate the percentages that would be paid and how the exchange would take plac e (Secreto, 2007) In theory, these reforms should have led to great changes in the relationships between patrons and tappers, but in practice the government had little or n o capacity to enforce these laws, and much of the abuse associated with the debt peonage system continued. When Napoleo Oliveira arrived as a migrant from the Northeast, the rubber estate patrons still had the contacts with merchants who furnished suppli es, with the government agencies that had replaced the aviadores as a source of credit, and with the local police and politicians who were charged with creating, implementing, and enforcing legislation. In 1958 Napoleo Oliveira moved his family to t he Seruini River, which was considered part of the Seringal Meteripu, and built his house. A village of extended family households began to form around this small settlement, which came to be known as Vila Limeira. The inhabitants of Vila Limeira contin ued to live and work in the patron client regime of the seringal system that existed virtually everywhere in the interior of Lbrea at the time. When rubber was still a valuable commodity they were engaged in the regime do toco in which all merchandise wa s bought from the owner of the Seringal Meteripu in exchange for rubber and other forest products. As noted in Chapter 3, much of the power of the seringal owners was based on their monopoly of key resources and their ability to provide supplies in exch ange for forest products. As previously noted the regime do toco consisted of a system of
129 generalized exchange' in which both parties had expectations about what they would provide to and receive from the other party. The government investment resultin g from the Washington Accords and Soldado da Borracha program during the 1940s revitalized the rubber industry and in many cases revitalized the power of seringal owners, in spite of the government's rhetoric about eliminating the abusive practices of the seringal system. When Oliveira arrived at Meteripu it was still owned by the infamous Coronel Umbelino Bezzera. In his travels up the Purus in 1921, Joaquim Gondim listed Umbelino as the owner of the Seringal Meteripu, although there is no mention of h is conduct vis vis the tappers working there at the time (Gondim, 2001) He was still the owner in 1958 when the elder Oliveira arriv ed, and his fame as a particularly perverse patron persisted nearly seventy years later when this research was conducted in 2008 and 2009. Individuals living as far away as the Ituxi River told a nearly identical story in which Coronel Umbelino had once c aught one of his tappers selling a backpack full of sernambi scrap rubber to a river trader, and had decided to have his capangas pour kerosene on the sernambi and light it on fire while it was still in the pack on the tapper's back. According to the stor y, the man was burned to death before he could extinguish himself in a nearby stream. This is just one example of the lore and legend that still informed the way inhabitants of the riverine communities in Lbrea understood the patron client relationships of the Rubber Boom period. It is interesting to note that two of the older gentlemen with whom I conducted timeline interviews had actually met this Coronel near the end of his life. One of the men, who also repeated the story of Coronel Umbelimo burnin g the tapper to death, said without hesitation about five minutes after
130 telling the story that he considered Umbelino to be a "good" patron and that he had had no problems when he worked for him in the 1930s. This strange mixture of potentially apocrypha l stories describing patrons brutally treating their clients does not seem to match many of the first hand accounts of more benevolent treatment. Even elderly respondents always explained that they had not witnessed these episodes of extreme cruelty and t hat this behavior had either happened "before their time" or "somewhere further upriver." However, there are certainly enough first hand accounts of inhuman treatment of rubber workers from the boom period (Lange, 1 914; Mathews, 1879; Woodroffe, 1914; Yungjohann, 1989) that I am reluctant to completely dismiss the oral histories that I collected during my research. However, many of the apocryphal stories may also have been part of what James Scott (1990) has called the "hidden transcripts", which amount to stories, legends, songs, and other forms of expression that are used by the powerless in certain settings as a discourse that challenges real (or perceived) imbalances in their interactions with more powerful actors. Therefore, embellished or exaggerated stories and legends about cruel patrons could have simply been another means of struggling against the real or perceived injustices of the social system in which riverine people found themselves. In Chapter 6 I delve further into the different emic and etic understandings of the word "patron," how they have changed over time in Lbrea, and what these changes may mean in other parts of the Amazon and Brazil. Respondents in many of these interviews grouped historical events into periods of time instead of responding to timeline questions with exact dates. People said that the
131 patrons like Coronel Umbelino existed in their grandfather's time Period of the Coronis and patron client relationships that included these kinds of abuses no longer existed when they were born. Another team of researchers who interviewed individuals in the area in the late 1990s also noted how their understanding of historical events was centered around periods of time, instead of concrete dates (F. S. D. d. Santos & Muaze, 2002) The tendency of responde nts to group events into general periods instead of specific dates complicated attempts to develop linear timelines of events. Coronel Umbelino died, presumably of old age, in the late 1950s or early 1960s at the Seringal Meteripu and was replaced by anot her patron, his son in law, Eurico Castelo Branco, to whom the Oliveira family in Vila Limeira continued to sell their products. At some point after the death of Coronel Umbelino the arrangement shifted to a "rent" system in which the inhabitants of Vila Limeira would pay the patron thirty kilos of "rent" for each rubber trail for each tapping season. The families in Vila Limeira began to feel short changed in this new arrangement as they received less and less from the patron and the exchange felt more a nd more lopsided in his favor. As described in Chapter 3, the ability of patrons to provide supplies on credit and their control of certain resources was diminished in periods when key commodities lost market value. Hence, their ability to provide all of the services that the tappers' had come to expect was also diminished. In spite of this, patrons continued to demand rent from people living on the seringal As the ability of the patron to provide necessary supplies and control key resources diminished the tappers began to form relationships with other individuals in order to meet these needs.
132 When Eurico Castelo Branco passed away his son, Francisco Castelo Branco, took over the area and the families in Vila Limeira paid the rent to him. They paid r ent on all of the extractive activities that they were engaged in, which included rubber tapping, Brazil nut collecting, and sorva tapping. Inhabitants told me these patrons prohibited them from planting permanent crops, like fruit trees, that could be us ed to establish squatter's rights to the land. They were only allowed to plant annual crops, like manioc, that were not seen in Brazilian law as the kind of benfeitoria or improvement that might be used to establish a land claim. The patrons used their c onnections with the police to threaten the elder Oliveira when he attempted to plant orange and lemon trees in the area. Around 1974 Napoleo Oliveira decided to stop paying rent to the Castelo Brancos after nearly fifteen years. His children explained t hat he simply decided to do this of his own volition but that his sons, who while working outside the community had come into contact with people who had informed them about their land rights, had influenced him. They also mentioned during interviews that part of the seringal had been sold at this time to absentee owners who never appeared to establish their claim. Family members of the old patrons kept trying to charge rent and control key resources, like the Brazil nut groves nearby, but the Oliveiras k new that part of the land had been sold and it was not clear who owned and controlled the land. When patrons lose control of a key resource, particularly land, it becomes more and more difficult for them to maintain the exclusivity of their exchange relat ionships with the client because they are less able to sanction clients for breaking the "one patron" rule by withholding the resource (Martin, 2009) This is what appears to have happened in communities like
133 Vila Limeira as the old seringal owners became less able to provide credit and as the in habitants began to become aware of the tenuous nature of the patron's land claims. In 1975 or 1976 Francisco Castelo Branco moved to Manaus and left his son Edmar Castelo Branco in charge of the area to collect payments from the families that would still pay rent, and determine how resources would be used. This patron maintained control of the Brazil nut groves near Vila Limeira, which by this point had become the most valuable commodity with the continued fall of rubber prices. He claimed to own the Bra zil nut groves and would not allow the Oliveira family to harvest there, bringing his own workers to harvest the Brazil nuts. In the 1970s and 1980s many of the old seringal owners began to move to the cities. Some left managers in charge of their seringa l areas or sent them periodically to collect rent. Some of the old seringalistas simply died and left no one to run their lands. Edmar Castelo Branco, the last patron to charge rent and determine resource use in and around Vila Limeira, moved to Manaus in 1993 or 1994. This exodus of the last remaining old time seringal patrons was almost complete by the time I arrived to conduct research in 2008. There were a couple of old time patrons who continued to charge rent on Brazil nut production in other commu nities along the Purus, but most had long ago abandoned the area and moved to Manaus, Porto Velho, or Lbrea. The exodus of the seringal owners led to a situation in which river traders began to replace them as a source of supplies for many families. New relationships began to form based on the same old aviamento debt based exchange system. As described in Chapter 4, riverine people in Lbrea have long been engaged in production activities that changed from one season to the next, and often from one commo dity to the next
134 based on periodic boom/bust economic cycles. In the face of the continuing collapse of rubber prices, families once again changed activities and began to emphasize production activities that had previously been secondary or tertiary, like sorva, Brazil nuts, animal pelts, copaiba oil, fish, and other forest products. Increasingly people began to shift their production activities to agriculture and fishing to replace the faltering extractive industries. In the late 1970s a timber boom be gan in the riverine communities along the Purus and Ituxi Rivers. By the middle of the 1990s approximately 680,000 cubic meters (24 million cubic feet) were being logged along the rivers in Amazonas state (Hemming, 2003) Most of the people in Vi la Limeira worked as loggers during this timber boom in response to the collapse of rubber. I was informed that the system worked much as it had with rubber and Brazil nuts; they were sold supplies, tools, gasoline, etc. on credit and they were expected t o repay the loan with the timber. Many inhabitants combined their logging activities with copaiba oil production. Copaiba oil prices collapsed around 1995 and Brazil nut prices also slumped for several years. I learned that the changes in resource use patterns also led to changes in settlement patterns. When rubber and Brazil nut prices were high, there were families living in remote areas up the many igaraps or streams that feed into the Purus and its tributaries. Families often lived many miles fro m the main course of the river, working the remote Brazil nut and rubber areas, known as centros Some families would spend the dry season in the vrzea near the main course of the Purus, which has better rubber tapping areas, and the wet season would be spent in the uplands where the Brazil nut groves were located. With the collapse of rubber, people began to spend less time in
135 the centro and established more permanent residence along the main course of the river where they could fish and do beach agricu lture. It became harder and harder to get supplies in the more remote areas and it made sense to move to the main course of the river where the river traders would pass. According to the inhabitants of Vila Limeira, the 1990s were a time of economic har dship in their community and in the other communities in the interior of Lbrea. Rubber prices collapsed between 1986 and 1992, according to the inhabitants of Vila Limeira due to changes in fiscal policy, the elimination of subsidies, and the removal of protective tariffs. The timber industry that had supplanted rubber in many of the communities during the late 1970s began to slow down as most of the trees that could be easily floated to the main waterways had already been cut. At this time, the governm ent began to restrict timber production. The remaining seringal owners were often involved in the timber business and many allowed loggers to cut trees on their land for part of the profit. These economic and social changes led to the departure of sering al owners, which forced inhabitants to expand their interactions with river traders to supply their needs. The economic changes also led to changes in resource use, which often led to conflict between the different groups. Negative Social Currents Res ource Conflict Fishing Conflicts The meandering bed of the sediment rich Purus River periodically changes course over a period of years, creating hundreds of oxbow lakes rich in aquatic resources. Since the colonial period the Purus has been an important source of fish and turtles for urban centers like Manaus. With the population increase in urban centers like Manaus, Porto Velho, and Rio Branco since the 1950s, the demand for fish for these markets
136 skyrocketed. As the fisheries near these urban centers became depleted, fishing boats began to go further and further to satisfy market demands. Improvements in refrigeration equipment allowed the larger fishing boats to travel hundreds of miles in search of fish. At the same time that over fishing and inc reasing market demand were pushing commercial fishermen to explore areas farther from market centers, families along the Purus, like those in Vila Limeira, were increasingly turning to agriculture and subsistence fishing to offset the collapse of rubber an d other forest products. Inhabitants of Vila Limeira informed me that in the 1980s and 1990s commercial fishing boats began to arrive from distant cities to exploit the lakes upon which communities like theirs depended for their main source of protein. People in Vila Limeira recalled large fishing boats coming down the Purus from Acre to trawl in the lakes. Sometimes these fishermen paid the seringal owner or even the people in the communities and other times they simply arrived and began fishing. With trawling nets and other equipment these fishing boats could remove virtually all of the fish from a lake in just a few days. In addition to the conflicts with commercial fishing boats from urban centers, the local collapse of extractivism in the area led many families to move to the city of Lbrea in search of economic opportunities. Many of these individuals who moved to the cities found that small and medium scale commercial fishing was one of the few economic opportunities available to them. Those of more modest means fished from canoes with basic equipment or fished for someone else. Others invested in small boats with some storage capacity capable of capturing relatively large amounts of fish in little time with
137 trawling nets. The increasing avail ability of relatively inexpensive nylon fishing nets and other equipment increased the capacity of fishermen to catch fish and enabled even riverine people with small amounts of capital to buy fishing equipment. Many of these fishermen who had once lived in communities in the interior felt that they still had rights to fish in these areas and that the laws of free navigation permitted them to take their boats wherever they wished. Increasingly this led to conflicts between these small and medium sized fis hing operators and the families that continued to live in the rural communities. The uncertainty and confusion regarding land title and resource use rights in the Brazilian Amazon further complicated the situation. Riverine communities generally had no r egistered land titles and many of the old patrons either did not have title themselves or had not paid land taxes on these old properties for many years, leaving the ownership of these resources in question. At best, most riverine communities had simple p ossession rights to the areas where they lived, which offered limited legal security and tenure rights. Brazilian laws regarding the free navigation of rivers led fishermen to believe that they could fish anywhere, and legislation stating that seasonally flooded areas were officially under the control of the Brazilian Navy also complicated resource use and land tenure claims (Benatti, Surgik, Treccani, McGrath, & da Gama, 2005) In most cases the land tenure situation was unclear, fraudulent titles were common, and overlapping land claims were the norm. Traditional Patrons and Resource Use The transition from a patron dominated seringal system to "free" communities was also related to the resource use conflicts and resource mining that were occurring along the Purus when this research was conducted. The case of the Seringal Realeza is an
138 example of this phenomenon. The Seringal Realeza was one of the better known seringais on the Middle Purus and was once bustling with rubber tappers. Since the fall of rubber prices many of the families that once lived there moved to other areas or to the cities. However, a group of families stayed, forming a community also known as Realeza. On the land that was onc e part of the Seringal Realeza there was a large lake that the former patro managed. He built a floating dock on the lake and maintained guards there to keep the communities and anyone else from fishing on the lake, which led to the build up of sizeable fish and turtle populations. The patron client power structure on the seringais was established over more than one hundred years in Lbrea and people had come to respect a code of rules associated with this structure. Older inhabitants informed me that everyone was aware of these rules. A fregus or client had permission to use certain resources on his patron's seringal but would never hunt or fish on another patron's land without permission. This all began to change when the patrons started to abandon t he area after the rubber crash in the 1970s and 1980s. Patrons often no longer maintained a physical presence on their lands and began to rent out their lakes and forests to fishermen and loggers. In other cases, like in Vila Limeira, seringais were sold to absentee landowners who rarely, if ever, visited the area. In still other cases fishermen and loggers simply began using the areas without permission because the only people left were the patrons' former clients who did not "own" the land either. In t he case of Realeza, when the patron decided to abandon his claim to the area and stopped protecting the lake, the rules that had been established over so many
139 years regarding the use of the lake resources were suddenly considered invalid, and many of the s urrounding communities and even indigenous communities from across the river started to exploit the fish resources in the lake. In effect it became an "open access" resource because the old patron left and no one knew to which of the nearby communities th e lake "belonged." People began trying to take as much out of the lake as they could before the other nearby communities did the same. In addition to the conflicts with other resource users, inhabitants of riverine communities found difficulties in enteri ng the fishing market in a way that enabled them to profit from this activity. Commercial fishing on a medium or large scale required equipment that the inhabitants of the communities did not have at their disposal. Boats, fishing nets, motors, and other equipment were often financially out of their reach. Certain inputs were also necessary to engage in commercial fishing in the area. For fishermen selling fresh fish, ice was absolutely crucial and at the time of this research there was only one ice pro duction facility in Lbrea, which never seemed able to produce enough ice for all the fishermen. In addition, ice storage was only feasible for individuals living near enough to the city to obtain ice and use it before it melted. Communities further fro m Lbrea produced dried fish for commercial markets. Although they were not dependent on ice to engage in this market, they usually depended on considerable amounts of salt that they applied to the fish before drying it in the sun. Without a constant sou rce of salt this activity was also difficult. Without their own boats to transport fish to market, communities often found themselves once again dependent on river traders who would buy their fish at low prices in exchange for high priced supplies.
140 Ma ny communities in the 1980s and 1990s found themselves in a similar situation as Vila Limeira in which their ability to protect and manage the fishing resources they had come to depend upon was complicated by the unclear and unenforced de jure property leg islation in the Brazilian Amazon. In addition, the de facto property and resource use rules that had become established over nearly one hundred years on the seringais were disintegrating as economic collapse forced old patrons to abandon these areas, lead ing to a situation of "open access" resource use in many cases. Communities like Vila Limeira found themselves facing the unenviable task of trying to defend these resources from both local and regional fishing operators who believed that they were also w ithin their rights to fish in these areas. Their limited access to necessary equipment and inputs meant that the possibilities for the inhabitants of Vila Limeira, like most riverine communities, to engage in commercial fishing were generally limited to e xchanges with river traders, which they felt were lopsided in favor of the traders. Negative Economic Currents The collapse of extractive industries along the Purus and its tributaries forced riverine inhabitants to search for alternative economic activit ies. As we saw in the previous section, fishing was an important alternative, albeit a complicated one often associated with its own complications. The timber boom from the late 1970s to the late 1990s provided some employment opportunities, and occasion ally prices for Brazil nuts and copaiba oil were high enough to convince them to engage in these activities. However, agriculture increasingly became the main production activity for the inhabitants of Vila Limeira.
141 For most of the riverine communities in the Middle Purus beach agriculture was the only viable option. Most of the communities did not have terra firme uplands and depended on the uncertain flood cycles of the Purus. Since the period of the first rubber bust these communities had experience with agriculture, upon which many depended for survival in the periods when rubber prices crashed (Gondim, 2001) In spite of their ex perience and expertise producing manioc, beans, and other crops, much as the case with their fishing activities, they faced limitations due to a lack of equipment and infrastructure, transportation, and storage. Even manioc production, which is ubiquitous throughout the Brazilian Amazon, required basic equipment. A forno de farinha or manioc roasting equipment was required and implied some investment. Production of crops like sugar cane required even more equipment and investment. As described in Chapte r 3, the opportunities to access loans to buy expensive farm equipment and transport it to isolated communities with no road access were generally non existent. Riverine communities also confronted a lack of seeds and had to contend with the ravages of pe sts that could also impede agricultural activities. Again, the lack of transportation often left riverine communities dependent on river traders to sell their products. Land Tenure and Indigenous Lands Fishing conflicts and the transition to agricultura l activities were not the only worries of the people in Vila Limeira. In addition to the conflicts with those who claimed to be the owners of the Seringal Meteripu and the surrounding area, they were increasingly at odds with indigenous groups on the upp er Seruini River that also claimed the lands and resources in the area. An indigenous post had existed on the upper Seruini River since at least the 1920s, when it was visited by Joaquim Gondim
142 during his travels up the Purus (Gondim, 2001) This post was run by the Servio de Proteo aos ndios or Indian Protection Service (SPI) and included indigenous schools and production facilities for sugar and manioc flour (Gondim, 2001) A group of Apurin lived and worked at this post, which came to be known as Pedro Dantas or Marien and was managed by a former military officer from the Northeast of Brazil who later became the owner of the Seringal Caado (Kroemer, 1985) A brief description of the post from 1921 gives the impression that the indigenous inhabitants were expected to work at the post in hunting, fishing, and agriculture and may not have been the beneficiaries of the items produced there (Gondim, 2001) The agricultural production was insufficient to support the 85 Apurin that lived at the post in 1927, much less the 300 that lived in the area around the post (K roemer, 1985) The SPI furnished food, supplies, and other necessities to the indigenous inhabitants at the post and in 1929 there were 91 Apurin living there (Kroemer, 1985) The SPI post fell on hard times in the 1930s and by 1940 only 25 Apurin were living nearby; the rest had become incorporated into the seringais along the Purus as cheap labor (Kroemer, 1985) The post was deactivated around 1950 and its lands wer e sold to the MANASA company, which was developing timber extraction activities in the area, by the son in law of the official in charge of the post (Kroemer, 1985) Indigenous groups in Lbrea began to organize in the 1970s, and with the help of the Catholic Church, CIMI, OPAN, FUNAI, and other organizations to reclaim areas that they had occupied prior to the rubber boom. They began to form relationships with these groups and agencies that were interested in help ing them, in much the same way that riverine communities would later build their own social capital. One older resident
143 of Vila Limeira remembered that around 1976, lands that had been part of the Marien indigenous post on the Upper Seruini became the Ar ea Indigena Marien This indigenous area was quite some distance up the Seruini River and apparently did not have large impacts on the resource use of the inhabitants of Vila Limeira. However, the tensions over resource use were coming to a head when I conducted my research in 2008 and 2009. The inhabitants of Vila Limeira informed me that the indigenous people living in the Indigenous Lands in the Upper Seruini were moving downriver to be closer to the main channel of the Purus, and wanted to take cont rol of resources to which the people in Vila Limeira believed they had a claim. Two important Brazil nut groves on the Seruini were at the heart of this conflict. The inhabitants of Vila Limeira claimed that they had rights to the area and that the indig enous people were trying to expand the size of their Indigenous Lands to take over an area that they already occupied. I was not able to interview people from the indigenous community to get their side of the story. However, regardless of who had the bet ter claim, the end result was a conflict over resource control. The creation of Indigenous Lands along the Purus and near Vila Limeira began in the 1970s. An older resident informed me that the first to be created was the T.I. Catipari (So Jerome), cre ated in 1974. In 1980 the T.I. Marien was created and in 1986 the T.I. Mamoria was created, which was an area occupied by the Jamamadi. In the 1980s and 1990s there was an explosion of indigenous land creation in Lbrea. In many cases the creation of t hese areas led to the expulsion of all non Indian peoples, including riverine people. Communities that had existed in an area for two or three generations were suddenly informed that they would have to move elsewhere. In
144 some cases they were told that th ey would be compensated for infrastructure and permanent plantings, like fruit trees, but not for the land itself. Under Brazilian law, when a indigenous area is created all land titles existing in the area are annulled and the land becomes property of th e state, to be managed for the benefit of the indigenous group for which it was created. Private lands can be expropriated to create Indigenous Lands, even if the land title is legitimate and all property taxes have been paid. Some people received a smal l compensation and attempted to move to the city, where the money ran out quickly. Others were never compensated at all. Some communities stayed in spite of constant conflict with the indigenous people, waiting until the government provided their compens ation. In the attempt to protect indigenous cultures and right a historic wrong, the creation of these indigenous lands represented a major threat to the riverine communities, many of whom were terrified of being forced to move to the cities where they we re ill equipped to survive. The threat of being expelled from their lands due to the expansion of existing Indigenous Lands or the creation of new ones was a cause for extreme concern for the inhabitants of Vila Limeira, who felt they would have nowhere to go once expelled from the area. This threat served as a major motivation for the inhabitants as they began forming networks with government agencies and social movements to ameliorate the constant threat of being pushed off the land they lived on or of losing access to key resources, like Brazil nut groves, that helped them eek out a living. Vila Limeira was not the only riverine village engaged in conflict with indigenous groups over the control and use of key resources. The nearby community of Ajur icaba, which had also formed on the site of an old seringal was also engaged in a conflict with
145 indigenous communities from the nearby Catipar/Mamori Indigenous Land over the control and use of a large Brazil nut grove. The boundaries of the indigenous area were unclear, and both sides believed that the Brazil nut grove belonged to them. Indigenous lands, national parks, and other conservation units with unclear and overlapping boundaries were unfortunately not unique to Lbrea, nor were the resource c onflicts that can be exacerbated by these boundary problems, and were common throughout the Brazilian Amazon. The riverine community of Vrzea Grande also had a conflict with indigenous communities from the Catipar/Mamori Indigenous Land. Several indige nous fishermen began predatory fishing operations in a lake that the community was attempting to manage. The indigenous fishermen claimed that the lake was going to be included within another Indigenous Land that was going to be created. In the communit y of Madeirinho there were also conflicts between riverine communities and indigenous communities over the Sacado Lake, an important fishing area. The indigenous community claimed that the riverine community was located in an area that was going to be inc luded in their Indigenous Land, which was due to be augmented. The riverine community of Jurucu, located on the site of another old seringal was informed by the land agency INCRA that their community was located in the Jarawara/Jamamadi/Kanamati Indige nous Land and that they would be ineligible for Incra assistance until their community either moved out of the area or proved that it was not located in the Indigenous Lands. The limits of this Indigenous Area on the ground
14 6 were once again unclear, which was a common problem throughout the Brazilian Amazon (Ricardo, 2004) The inhabitants of Santa Rosa lost virtually all of the land behind the community with the creation of the T.I. Ilha Verde around 1998. T hey had received no notice about the creation of this indigenous area until one day when they found officials demarcating a boundary behind their community. The creation of this Indigenous Land also negatively impacted the community of Praia do Buraco, wh ich lost the terra firme upland areas where they had always planted manioc to produce the maniva cuttings used to plant their manioc fields. The creation of indigenous lands on the Sepatini and Caina Rivers also displaced riverine peoples, who were forc ed to leave their homes, crops, and other resources and move elsewhere. I was unable to interview indigenous people in these communities and my goal is not to demonize them in any way. Indigenous peoples along the Purus were brutally displaced from thei r lands, massacred in correrias subjected to death from diseases to which they had no immunity, and often forcibly assimilated into the extractive economy and Brazilian society (Kroemer, 1985; Ribeiro, 1996) However, it is important to note the impact that the creation of indigenous lands have had on riverine populations, who are also a poor and often marginalized group. The threat of being expelled from their commun ities due to the creation of indigenous lands was one of the catalysts that inspired riverine people on the Purus to form relationships with government agency personnel and social movement groups. They believed that these agents could use
147 their position a s brokers and intermediaries to higher government officials to help them resolve their resource and land tenure problems. Networking, Social Capital, and Positive Social Currents In Chapter 4 I described the difficulties that the inhabitants of riverine communities face in confronting seasonal production cycles, lack of health care, and lack of access to formal loans and credit. In this chapter I have described resource conflicts and limitations to production activities specific to the community of Vila Limeira, but also impacting other riverine communities along the Purus. These are examples of the negative social and economic currents that riverine inhabitants faced in their difficult existence. However, as I previously stated the inhabitants of thes e communities were not passive actors, content to sit by as these currents swept them away. For many of the former inhabitants of these communities migration was seen as the most appropriate solution to these hardships. The rural urban migration that beg an in Lbrea in the 1970s had totally reversed the population dynamic by the time this research was conducted in 2008 2009. In 1970 80% of the population lived in rural areas and 20% lived in the urban center. By 2000 the rural population had shrunk to 2 0% and the urban population had risen to 80% (see Table 3 3). The move to the cities was itself fraught with uncertainty as people found themselves in a new environment for which they were not prepared. The skills they had honed over several generations to survive in the interior did not necessarily serve them in the city where education and a trade were generally prerequisites for finding employment. For those without skills and education, a life in the poorest, most crime ridden neighborhoods was usual ly the only option. Work in the informal sector selling
148 candy on buses or washing cars might be the only type of employment available for people attempting to make the difficult transition from rural to urban existence. Some families gave up on the urba n dream and moved back to rural areas, with many stories about the hardships of trying to survive in the city, but the lure of health care and education were powerful draws that still inspired people to move to the cities. Many families found themselves i n the position of making a difficult choice between an increasingly difficult rural poverty and an unfamiliar urban poverty with its own set of uncertainties. As word of the difficulties of making the transition to an urban existence filtered back to the c ommunities in the interior, many inhabitants decided that flight was not a viable option and that they would have to find solutions that would enable them to remain in their communities while at the same time improving their standard of living. In the fol lowing section I describe how the inhabitants of Vila Limeira have formed strategic relationships in order to attempt to resolve the fishing conflicts, production difficulties, and creation of indigenous lands that they perceived as threats to their existe nce. Patron Conflicts and the Catholic Church The relationships between the extended Oliveira family in Vila Limeira and the various patrons of the Seringal Meteripu were not the only interactions the Oliveiras had, in spite of the extreme distances and isolation that separated the community from urban areas. The economic downturn for extractive products like rubber significantly weakened the ability of the seringal owners to maintain their control of the resources that were the basis of their patronage arrangements. However, at the same time, dramatic social changes were taking place that led their clients to further question these
149 arrangements. Even as the old patrons were abandoning the area due to the collapse in rubber prices, new actors began to a ppear to whom people in Vila Limeira could turn for help and information. The Catholic Church in Lbrea was a key actor in helping to establish the first social movement organizations in the area that would challenge the old patron client relationships t hat had become so entrenched after nearly 100 years. The Prelacy of Lbrea was the headquarters for the Catholic Church for a sub region of the state of Amazonas, consisting of the municipalities of Lbrea, Tapau, Canutama, and Pauini, and covering a geo graphic area of 230,642.23 square kilometers. Each municipality had a sede or headquarters with a coordinator. In Lbrea there were several religious orders that formed the prelacy, including: Dios es Sano Maristas and Agostianas Recoletas At the tim e this research was conducted there were eight priests and four nuns working in the prelacy, with several thousand lay members assisting with a variety of tasks. The Catholic Church was one of the oldest organizations in Lbrea; the Paroquia was first est ablished in 1878, just seven years after Colonel Labre had founded the city (Ferrarini, 1981) The Vatican recognized Lbrea as a prelacy in 1925 and the Agostinianos Recoletos have run the organization since that time ( Parquia Nossa Senhora de Nazar Lbrea AM 125 Anos Evangelizando 2003) Christian Base Co mmunities The first attempts at social movement organization in Lbrea were centered on the formation of Comunidades Ecleisias de Base or Christian Base Communities (CEBs), which began in the 1970s. CEB formation is an example of the how the Catholic Chur ch shifted to a pro poor approach to evangelization in the late 1960s (Bruneau, 1986) The CEBs were designed to offset the lack of priests by teaching individuals
150 who showed leaderships qualities in the communities to read and conduct basic sermons and church servic es (Boff, 1985) The CEBs were a strategy used throughout Brazil by the Catholic Church in response to t he tremendous expansion of evangelical churches, and to offset the chronic shortage of priests and nuns (Adriance, 1995) T he church realized that preaching alone might not be very successful, and they also incorporated community organization and attempts to improve basic services like health and education (Bruneau, 1986) I was informed that the Prelacy of Lbrea had helped organize 215 CEBs along the Purus between 1975 and 1982. The CEB program came at a crucial time when patrons were abandoning the seringais and when former rubber tappers were beginning to cluster in riverine villages instead of spread out colocaes In many ways th e CEBs appear to have been one of the first seeds of organization in these communities. One of the lasting results of the CEB program was the formation of community leaders who in many ways began to serve as brokers between the newly forming communities a nd the outside world. Although the inhabitants of Vila Limeira remembered the organization of CEBs and the work of priests in helping communities, they had become an Evangelical community around 1997, and perhaps for this reason they did not speak much a bout the formation of a CEB in their community. Religious affiliation appeared to have an effect on the contact that communities had with different religious groups. It is understandable that when allocating extremely scarce resources organizations tend to prioritize the communities that are more likely to be receptive to their message. For this reason there was a tendency in some cases for Evangelical organizations to focus on Evangelical
151 communities, and for Catholic organizations to prioritize Catholi c communities when they traveled to the interior. Rural Workers' Union At the same time that the Catholic Church in Lbrea was inspiring organization at the community level, they were also building links between the inhabitants of riverine villages and wider peasant movements in Brazil. Rural Workers' Unions became one of the most important avenues for peasant organization in the 1970s and 1980s in Brazil. These organizations were set up to help peasants in their struggles for land and workers' rights. I was told that in Lbrea the main motivation for the formation of the sindicato was the constant struggle between rubber tappers and patrons on the seringais Tappers felt that they were being cheated on the debt ledgers and in the prices that they pa id for supplies. They felt that it was unjust that they could be thrown off their production area if they sold to someone other than the patron. I learned that the priests with their new pro poor approach to evangelization increasingly took the side of t he tappers in this struggle and felt it was their duty to help the tappers receive the rights as workers they were due under Brazilian labor laws. A former member of the union directory informed me that in 1978 an influential priest decided to help organi ze a rural workers' union in Lbrea and began the process of holding meetings about the concept. Meetings were held and contacts were made with state level organizations like FETAGRE and national organizations like CONTAG, who sent individuals to hold mee tings to discuss the creation of the union. The Sindicato de Trabalhadores Rurais de Lbrea was legalized in November 1979.
152 I learned that the stevedores who loaded and unloaded the ships that arrived in the port of Lbrea were also interested in this t ype of organization and there was a movement to form the Sindicato dos Estivadores at the same time. The Stevedores Union actually began to organize first, but never really picked up momentum and quickly faded. I was told that the prelacy also attempted to help form an Associao de Lavadeiras or Washerwomen's Association to help these chronically underpaid domestics obtain a just wage for their work. The formation of the rural workers' union meant that tappers now had an organization that they could go t o for help with injustices on the seringais and from 1979 through the mid 1980s the union had a significant impact on the patron client relationships in Lbrea. I was told that union leaders would review tappers' debt ledgers to determine if they were be ing cheated and would help them obtain legal counsel in the event that a patron was trying to throw them off the seringal The leadership of the union was connected with other unions in the area through regional meetings with FETAGRE at the state level an d with national organizations like CONTAG. I was informed that at the time the union was formed the only other organizations that tappers could turn to for help were the Prelacy and the CDH Commiso de Direitos Humanos (Human Rights Commission), an orga nization that existed for a time in Lbrea but was long since defunct by the time this research was conducted. The union had approximately 300 to 500 paying members at the height of its power. In the late 1980s the directorate of the union changed and th e new president took it in a different direction by making a pact with the municipality, in effect turning the
153 union into a pelego 2 union that no longer looked to challenge the imbalances of power that existed between landowners and workers. The Rural Wo rkers' Union of Lbrea still existed in 2008 2009, but it was no longer considered to be a strong social movement organization when this research project was carried out. Its main function had become assisting individuals to obtain the necessary paperwork to receive their retirement benefits as agriculturalists, and to access other government benefit programs. In order to receive retirement benefits as agriculturalists from the Instituto Nacional de Seguro Social (INSS) individuals had to present a series of documents proving that they had worked as small farmers. Proof of membership in a Rural Workers' Union was an important element of this bureaucratic process. Individuals were permitted to apply for these benefits, which consisted of a monthly stipend of 415 reais per month, at 55 years old (women) and 60 years old (men). People also sought out the Workers' Union to obtain paperwork to access auxilio de maternidade (maternity benefits) and auxilio de doena (disability benefits). When this research w as conducted the union had approximately 3,300 members who were supposed to pay a monthly membership fee. The fee had been 7 or 8 reais per month, but at the time research was conducted the fee had been dropped to 2 reais per month to try to convince memb ers to pay their dues. The money from these fees was sent to the FETAGRE and CONTAG headquarters, which would return approximately 5% of the money for the operational costs of the union in Lbrea. The office was in a state of disrepair and had extremely limited staff and equipment to conduct its business. The 2 A peleg o union is one in which the officials have be coopted or coerced to favor rich landowners. Pelego means "sheepskin" and this term is similar to the expression "sheep in wolf's clothing" (Adriance, 1995)
154 president of the union in 2008 2009 informed me that it was now very rare for individuals to seek out the union to resolve problems with patrons and to resolve other labor conflicts. He explained t hat people generally went to other social movement organizations, like the Pastoral Land Commission, to try to settle the types of problems that the union had once helped resolve. The main reason that people sought out the Rural Workers' Union was to obta in paperwork to access government benefits. Due to its location at the extreme western end of the municipality, Vila Limeira was not often visited by Rural Workers' Union of Lbrea. The union, like most of the social movement and even government agencie s was chronically short of the financial means to make long journeys into the interior to visit communities. However, inhabitants of Vila Limeira did remember that around 1988 the Sindicato de Trabalhadores Rurais de Pauini which is much closer despite b eing in a different municipality, began to come to the community to help them get organized. The union helped them in their struggles with the patrons, who were still attempting to charge rent, as well as with fishermen and loggers from outside the commun ity. One of the most important outcomes of the formation of the CEBs and Rural Workers' Union was the diffusion of information regarding peoples' land tenure and labor rights. The inhabitants of Vila Limeira, and countless other communities on the now d efunct seringais were shocked to find out that in many cases they were being charged rent for occupying land that their patron claimed was his but was not included in any official land title. As previously mentioned, when patrons lose control of a key res ource, particularly land, the "one patron" rule becomes more difficult to enforce, which often leads to a change in social structure (Martin, 2009) When riverine
155 inhabitants realized that the patrons often had no legal authority to charge rent or determine they way they used resources, community inhabitants became emboldened to challenge the longstanding patron client relationships. Pastoral Land Commission The creation of the Comisso Pastoral da Terra or Pastoral Land Commission (CPT) in Lbrea was another very important result of the initiativ e of Catholic priests and lay members in helping communities in the interior fight for their rights. CPT was established in Brazil in June 1975 during a meeting of the Pastoral da Amazonia convened by the Conferncia Nacional dos Bispos do Brasil or Natio nal Conference of the Bishops of Brazil (CNBB) ( O nascimento da CPT 2010) The organization was created in response to the land conflicts that were worsening in the Amazon i n the 1970s and focused on helping posseiros 3 that were being subjected to violent expulsion and unjust working conditions ( O nascimento da CPT 2010) I was informed that th e CPT office was established in Lbrea soon after the creation of the Rural Workers' Union. In part it was established to assist the union, which had begun to lose its effectiveness as a force of social change. When the CPT arrived in Lbrea the BR 230 T ransamazon Highway was also arriving; it reached the city in September 1976 (Ferrarini, 1981) However, the arrival of the road did not lead to the huge land conflicts that characterized the building of the Transamazon in Par and other parts of the Amazon (Schmink & Wood, 1992) Instead, the CPT office in Lbrea dealt mainly with conflicts between patrons and clients, riverine and indigenous peoples, and between riverine communities over shared resources. With the economic 3 Posseiros were typically poor farmers who made claims to unoccupied land in frontier areas by clearing it, planting crops, and building rudimentary structures.
156 collapse and abandonment of the seringais CPT increasingly began to try to help resolve conflicts over aquatic resources. Fishing Conflicts and Social Movements In response to the persistent conflicts with local and regional fishermen riverine people went first to local authorities like the municipality, the police, and IBAMA to denounce these problems. Unsatisfied with the general lack of action or capacity to resolve these problems, they next went to groups like the Pastoral Land C ommission (CPT) to try to resolve the problem. CPT was faced with the challenge of helping people resolve a variety of different types of conflict over fishing resources. They were asked to help with conflicts: 1) within communities between different hou seholds; 2) between different riverine communities; 3) between indigenous and riverine communities; 4) between riverine communities and local fishermen, many of whom had once lived in the communities before moving to the city of Lbrea; and 5) between rive rine communities and large scale fishermen from Acre and from Manaus. In response to these community requests to help resolve fishing conflicts, CPT first suggested two programs: 1) the Lagos de Preservao or Lake Preservation Program, and 2) the Agent es Ambientais Voluntarios or Volunteer Environmental Agents. The CPT staff in Lbrea decided to offer these solutions as a result of their interactions with other CPT offices in the state of Amazonas, who were developing these programs to help resolve sim ilar problems in other parts of the state. The Lake Preservation Movement began in the Prelacy of Tef in the 1970s in response to fishing conflicts that had arisen with the skyrocketing demand from Manaus, which had seen its population explode with the creation of the Zona Franca de Manaus in 1967 (da Costa, Ramos, & de A. Cunha, 2007 ) Through the 1970s and
157 1980s the Lake Preservation program spread to other municipalities in Amazonas including: Parintins, Coari, Itacoatiara, Silves, and Iranduba (da Costa et al., 2007) In 1987 the community of Santa Helena became the first community in Lbrea to participate in the program (da Costa et al., 2007) The CEBs played a key role in the first experiences with the program in the 1970s and in 1975 CPT became involved (da Costa et al., 2007) In Tef, where the program began, r iverine communities began to form Preservation Committees in which certain community members were charged with monitoring community lakes and preventing predatory fishing operations (da Costa et al., 2007) This led to the creation of the Agentes Ambientais Comunitrios or Community Environmental Agents, individuals who would receive b asic training from governmental environmental agencies on how to monitor the use of the lakes and report predatory fishing activities (da Costa et al., 2007) As a result of the early experiences with the program participants decided that they best way to protect their fishing resources was to create distinct categories based on the ty pe of use and protection required. This included Sanctuary or Breeding Lakes, Maintenance Lakes, and Open Lakes (da Costa et al., 2007) In the 1970s legal provisions began to be adopted at the municipal level to try to regulate fishing activities. In 1997 CPT Amazonas signed an agreement of cooperation with the Brazilian Environme ntal Agency IBAMA creating the Agentes Ambientais Voluntrios or Volunteer Environmental Agent program in which they would train community inhabitants to monitor natural resource use in and around their communities (da Costa et al., 2007) Acordos de Pesca or Fishing Accords were formalized with the IBAMA
158 Instruo Normativa 29/03 that formalized the process of officially designating lakes within the protection categories listed above (da Costa et al., 2007) When communities like Vila Limeira and others along the Purus and its tributaries came to the Prelacy and CPT in search of help in resolving fishing conflicts, the groups responded by offering the Lake Preserv ation and Volunteer Environmental Agent programs that they had learned about through their network of contacts in other CPT offices in the region. This is an example of how riverine communities began to seek out other sources of help in the wake of the de parture of seringal owners, and how the patronage pyramid of the seringal system began to change to a multi hub system in which clients sought help from several brokers and intermediaries. Indigenous Missionary Council (CIMI) The policy of the Catholic C hurch regarding indigenous peoples has also changed over the years; although at certain times they had tried to protect indigenous peoples from being overwhelmed by the social changes associated with colonization, their focus for much of their history was to evangelize these groups and incorporate them into Brazilian society. By the 1970s this approach had changed, and the Catholic Church began to recognize the importance of helping indigenous people to maintain their culture, language, and territory. Ind igenous peoples had been pushed off of their lands and incorporated into the extractive economy in Lbrea since Brazilian expansion into the area in the 1850s. The building of large infrastructure projects in the Amazon in the 1970s continued to open up n ew lands, and indigenous groups continued to face expulsion, death from disease, and cultural disintegration (Davis, 1977; Hemming, 2003) In 1972 the CNBB created the Conselho Indigenista Missionrio or Indigenous Missionary Council (CIMI) to help i ndigenous peoples organize and demand their rights
159 to maintain their cultural integrity and the land and resources they depended on for survival ("CIMI Quem Somos", 2004) The organization was created as an alternative to the official policy of the Brazilian government at the time, which was to integrate indigenous peoples into the dominant majority culture ("CIMI Quem Somos", 2004) The Catholic Church was also fighting on behalf of indigenous people in Lbrea. First, in the 1970s lay members and priests conducted studies regarding the situation of indigenous groups in t he area, visiting the indigenous communities, making contact, and collecting basic information. The CIMI office was opened in Lbrea in 1978 and began to help indigenous peoples to organize and demand their rights. A group of indigenous men from Lbrea t hat had served in the Brazilian military returned in the late 1970s and decided to organize themselves to reclaim lands that they had lost during the expansion of extractive industries in the second half of the 19 th century. CIMI began to help these leade rs organize and arrange meetings to discuss their problems and think about solutions. One of the solutions that they increasingly came to regard as imperative was the creation of Indigenous Lands in the municipality. Like riverine peoples, indigenous peo ple had no title to the land they occupied and no guarantee that they would not be expelled or lose access to key resources. The creation of indigenous lands would help resolve this problem, and in the late 1970s and early 1980s they began to push for the creation of these areas. The federal agency in charge of indigenous questions in Brazil, including the creation and management of indigenous lands, is the Fundao Nacional do ndio or National Indian Foundation, better known by the acronym FUNAI. This organization replaced the SPI in Brazil in 1968 after a national scandal involving the SPI
160 administration and an effort to update indigenous policy to include statutes reflecting United Nations and International Labor Organization statements regarding hum an and minority rights (Wagley, 1977) I was informed that the FUNAI began to work in Lbrea in 1981 in connection with the office in Porto Velho, Rondnia and by 1983 they began to demarcate proposed indigenous lands in Lbrea. The FUNAI office in Lbrea was officially e stablished in 1987. The efforts by indigenous leaders and CIMI to create an indigenous area along the Paci River near the city of Lbrea led to conflicts with the business interests that claimed ownership of productive Brazil nut groves and timber in th e proposed indigenous area. Violence ensued as these powerful business interests attempted to eliminate the indigenous leaders that were demanding the creation of the area. There were attempts on indigenous leaders lives, and several family members of on e of the key leaders, the Apurin tuxuau or chief Agostinho, were murdered. On April 28,1985 Sister Cleusa Rody Coelho, a nun from the Augustinian order who had worked off and on in Lbrea since 1954 and had been working with CIMI since 1982 to help organ ize indigenous communities, went to help mediate the conflict on the Paci River (Ferrarini, 2009) In spite of warnings about threats to her safety she went to the indigenous community and was murdered b y an assassin named Raimundo Podivem, who was also involved in the murder of Agostinho's wife and one of his sons ( Parquia Nossa Senhora de Nazar Lbrea AM 125 Anos Evangelizando 2003) Podivem died soon after of an illness and the instigators of the murders were never brought to justice, although t here is speculation that he was hired by local business owners who wanted to maintain control of timber and Brazil nut groves in the area.
161 In 1985 the Terra Indigena Caititu was created, and many believed that the creation of the indigenous area was due in large part to the backlash from the murder of sister Cleusa. From this point on a series of indigenous lands were established in Lbrea. CIMI continued to work on indigenous issues in Lbrea, and in addition to helping with land and resource conflicts they often served as a mediator between government agencies and indigenous communities. The CIMI staff was able to communicate in indigenous languages to a much greater degree than staff from government agencies, and often accompanied government agents on trips to the communities to serve as translators. CIMI agents told me that they felt that another one of their key tasks was to monitor the government agencies, like FUNAI, to make sure that they were doing their work correctly with the communities. CIMI and CPT were officially considered part of the Prelacy of Lbrea, but they were run at times by lay members, and in many ways they operated with a certain degree of autonomy. CPT had a state level headquarters in Manaus, as well as national offices in Brasilia. CIMI also had regional and national headquarters. Typically riverine communities like Vila Limeira had little direct contact with CIMI, unless they were at meetings or other events in which social movement groups gathered to support one ano ther. CIMI also had limited resources and had to prioritize the way it used these resources. However, there was significant contact between CIMI, CPT, and other social movement organizations in Lbrea. They would occasionally meet to discuss and try to help resolve conflicts between their constituencies, and often supported one another in their respective efforts. This is an example of the horizontal
162 links between brokers and intermediaries that can proliferate after the collapse of patronage pyramids (Martin, 2009) Children's Pastoral Pasto ral da Crian a The Pastoral da Criana or Children's Pastoral was created in September 1983 in Florestpolis, Paran, Brazil to help end the problem of child mortality in Brazil. Dr. Zilda Arns Neumann, a pediatric and sanitary physician, was the founder and president of this organization, a task she had carried out at the request of the CNBB. I was informed that the Pastoral da Criana in Lbrea was founded in 1993 by Father Miguel Mohoy, and focused on providing natural "traditional" medicines to poor c ommunities as an alternative to expensive manufactured medicines. The Pastoral da Criana was also part of the Prelacy of Lbrea but with direct links to the organization's headquarters in Curitiba, Paran, Brazil. They had a program to monitor the nutri tional status of children in urban and rural areas and try to provide information on basic hygiene, pre natal care, and other topics associated with children's health. The Pastoral da Criana also served as a channel for donations and some food that was p rovided by the central office. I was told that in 2008 they sent 1.73 reais for each child weighed in the area. When this research was conducted the Pastoral da Crian a was trying to increase its presence in the rural riverine communities along the Purus and its tributaries. I learned that lay members, approximately 126 volunteers, did the vast majority of the work done by the Pastoral da Criana in Lbrea. I learned that the Catholic Church in Lbrea has set up a series of support organizations, like the Pastoral da Criana to assist other social groups. These included the Pastoral da Familia Pastoral do Idoso Pastoral do Menor and the Centro Esperana The Centro Esperana created in 1994, was an organization that
163 attempted to provide a safe hav en for adolescents in Lbrea. They learned a trade: embroidery, sewing, cooking, metal working, woodworking, etc. The 320 young people enrolled in the center also learned discipline and received meals at the Centro Esperana They had to be enrolled in school in order to participate in the center, and participants were chosen to a large degree based on their situation. The main goal of the Center was to target at risk youth and to keep them off the streets and prevent problems with drugs, crime, prostit ution, etc. Demand to participate in this program was high, and I was informed by the priest in charge that if they had the resources they could easily have as many as 2,000 young people participating. The Marista priests that came to Lbrea in 1967 to i mprove the educational system continued to play an important role in the town's educational system, although by 2008 2009 most of the schools were run by state and municipal government agencies. The prelacy also had begun a housing program to try to assis t destitute families fleeing the interior. They were building simple houses for these families, which they could live in, but did not own, until they were able to find employment and housing. Networking with State Agencies to Resolve Agricultural Proble ms The patron's barraco had long been the place where riverine inhabitants had obtained rubber tapping equipment and other implements. With the departure of the patrons and the fall of rubber prices, riverine inhabitants began searching for contacts that could help them obtain the equipment they would need for agriculture and fishing activities. In 1998 the inhabitants of Vila Limeira heard that the state government of Amazonas was providing credit and implements to communities that wanted to engage in a gricultural activities. They received word that in order to receive these benefits community members had to be organized into legalized community associations. Many
164 Brazilian government organizations demanded that communities organize themselves into leg ally registered associations with a directorate (president, vice president, secretary, treasurer) in order to receive government benefits. Community leaders from Vila Limeira decided to take action to create a community association to access the governmen t benefits. Community Leaders One of the legacies of the CEB program along the Purus was the formation of leaders in many of these communities, who began to take action and attempt to bring benefits and improvements to their communities. With the departur e of the patrons and the formation of leaders through programs like the CEBs this dynamic began to change. The CEBs were certainly influential in this process but I was also informed during interviews that many communities had come to the conclusion that they needed representatives to obtain help from the outside world and were forming leaders even in areas where CEBs were not created. When this research was conducted in 2008 2009 most of the communities along the Purus and its tributaries had some form of democratically elected leader who would represent the community in the outside world. Visitors from social movement groups and government agencies would typically seek out these leaders upon arriving in the community. These leaders were typically the ones who would be invited to represent their communities in meetings in Lbrea and Pauini, and were often expected by other community members to seek out assistance for the community, resolve conflicts within the community, and seek out solutions to many o f the larger problems previously mentioned. Since they were generally the ones invited to meetings and sought out by visitors, community leaders became key sources of information for other community
165 members. In my visits to the communities I witnessed co mmunity leaders hold meetings in which they would relay to other community members information they had gathered during their visits to agencies in town. The officials in government agencies, social movements, and local politicians also reinforced the pos ition of community leaders by seeking them out when they wanted to pass information on to communities, secure votes during elections, etc. They obviously knew that this was an effective strategy in reaching a larger number of community inhabitants. In F igure 5 1 we can see that the community leader has replaced the patron as a source of information and advice. Being invited to all of the meetings and having access to information that the other communities members did not have put these leaders in a "bro ker" position between community members and the outside world, that had previously been filled by the patrons. Figure 5 2. Post Seringal Networks However, while the community leaders filled a broker role for information, they had not assumed th e role of linking community members to markets and sources of supplies
166 and credit like the old patrons. Instead, (as I explain further in Chapter 6) community members were most likely to use their relationships with river traders, merchants in town, and g overnment extension agencies to access credit and supplies. Community leaders in Vila Limeira realized that they would need assistance if they were going to negotiate the bureaucracy of the Brazilian government and set up a community association. They b egan to turn to existing relationships outside their community and also began to form new relationships with individuals in social movements and government agencies that could help them achieve this goal. Their first stop was to see the prefeito or mayor of Pauini. Municipal Politicians In the remote towns of the Amazon the city mayor has long been in a position of influence. In much the same way that community leaders had become the link between their village and the outside world, in many ways the mayor s of the small towns of the Amazon were important brokers between their towns and the wider world. The legal responsibilities for creating municipal legislation and representing the Brazilian government at the local level made municipal mayors one of the most important channels through which resources from state and federal agencies reached remote towns. The following quote by Darrell Miller regarding Amazon communities in the state of Par in the 1970s was still very true for towns like Pauini and Lbrea in 2008 2009. Traditionally, the office of mayor has been extremely important in the Amazon. Given the boom bust nature of the economy of that region, the municipal budget represents the one continuous and assured source of income although, granted, it is relatively small when compared to more 'developed' communities. In times of economic hardship, having a position or job with the municipal government means that one has an assured cash income to buy foodstuffs and manufactured goods not locally produce d, whereas others must turn (or return) to subsistence farming and hunting to feed their families (Miller, 1985)
167 I would argue that with the collapse of extractivism and the massive rural urban migration that occurred along the Purus after 1970, the office of prefeito became more important than ever. In 2008 2009, jobs in the municipality were highly coveted due to the lack of employment opportunities and were a source of influence for those in charge of the municipality. A may or was expected to know how bureaucratic procedures, like legally registering a community association, worked and it is therefore no surprise that the community leaders of Vila Limeira began their search for assistance with the mayor of Pauini, the town in closest proximity to their community. City councilmen and councilwomen, vereadores as they are called in Brazil, were also sought out because of their position within the municipal government. Their official tasks, in addition to proposing municipal le gislation, were to monitor the work of the mayor (particularly the municipal budget) and to serve as interlocutors between the public and the municipality. Their position within the municipality, proximity to the mayor, and perceived (or real) influence o n the allocation of municipal funds meant that vereadores were also sought out frequently by communities requesting schools, health posts, and other improvements, or individuals asking for money for food, medicine, and other items. At the time of this res earch Lbrea had nine vereadores who were elected every four years in municipal elections. The vereadores had weekly meetings in the city council chambers to discuss municipal business, propose legislation, and debate other municipal affairs. Municipal s ecretaries were also sought out because of their position within the municipal structure and their potential to provide access to resources and assistance. Municipal secretaries were appointed by the mayor and put in charge of the municipal
168 efforts in dif ferent sectors. At the time of this research Lbrea had municipal secretaries of: 1) Environment, 2) Health, 3) Education, 4) Municipal Administration, 5) Production, 6) Finances, 7) Communication, 8) Social Assistance, 9) Water and Sewers, 10) Sports, an d 11) Cabinet Chief. Municipal secretaries were often expected to be the official liaison between state and federal organizations and their municipality. For example, the municipal secretary of the environment in Lbrea was expected to be in contact with officials and agencies at state and federal levels that dealt with environmental issues. In order to facilitate bureaucratic processes and consolidate the offices of federal, state, and municipal government offices, in 2006 the municipality of Lbrea open ed the Centro Cidado or Citizen's Center. This building contained the offices of many of the offices that provided government benefits to the inhabitants of the municipality. In 2008 2009 these offices included: 1) Bolsa Familia program, 2) Ministry of Employment, 3) Military Services, 4) Identification, 5) CPF, 6) Federal Registry, 7) and some of the services of INSS. Inhabitants of the municipality could visit the Citizen's Center instead of going to offices throughout the town. Officers of the Citiz en's Center would periodically accompany other municipal agencies on visits to the interior to help people obtain the citizen documents necessary to conduct almost all official business. Social Movement and State Agency Networks When the mayor of Pauini refused to help them because they were not in his jurisdiction, the leaders of Vila Limeira went to the CPT office in Pauini and the Sindicato de Trabalhadores Rurais de Pauini to find out what they needed to do to create the association. The sindicato ha d been visiting their community since the late 1980s, helping them to organize, and they learned from the union and CPT that first they would need at least twenty people to form the association. Since the agricultural
169 program was being managed by IDAM the Instituto de Desenvolvimento Agropecurio e Florestal Sustentvel do Estado do Amazonas and there was no IDAM office in Pauini at the time, they decided that they would have to make the long trip to Lbrea to obtain the information and assistance they nee ded to create their association. IDAM was a key contact for riverine communities when this research was conducted because of the agency's mandate of improving agricultural and livestock production in the state of Amazonas. IDAM was part of the Secretari a Estadual de Produo SEPROR and had its headquarters in Manaus and branch offices in all 62 of the municipalities in Amazonas state. The IDAM office in the city of Lbrea was one of two in the municipality, the other being located in Nova California in the southern part of Lbrea, which was accessible from Rondnia and Acre but effectively cut off from the municipal center due to the lack of roads. The IDAM office in Lbrea began as a state branch of the federal government agency known as ACAR, the Asso ciao de Crdito e Assistncia Rural which was created in the state of Amazonas in 1966. This organization worked in conjunction with the SUDHEVEA rubber agency office in Lbrea. The SUDHEVEA program was created in 1965 and ended in 1989. ACAR was ren amed EMATER and in 1977 the government of Amazonas created the state level EMATER, Empresa de Assistncia Tcnica e Extenso Rural do Estado do Amazonas an agency whose work began to reflect the desire of the state government to deemphasize rubber product ion and emphasize family agriculture and animal husbandry. Finally, in 1993 IDAM, which had been created in 1966 as part of EMATER, replaced EMATER and added social and environmental components to their extension work.
170 At the time of this research the I DAM office managed a series of key resources important to riverine communities. First, they were in charge of the PRONAF agricultural credit program, which provided low interest loans and subsidies for family farmers and other producers. The IDAM office also had a program to distribute seeds to family farmers, who often lack seeds in certain years due storage problems. The IDAM office provided extension for livestock production, including the vaccinations and pest control measures important in tropical a reas. Extension agents from IDAM conducted training workshops to help family farmers improve their production techniques so they could get higher prices for their products in agricultural markets. The extension agents and officials in IDAM were much more likely to work directly with community inhabitants than their predecessors in SUDHEVEA, who tended to work mostly with seringal owners. On April 10, 1998 community leaders arranged with the pastor who was working in Vila Limeira to take them in his boat to Lbrea so they could make contact with the IDAM office to obtain assistance in creating the association. In the 1990s many of the inhabitants of Vila Limeira had worked for MANASA, the company that had purchased the land that had been part of the Mari en Indigenous Post that SPI had established on the Seruini River in the 1920s. This timber company had bought lands in the area as part of a reforestation agreement they had with the government. They were required to reforest areas to compensate for the ir timber extraction elsewhere. On the Seruini they had decided to plant an enormous area with Brazil nut trees, and the inhabitants of Vila Limeira were often hired for this work. The company had installations on the Seruini River that included an Evang elical church, and the pastor began to periodically visit Vila
171 Limeira in the 1990s. In 1997 the majority of the inhabitants of Vila Limeira converted to Protestantism and eventually built their own Assembleia de Deu s chapel. Evangelical Church Networks Since 1970 there has been an explosion of Evangelical churches in Lbrea, and in Brazil in general (Chesnut, 1997; Stoll, 1990) These churches and the pastors that work in them represented yet another potential s ource of assistance that the inhabitants of riverine villages often turned to in times of need. At the time of this research there were at least 17 different denominations of Evangelical churches in the city. However, the biggest of these was the Assembl ia de Deus or Assembly of God. I learned in an interview with church leaders that the Assembly of God was the Evangelical church with the largest presence in Lbrea, with five chapels in the city and eleven more in the interior. Manaus had 530 Assembli a de Deus chapels and there were approximately 3,000 in the state of Amazonas. The evangelical churches had a reputation for being much more conservative in their approach to evangelism than the Catholic priests inspired by Liberation Theology who had bee n so instrumental in forming the CEBs and the rural workers' unions. The evangelical churches generally did not challenge social imbalances or look to change social hierarchies. However, pastors were often involved in other types of community development projects, and the pastor in Vila Limeira actively supported their attempts to form their community association. Ireland (1993) explains this by making a distinction between "church crentes" and "sect crentes." "Church crentes" were evangelicals who depended on the church hierarchy and depended on them to take action in their benefit, while "sect crentes" believed in individual faith and were more likely to take individual action without deferring to the church hierarchy (Ireland, 1993) The activities of the
172 pastors and their followers who actively challenged the status quo and engaged in social movement activities would seem to dispute the notion that evangelicals are not as likely as Catholics to engage in activities that address social imbalances; at least in Lbr ea there were a good number of evangelicals who fit into the "sect crente" category. With the help of their pastor, the inhabitants of Vila Limiera made the trip to Lbrea and went to the IDAM office to begin the process of formalizing their community ass ociation, the Associao dos Produtores Agro Extrativistas de Vila Limeira (APAVIL). The IDAM office, in addition to providing important resources like agricultural credit and seeds, also helped rural inhabitants with government paperwork. In the 1990s t he state government began to require that many official benefit programs be channeled through legally registered community associations. Although the idea of community organization dated back to the CEBs, now that it had become a government requirement to access resources, communities had more incentive to organize in this manner. The CPT office was also actively involved in helping communities to form these associations. The first community associations in Lbrea were legalized in the late 1990s, around the same time that they inhabitants of Vila Limeira were forming theirs. IDAM was actively helping other communities to form associations in the municipality, and at the time of this research there were approximately twenty community associations. Figur e 5 4 below shows the networks that were constructed by leaders in Vila Limeira to create their community association. Accessing Resources through Community Associations Once their association was created, the inhabitants of Vila Limeira received machine ry and equipment to plant and process sugar cane from the state government. The elder Oliveira had experience producing sugar based products, like rapadura and
173 cachaa from his youth in the Brazilian Northeast. Vila Limeira is one of the few communities in the area with significant amounts of terra firme uplands near their community and they rapidly cleared forest behind the community and began planting sugar cane. Their production expanded rapidly and they formed a relationship with buyers from Coca Col a for the sugar they produced. However, technical problems with their sugar mill and a lack of replacement parts shut down their sugar operation. I was informed that the director of IDAM also convinced them to participate in a coffee project in which th ey would take out government loans to buy seeds, implements, and fertilizers to produce coffee commercially. Unfortunately, this program was not based on economic or marketing studies and the inhabitants found themselves unable to sell their coffee crop a nd unable to pay back the loans for the program, making them ineligible for more government loans until these debts were paid. By 2009 the failure of these two agricultural programs led the people of Vila Limeira to once again rely on manioc flour and Bra zil nut production as their main sources of income. Their relationship with IDAM had resulted in their participation in workshops on how to improve the quality of their manioc flour, which did enable them to get better prices for it on the market. In spi te of the setbacks in the development projects they tried, the people of Vila Limeira had developed a community association to organize their activities, and their leaders formed a series of relationships with social movements and government agencies that helped them access resources to benefit their community.
174 At this point we leave Vila Limeira and pick up the story of two more communities, which are indicative of how riverine communities responded to the collapse of extractivism and the changing nature o f patron client relationships in Lbrea. Vila Vitria Along the Ituxi River, one of the major tributaries of the Purus, a similar story of community organization in response was playing out. The Ituxi River was also a major rubber producing area that ha d been explored by Colonel Labre, first as a potential route to the Madeira River and later because of the river's rich rubber resources. The Ituxi had a different resource base than the Purus, which has impacted the human occupation patterns over the yea rs. The Ituxi, a black water river, did not have the rich sediment load that made the beaches of the Purus so good for agriculture. The beaches of the Ituxi were also much smaller and exposed for shorter periods of time than those of the Purus. On the o ther hand the Ituxi had more areas of terra firme uplands closer to the main channel than the Purus. The Ituxi was known as an area rich in Brazil nuts, which had long been an important activity in the rainy season when rubber could not be tapped. The I tuxi also passed through various boom bust phases of resource use over the years in response to changing market demands. Although rubber was the mainstay of the inhabitants' production activities, they also combined rubber production with: providing firew ood for steamships, Brazil nuts, animal pelts, and fishing. In the 1970s there was a sorva boom on the Ituxi River and its tributaries that employed a number of the families that had once tapped rubber in the area. Sorva is the latex derived from the sorv a tree ( Couma sp. ) and was once used as a natural base for chewing gum. Amazonas state was the number one supplier of Brazilian sorva, and 1976 was Brazil's
175 peak year of production, with production reaching just over 6,000 tons (Coppen, 1995) For many years the sorva trees were felled in order to extract the latex, which depleted the resource base and forced sorveiros further and further into the forests to collect the latex. The Siriquiqui River, a tributary of the Ituxi, was the epicenter of sorva production, while the Cuerequete and Punicici Rivers, also tributaries of the Ituxi, were known for their Brazil nut groves. Near the end of the sorva boom cycle a new technique of using steel spurs and a rope to climb the sorva trees and score the bark using a special sorva tapping tool was developed to save the trees from destruction. The sorva business, like most of the extractive activities, began to collapse in the early 1990s and by 1992 Brazil's production had dropped to 500 tons (Coppen, 1995) By the time this research was conducted no one was producing sorva on the Ituxi. Copaiba oil was a produ ct that many families extracted on the Ituxi, but as previously mentioned the fluctuations in the market for this oil and difficulties in providing a quality product that merchants would buy complicated inhabitants' ability to continue to engage in this ac tivity. At the time this research was conducted, the inhabitants of the Ituxi were trying to eek out a living with Brazil nuts, manioc flour production, and timber extraction. Due to the lack of productive beaches, the inhabitants of the Ituxi were limit ed to the more labor intensive terra firme upland farming that required them to clear either secondary capoeira forests or primary forests in order to plant manioc. The Brazil nut groves were a contested resource, and due to their distance from the market s in Lbrea, implied high transportation costs. Timber extraction on the Ituxi was complicated by the extreme difficulties small scale loggers faced in accessing the government permits and
176 management plans necessary to engage in legal timber extraction. Most inhabitants were forced to log illegally, and ran the risk of being fined as they brought planks into Lbrea to sell to furniture makers. Interactions between patrons and clients appeared to have been a little different on the Ituxi than on the Purus The Ituxi did not have as many barraces lining its banks as the Purus. Instead, there was a large barraco on the Purus at the point where the Ituxi drains into the Purus, known as Boca do Ituxi. Then further up the river was a large barraco at the confluence of the Iquiry and Ituxi Rivers, known as the Fortaleza do Iquiry. This was once a large seringal and had rubber storage warehouses made of iron and cement that were still standing at the time this research was conducted. In between these two large barraces were several small barraces but for many years patrons that came periodically in boats supplied the tappers on the Ituxi in exchange for forest products. The collapse of extractive industries impacted the Ituxi much more than the Purus in terms of the out migration of the population. Inhabitants informed me that in the fifteen years before I began research there in 2008, between 50% and 75% percent of the population of the Ituxi had abandoned the area, moved to Lbrea, and then moved o n to Porto Velho and Manaus. In addition to the higher rate of out migration to the cities than the Purus, it was much more common for the inhabitants of the Ituxi to live most of the year in the city and return to the Ituxi for part of the year for the B razil nut harvest or to plant, tend, and harvest their manioc plots. Fishing became even more important for communities on the Ituxi in the wake of the collapse of rubber and sorva prices. However, the fishing resources on the river were not as productiv e as the Purus
177 and there were fewer lakes. The inhabitants of the Ituxi also had similar types of conflict over these resources: conflicts existed within communities, between communities, and between communities and fishermen from urban areas. Similar to the situation on the Purus, the uncertainty regarding resource ownership complicated these situations. The migration of some families to the city and their continued use of resources in the area put them into conflict with families that had stayed on the river. The use of Brazil nut groves and timber extraction faced the same complications. The extra distances from the city of Lbrea, and the smaller population, in many instances meant that the Ituxi received many fewer visits from social movement grou ps, government agencies, and religious groups than the Purus. For example, relatively sparsely populated tributaries like the Ituxi were often not included in projects like the CEB program, because the high cost of fuel meant that it made more sense to fo cus on the heavily populated Purus for these types of programs. Inhabitants told me that between 1980 and 2003 the Catholic Church had an extremely limited presence on the Ituxi. The collapse of extractive industries caused most of the patrons to abandon the area, and when this research was carried out even the river traders were visiting less and less, making it more and more difficult for the inhabitants to obtain needed supplies. The proliferation of rabeta motors had offset this problem to some degre e, enabling inhabitants to make the trip to town to buy and sell, but fuel costs were still high and their small canoes did not enable them to take large quantities of their products to sell. In the 1990s when this final collapse of extractivism was taki ng place there were still no schools on the Ituxi and very limited health facilities. Basic health posts existed
178 periodically during the mandates of certain mayors; even these had few medical supplies and no trained health care providers. Some of the fa milies interviewed on the Ituxi had inhabited their communities for at least three generations. One such community was Floresta, located near the confluence of the Ituxi and the Punicici Rivers. The history of this community is similar to many of the com munities in the interior of Lbrea. The families traced back to the arrival of men from the Northeast of Brazil that came as part of the various waves of migration to the area. In the case of Vila Limeira this occurred during the 1940s and in the case of Floresta the formation of the community dates back to the arrival of Barros Maciel in the 1910s. Barros Maciel came to the area drawn by the rubber industry like so many others, but he decided that growing sugarcane and producing cachaa was the best str ategy for establishing a presence along the Ituxi. In addition he actively explored the Brazil nut groves on the Punicici River and was involved in rubber tapping. Barros Maciel was involved in the process of incorporating local indigenous groups into th e extractive economy. Although the Ituxi had been explored and colonized in the 1870s Barros Maciel's descendents tell that there were still indigenous groups living in the area when he arrived. He is said to have helped pacify several of these groups al ong the Ituxi and Punicici Rivers in order to build a labor force for his activities. As in other parts of the Amazon the indigenous peoples of the Ituxi and tributaries like the Punicici did not fare well as a result of this contact, dying of diseases or being forcibly assimilated into the extractive industries. Barros Maciel's descendants engaged in changing livelihood strategies like so many other riverine communities, trading whatever products had market value for the
179 much needed supplies that had to be brought from urban centers. Rubber was the mainstay for many years, but they also tapped sorva, engaged in the pelt trade hunting alligators in the nearby lakes, fished, hunted, and collected Brazil nuts. With the collapse of extractivism in the 1990s the family began to move to Lbrea so that their children could get an education. However, after three generations on the land the Barros Maciel family felt a deep attachment to the land and resources and maintained a presence in Floresta. Similar to th e case of Vila Limeira, the inhabitants of Floresta did not have a land title that was considered valid by the government. Titles that had been created during the first rubber boom in many cases still existed but had been transferred, divided, and subdivi ded so many times that it was difficult to determine who owned what. In many cases a title had been divided or transferred to a new owner but had not been properly registered or the land taxes had not been paid. The inhabitants of Floresta, like most of the villages on the Ituxi, were left trying to depend on simple possession rights to defend their claims. Unlike the Purus, inhabitants of the Ituxi did not appear to have had to pay rent on their rubber trails to the same degree. Patron client relations hips certainly existed, but it appears that these relationships were mostly limited to the exchange of forest products for supplies and were less likely to include the payment of rent. In contrast to Vila Limeira on the Purus, the social organization tha t was in full bloom on the Ituxi when this research was conducted was due to the presence of an outsider that arrived in the area in 1995. Pastor Antonio Vasconcellos de Souza was himself born on the Purus and his family, like so many others, had arrived from Cear, drawn by the lure of the first rubber boom. Antonio grew up on a seringal on the Purus
180 and then moved to Lbrea when his family decided to leave the interior. Like so many others he found the transition difficult. He fished and worked other jobs while taking advantage of the educational opportunities available in Lbrea at the time. Antonio had been involved in the Catholic Church, but eventually decided to convert to evangelism, like so many inhabitants in Lbrea. He was invited to study t o become a pastor and began the theological training that would enable him to engage in this profession. Having completed his training, Pastor Antonio was assigned to work on the Ituxi River in 1995. He replaced another pastor who could no longer stand t he constant bouts of malaria that he was suffering on the Ituxi. Pastor Antonio was shocked by the disorganized nature of the communities he encountered on the Ituxi, where domestic violence and alcoholism were common. The total lack of education and hea lth facilities also alarmed him. Pastor Antonio felt that the best way to bring help to the people on the Ituxi was to engage in politics and run for the city council. In 1997 he decided to run for the city council, investing his savings in the campaign He was unaware of the necessity to have the official backing of a political party and by the time he realized this all the parties had given their support to other candidates. He had invested all his savings in the campaign and now had no hope of being elected. Similar to the case of Vila Limeira the word was in the air that the government was offering implements and projects for communities organized in associations. Pastor Antonio saw this as an opportunity to bring improvements to the Ituxi. He h ad written a letter describing the situation on the Ituxi and the hardships of the inhabitants but had no idea to whom he could send the letter. One evening he was asked to preach at his
181 church, the Assemblia de Deus and he decided to include the hardsh ips of the inhabitants of the Ituxi in his sermon. When he concluded, a man that he had not recognized approached him, interested in discussing his sermon. The man turned out to be the newly arrived director of IDAM, and was actively looking to help comm unities organize associations and access government benefits. Pastor Antonio gave him the letter and the IDAM director began to help him organize the communities on the Ituxi. Pastor Antonio gathered 32 inhabitants on the Ituxi, who became founding memb ers of APADRIT, the Associao dos Produtores Agro extrativistas Assemblia de Deus do Rio Ituxi Pastor Antonio moved to an old rubber tapping colocao on the Ituxi called Ba and renamed it Vila Vitria. Pastor Antonio also had a relationship with the mayor of Lbrea and was put in charge of an education program on the Ituxi that led to the building of several schools, at which Pastor Antonio taught. As in the case of Vila Limeira, Pastor Antonio actively sought out relationships with individuals in o ther social groups as a way of accessing needed resources for the communities of the Ituxi. He used these bridging ties, like the one with the director of IDAM, to access information and other resources that the inhabitants of the Ituxi did not have withi n their kin and community networks. He was also organizing the communities of the Ituxi to form the association. This is an example of bonding ties. Pastor Antonio became a key broker between the community members and the government through this relatio nship with the director of IDAM. APADRIT was legalized in 1997, and as a result of their participation in the association, several communities on the Ituxi received farming implements and motors from the government. The director of IDAM also convinced t hem to participate in the
182 coffee project, which was even more of a disaster than on the Purus. The participants on the Ituxi did not receive proper instruction about applying the chemical fertilizers that had to be purchased with the coffee project, and m ost of the coffee died. The families that participated on the Ituxi, like those in Vila Limeira were left in debt as a result of this program. Problems and Challenges The families on the Ituxi faced similar problems to those of the Purus in the 1990s: a collapsing extractive economy, the difficult transition to agriculture and fishing, and conflicts over key resources. Several indigenous areas were created at the mouth of the Ituxi, but the pressure near Floresta and Vila Vitria was not as acute as on the Purus around Vila Limeira. In 1997 the director of IDAM first mentioned to Pastor Antonio the idea of creating an extractive reserve. They had several meetings in which the benefits of the extractive reserve model were explained. In these meetings i t was explained that in the same way that organizing and creating the community association was a way to facilitate the arrival of agricultural assistance, the RESEX was another way to organize and bring basic services to the area. In 1997 the IDAM direct or helped Pastor Antonio write a letter requesting the creation of an extractive reserve on the Ituxi. This letter may not have found the right office or the right individual, but apparently Pastor Antonio received no response regarding his request. Aro und the time that Pastor Antonio and several communities on the Ituxi were forming their association and making connections to try to bring basic services to the Ituxi, a new threat was growing that would eventually put their continued existence on the riv er in jeopardy. The headwaters of the Ituxi and its tributaries come from Acre and from the southern part of the municipality of Lbrea, which is precisely the area that in
183 the 1990s was rapidly being colonized by ranchers and land speculators from Easter n Acre and Rondnia. Many of these grileiros bought the posse rights of riverine inhabitants on the upper courses of the tributaries of the Ituxi, and rapidly colonized these areas in the late 1990 and 2000s. This unplanned and often illegal colonization led to conflict because many of the most productive Brazil nut groves in the Ituxi watershed were on the uplands near the headwaters and all along the courses of the Punicici and Cuerequet Rivers. Brazil nut collection was one of the last viable economi c alternatives for the inhabitants of the Ituxi; even former inhabitants that had moved to Lbrea continued to depend on this activity as a key source of income, and the destruction of these groves or loss of access to them represented a major threat to th e inhabitants of the area. If we compare the Vila Limeira with the two communities on the Ituxi, Floresta and Vila Vitria, we see that they were facing common problems. Collapse of extractivism, difficult transition to agriculture and fishing, land ten ure uncertainty, internal and external conflict over key resources, a history of patron client relationships, exclusion from formal credit, problems of seasonal resource production, geographic isolation, and a general lack of basic services. In both case s we see the importance of their connections and contacts with outside groups in the process of trying to overcome these problems. Organizations associated with the Catholic Church and later government agencies and evangelical church movements played key roles in helping the communities to organize and reach out to obtain resources that they did not have access to within their kin and community networks. These early contacts helped them get organized into associations and
184 helped them access credits, agric ultural improvement programs, and some basic services. However, the land tenure problem had not been resolved and the resource conflicts continued in spite of attempts to do the Lake Protection and Environment Volunteers programs. In the late 1990s the i nhabitants of Vila Limiera, Floresta, and Vila Vitria began the process of demanding extractive reserves in the effort to resolve their land tenure problems, economic woes, and lack of access to basic services. The process in which they engaged reveals a gain how the patronage pyramids established during the rubber boom were increasingly replaced by multi hub social structures in which government agencies, religious organizations, and social movements played increasingly important roles as brokers and inte rmediaries. Extractive Reserves By the late 1990s the inhabitants of Vila Limeira had stopped paying rent to the old patrons, had formed their community association, and had obtained implements in the effort to make the transition from extractivism to a c ombination of agriculture and extractivism. But they still faced extreme uncertainty due to the land tenure problems they faced. Indigenous reserves had been created all around them and they were being threatened with the creation of an indigenous area t hat would push them off their land. Their lack of land title also prevented them from defending the resources they depended on from fishermen, loggers, and other resource users from outside the community. In addition, the community was lacking even the mo st basic health services and education. They decided to once again seek information and assistance from their network of contacts and continue to build this network to try to solve their problems.
185 Vila Limeira and the M dio Purus Extractive Reserve By th e end of the 1990s the inhabitants of Vila Limeira heard rumors that another Indigenous Land was going to be created on the Seruini River, and they felt that they were at a critical point and would have to either resolve their land tenure problem or be pus hed off the land. Their first step was to turn to the director of IDAM who had helped them so much with the formation of their community association. In 1999 he suggested that they speak to INCRA, the federal agency in charge of agrarian reform. The lea ders of Vila Limeira went to the INCRA office and began to inquire about the possibilities of legalizing their land claims. They were informed that it would not be possible to create individual plots because there were private lands in the area with legit imate titles. A large parcel of land known as the Gleba Mapu had a legitimate title that would have to be dealt with. This parcel of land was part of the old Seringal Meteripu that had been sold when Edmar Castelo Branco, the last patron that claimed t o own the area, moved to Manaus. The new owners never came to claim the land or work it, but the INCRA office said there was a valid title for the land that had to be respected. The second option that INCRA had was to create an assentamento or agrarian r eform settlement in which each family would receive a lot of a certain size. The inhabitants of Vila Limeira began the bureaucratic process of officially requesting an assentamento The bureaucratic process was extremely slow and the leaders in Vila Lime ira decided that they could not wait and should look into options to resolve their land tenure problem. In 2000 one of the leaders of Vila Limeira heard that a meeting was going to be held in Pauini about extractive reserves, another land reform program that could be a solution to the community's problems. This meeting included representatives of the National Council of Rubber Tappers, the Ministrio do Meio Ambiente and the Workers'
186 Union of Pauini. These representatives explained the extractive reser ve concept to the leaders and began to respond to questions about the steps the community would need to take to begin the process of creating an extractive reserve. On August 17, 2000 they sent a letter to the Coordenadoria de Agroextrativismo to request more information about the steps they would need to take request a reserve, and one month later on September 18, 2000 they sent their first official letter to CNPT in Braslia in the name of their community association, requesting that a study be done to d etermine the possibilities of creating an extractive reserve. They turned to their contact in IDAM to help them draft the official letters and paperwork petitioning the creation of the extractive reserve. At the same the IDAM office in Lbrea was helping Pastor Antonio and the inhabitants of the Ituxi to request an extractive reserve in the name of their association, they were helping another community on the Purus, Vila Acim, to send a request in the name of their community association. The inhabitants of Vila Acim were also organizing their activities with the help of a Pastor from the Assemblia de Deus These letters were sent separately, each community requesting its own reserve. The inhabitants of Vila Limeira requested that the 23,000 hectare Gleba Mapu be expropriated and turned into an extractive reserve to benefit the 19 families in the community. On September 19, 2000 they received a letter informing them that the IBAMA office in Braslia had sent a letter to the CNPT office in Manaus req uesting the completion of a study to determine if the conditions existed to create an extractive reserve in the area. The CNPT office in Manaus informed the inhabitants that the number of families that would be benefited within the Gleba Mapu was insuffi cient to
187 warrant the creation of an extractive reserve. So they decided to add the neighboring communities of Vila Ded (20 families), Praia da Conceio (15 families), Meteripu (30 families), and Santa F (70 families). Eventually the leaders were info rmed that the extractive reserve would have to benefit at least 500 families in order to be created, forcing them to expand and include yet more communities. These neighboring communities in some cases included family and friends, but in all cases the inh abitants faced the same types of threats: creation of indigenous lands, the same lack of basic services, land tenure and resource insecurity, conflicts with outside resource users, and a history of being enmeshed in patron client relationships in their pro duction activities. The grouping of these communities is an example of "bonding ties" "expressive actions" in which people of the same social group come together to defend resources (Lin, 2001b) However, it is interesting to note that this "bonding activity" was done in response to the demands of the government agencies for creating the reserve, and not solely based on the initiative of the c ommunities or their leaders. These types of horizontal linkages are more common in the multi hub, broker dominated social structures that replace patronage pyramids (Martin, 2009) I learned that in 2001 the coordinator of the CNPT office in Manaus visited Lbrea to conduct a preliminary study an d mapping exercise to determine the possibilities of creating the extractive reserve. This visit included meetings designed to involve more communities in the process of creating the proposed extractive reserve. In 2001 CPT became directly involved in th e process of requesting the reserves in Lbrea. In a regional meeting, CPT members from Lbrea heard from CPT members in other parts of the state about Extractive Reserves and how they could be used as a way to protect
188 traditional peoples, stabilize land tenure claims, and mediate conflict. The Lake Preservation and Volunteer Environmental Agents programs had very modest impacts on the conflicts that were taking place, and CPT staff began to suggest Extractive Reserves as a solution when community leaders came to them for help in resolving resource conflicts. CPT leaders had heard about the extractive reserve model through the same network of contacts that had passed on the word about the Lake Preservation and Volunteer Environmental Agents programs. The y held a meeting in 2001 in which they brought in experts to inform more communities about the extractive reserves and how they worked. After this meeting the original three extractive reserve requests were combined into one large request that would inclu de a huge area encompassing much of the Purus between Pauini and Lbrea as well as tributaries like the Rio Cana. This combined extractive reserve request totaled 1,450,000 ha. What had begun as two individual community requests (Vila Limeira and Vila Acim) for extractive reserves on the Purus had expanded to encompass more than 90 communities, while on the Ituxi the request submitted by Vila Vitria had expanded to include another 11 communities. Community members were forming "bridging ties" to orga nizations outside their communities to try to resolve their local problems at the same time they were forming "bonding ties" between communities in response to government guidelines. This process required first a set of "bonding ties" at the community lev el, forming the associations, but then as the size of the area for the proposed extractive reserve increased it required "bonding ties" between more and more communities that had similar general problems but often very specific needs and internal social dy namics. This is another example of the horizontal networks that often
189 form when patronage triangles change into multi hub broker driven networks (Martin, 2009) One problem that arose with the expansion of the extractive reserve boundaries was the dissemination of information to all of the commun ities that had been included in the reserve. In most cases, like the Encontro dos Ribeirinhos community leaders were invited to participate in the event in Lbrea and were expected to return to their community and relay the concept of the reserves, which is often presented in complicated technical jargon, to the other members of their communities. This led to a great deal of confusion about what the extractive reserve would mean in the daily lives of the inhabitants. This was complicated by the fact tha t individuals who stood to lose from the creation of the extractive reserves had begun to tell people that the reserves would be a national park that they could no longer live in or that they would be expelled and the land would be sold to foreigners. I l earned that at the same time that they were actively pursuing the extractive reserves, the inhabitants of Vila Limeira still had their official request for an INCRA settlement pending. In 2001 INCRA said it would be possible to create an assentamento but the INCRA office in Lbrea was in a transition phase and did not have a coordinator, so the request was shelved. In 2002 the new coordinator in Lbrea reactivated the request. In 2004 Z Maria went to the INCRA superintendent in Manaus to talk about the ir request, and in May June 2005 they sent technicians to do a study and census of the families in Vila Limeira and Vila Ded, which were the main communities in the request. Both communities were at least partially inside the area titled to the owners of the old Seringal Materipu. INCRA's policy for creating
190 assentamentos on private lands was to wait until the expropriation process was complete. The inhabitants of Vila Limeira knew that this could potentially take years so they decided to focus on the extractive reserve request instead. The extractive reserve legislation, on the other hand, did not require that land titles be expropriated before the area was created. The government could create the extractive reserve first and then attempt to indemnif y owners with legitimate titles later. This is an example of the fluid and ever changing nature of the networks of the community leaders in Vila Limeira. First they had gone to an initial group of government agencies and social movements to obtain infor mation about forming their association in order to access government resources. Next they went to INCRA to find out about creating an assentamento and to IBAMA to find out about creating an extractive reserve. They were essentially using whichever broker s and intermediaries would bring the desired result of resolving their land tenure problems. When the assentamento idea proved too complicated, they emphasized their interactions with IBAMA and focused on creating the extractive reserve. The leaders in Vila Limeira, Floresta, and Vila Vitria also had to adapt their networks to the changing realities within the organizations they were working with. In 2002 the coordinator of CNPT/IBAMA in Manaus changed positions and was replaced. The new coordinator w ent to Lbrea to continue the process of creating the extractive reserves in 2003. In 2003 this new coordinator took another job and was replaced. The changes in IBAMA staff caused the extractive reserve proposal to be shelved and forced the leaders in t he communities to seek out relationships with the newly appointed coordinators.
191 During this phase an association was created that was supposed to represent inhabitants of the Purus and Ituxi Rivers, called the Associao dos Produtores da Reserva Extrativ ista de Lbrea (APREMP). This was a provisionary association that they set up to do the petition that never existed in fact. In 2003 2005 CPT and the other social movement organizations, together with IBAMA organized a leadership course for the community leaders, as a way of once again passing on information about the extractive reserve concept to community inhabitants. In addition to the changes in IBAMA personnel, the community leaders had to adapt to the changing landscape of social movement organizati ons and government agencies in Lbrea. In 2003 the IDAM director that had been so active in helping the communities to form their associations and complete the paperwork for the extractive reserves was replaced, and the new director had a different focus for the agency. The IBAMA officials in Lbrea also changed periodically, forcing the community leaders to build new relationships. In 2003 the IBAMA office in Lbrea began to actively participate in petitioning the creation of the extractive reserves. A round the same time a Conselho Nacional dos Seringueiros (CNS) office was opened in Lbrea as well as a Grupo de Trabalho Amazonico (GTA) office. CNS, the official rubber tappers' social movement, grew out of the rubber tapper organization in the 1970s an d 1980s in the nearby state of Acre (Allegretti, 2002) The extractive reserve model grew out of interactions between ru bber tapper leaders like Chico Mendes, government officials, Brazilian academics, and international environmental organizations (Allegretti, 2002) By 2009 the extractive reserve model was the solution that CNS promoted and championed as a solution to the resource conflicts, land tenure problems, and lack of
192 basic services facing extractivist communities throughout the Amaz on. The model had even spread to areas along the coast, far from the Amazon realities that had generated it. GTA was created in 1990 as part of the PPG 7 program to serve as a liaison between community associations in the Amazon and government agencies. GTA links the associations together at the local level and is supposed to help them access funding opportunities with donors and government agencies. GTA was also actively championing the creation of extractive reserves as a solution to the types of com munity problems facing Vila Limeira, Floresta, and Vila Vitria. In 2004 IBAMA conducted socioeconomic studies along the Ituxi and Purus Rivers to determine which natural resources the communities within the proposed extractive reserves depended upon. On e of the results of the socioeconomic studies was the realization that the resource use patterns on the Purus and Ituxi Rivers were different, and it would be extremely difficult to manage an extractive reserve that covered such a huge area. In August 200 5 the decision was made to separate the proposal into two, one for the Mdio Purus Extractive Reserve and one for the Ituxi Extractive Reserve. In addition to dividing the proposal into two, at this time several large portions were removed from the origina l limits because they overlapped with the area of the proposed Iquiri National Forest and a study area for another Indigenous area on the Mamoria River. Finally, a third section that had been included in the original proposal was removed after a participa tory mapping meeting with groups in Lbrea that would potentially be impacted by the creation of the reserves. The Colonia de Pescadores in
193 Lbrea requested that a portion of the reserve close to Lbrea be removed so that it would not interfere with their commercial fishing activities. Between 2005 and 2007 there were several meetings in which community leaders, social movement coordinators, and government officials discussed the limits of the extractive reserves. In 2005 the Ituxi Extractive Reserve was enlarged to include Brazil nut groves upon which families along the river depended. In addition to the meetings involving the community leaders, government officials working on the petitions met with each other and with supervisors in Braslia. These officials then had the challenge of relaying decisions made in Braslia and Manaus to individual community inhabitants in the reserves. I learned that the creation of the extractive reserves had to pass through the government bureaucracy up to the Casa Civil, the president of Brazil's cabinet of ministries. Negotiations and meetings were held in which the different government ministries gave their input on the creation of the reserve. The ministries of mining and energy and the ministry of transportati on questioned the creation of the reserve; the Ituxi River has several waterfalls that were considered to have hydroelectric potential. The mayor of Lbrea at the time did not support the reserve, and campaigned within his political network to have the pe tition scrapped. Finally, on May 8, 2008, President Lula Igncio da Silva signed the decree creating the Mdio Purus Extractive Reserve, which had began years earlier in Vila Limeira and now included more than 90 communities. The potential of hydroelectr ic generation on the Ituxi River delayed the creation of the Ituxi Extractive Reserve until June 2008.
194 Network Interactions Government agencies with an interest in creating extractive reserves and other sustainable development reserves reached out to soc ial movement groups at regional and local levels. These groups had closer, more daily relationships to the local communities that actually lived in the areas in question. Typically the people who worked in the government agencies were based in cities lik e Manaus and Braslia, although there were often a couple of people working at the local level also. But even the government people working at the local level often turned to the social movement groups for assistance in working with the communities. In L brea locals who were either from the communities or had very strong relationships with the people who lived there staffed many of the social movement groups. They could "speak the language" of the people living in the communities and therefore they fille d an important role as intermediaries or "brokers" between government agencies and the people in the communities. The people who worked in the local social movement groups lived almost exclusively in Lbrea. The people in the local social movement groups had strong relationships and relatively constant contact with community leaders from the interior. The community leaders acted as "brokers" between the social movements and government agencies, and the community inhabitants. They attended all the meetin gs and workshops, and were supposed to return to the communities and inform the others about what had happened. Social Movement Networks Community leaders filled a broker role not only for the community inhabitants, but also for government officials and social movement leaders who wanted to pass information to community inhabitants. The government officials and social movement
195 leaders also had goals and objectives and their relationships with community leaders helped them reach out to a larger mass of in habitants. Being a broker confers a certain degree of responsibility and power. Leaders controlled information and were conduits that enabled inhabitants to access resources outside their communities. However, with this responsibility and power can com e great risk. Powerful forces wanted access to resources in the reserves and they also knew who community leaders and social movement actors were, and they realized that eliminating these leaders could disrupt these activities. This meant that while bein g a community leader conferred prestige and power, it could also bring great risk. In addition, community leaders often had to travel to meetings in town and had less time to be with their families, plant their crops, and engage in other economic activiti es. Traditional Patron Client Networks vs. Community Leader Networks As mentioned in Chapter 3, traditional patron client relationships in Lbrea were based on the control of key resources and revolved around a series of exchange expectations between pat rons and clients. In 2008 2009 riverine inhabitants continued to engage in modified versions of these relationships, particularly in their access to markets. River traders and storeowners in Lbrea had replaced the seringal owners and these "new" patrons had less control of the resource base than the seringal owners had once maintained. They had less control over their clients, who often engaged in relationships with more than one patron. However, these patrons typically offered less in the way of servi ces, less credit, shorter periods to repay loans, less variety in the kinds of supplies, etc. With the collapse of the "traditional" patron networks, community inhabitants had formed multi hub social networks with newly arrived social movement organizatio ns, government agencies, and other actors.
196 These new relationships had mutually beneficial tendencies, and the influence of the two groups depended upon where they were interacting. For example, government officials and social movement coordinators defe rred to community leaders when they were visiting the communities. In these instances the community leaders had more influence because: they knew the area, the officials needed them as "guides," and the community leaders could vouch for them and validate their presence in the community. Officials also realized that community leaders were important multipliers of information in the community and were therefore the best available conduit to spread their messages to as many community inhabitants as possible. The reverse was also true in these relationships; in urban settings the influence of the social movement coordinators and government officials was more pronounced. Their positions enabled them to access resources that the community leaders did not have at their disposal. They were able to plug community leaders into their own institutional networks and introduce them to people they would not otherwise meet. In essence the social movement coordinators and government officials became "guides" for commun ity leaders, shepherding them through the bureaucracies in much the same way that the community leaders had been their "guides" in the communities. Chapter Summary The case studies in this chapter show how social structures in Lbrea, based on the patron client relationships that were institutionalized during the rubber boom, changed dramatically between the 1940s and the 2000s. As we can see in Figure 5 3 4 below, supplies once came to the seringal through the owner's contacts with merchant 4 This figu re was inspired by the patronage pyramid figures in (Martin, 2009)
197 houses, who in turn often had ties to large banks. This is an example of a patronage pyramid, in which a patron uses his influence to maintain control of a series of clients (Martin, 2009) Figure 5 3 Seringal Patronage Triangle Seringal owner patrons were able to keep their tapper clients by providi ng supplies on credit on an exclusive basis. In the event that the tapper sold rubber to someone else, such as a river trader, the patron threatened to invoke sanctions, such as restricting credit or evicting the tapper and his family from the seringal Priests, government agents, politicians, and others respected the seringal hierarchy and would seek out the patro upon arriving at the seringal Economic collapse, particularly the decline of the rubber industry, made it difficult for seringal owners to provide supplies on credit and they were unable to maintain their land tenure claims. Patrons could no longer maintain exclusive exchange with their clients because they could not provide credit and could not threaten to evict them from
198 the land. They w ere no longer able to enforce the "one patron" rule, which is considered a key element of maintaining patronage pyramids (Martin, 2009) A new multi hub structure began to emerge in which community inhabitants began to seek out a myriad of actors to obtain resources and services. Figure 5 4 belo w shows my impression of how social networks were changing into more multi hub structures in which community leaders began to assume key roles as brokers between the village inhabitants and the outside world. Figure 5 4 Post Seringal Multi Hub Network As Martin (2009) has pointed out, although a multi hub network may be more egalitarian, it may take more effort to access resources through many patrons than through one Finding one patron who can provide the same resources and services may be more efficient, which may e xplain why patronage pyramids can persist or resurface (Martin, 2009) Therefore, although the patronage associated with the
199 seringal system appears to have disappeared, patronage in the area has not disappeared and new patronage pyramids form. In Chapter 6 I present the results of a survey de signed to determine which of the "new" actors were providing resources and what the perceptions of riverine inhabitants were regarding their propensity to provide assistance. The results of this survey show how riverine inhabitants' networks changed to mu lti hub networks in which they no longer depended on a single patron.
200 CHAPTER 6 NEW SOCIAL NETWORKS AND THE DISTRIBUTION OF SOCIAL CAPITAL The community we are discussing is a poor one in which only limited resources are produced locally. The 'extras' that make life more satisfying, as we shall see, are obtained from outside the local community. The means of obtaining them are the exchanges that are the basis of relationships of patronage and clientage (Greenfield, 1979) Although the quote above describes a community in a rural area of the state of Minas Gerais in the 197 0s, in many ways it is a perfect description of the situation of isolated villages along the Purus and Ituxi Rivers in the municipality of Lbrea in 2008 2009. In these villages, "traditional" patrons had once supplied the "extras" from outside the commun ity as part of the patron client exchanges upon which the aviamento credit system was based. Beginning in the 1970s and 1980s economic changes made it increasingly difficult for "traditional" patrons to supply the "extras" and "new" actors began to arrive providing new channels through which vital resources could be accessed. My attempts to determine how social networks in Lbrea had changed over time, and how these changes had impacted the ability of riverine inhabitants to access resources and services were inspired and guided by the following questions: 1. Who were the actors that made up the "new" social networks of riverine inhabitants in Lbrea? 2. What were the resources that the actors in these "new" networks had at their disposal? 3. How likely were the actors to provide riverine inhabitants with access to these resources? 4. What characteristics best explain the ability of some riverine to more successfully use their social networks than others? 5. Which of the providers of resources did the inhabitants cons ider to be "patrons?"
201 As described in the previous chapter, in villages in the interior of Lbrea some inhabitants were more successful than others at obtaining the "extras" through the "new" networks that arose in the wake of the changes in the "tradition al" patron client system. In Chapter 5 I presented information about how community leaders used their social networking and social capital building abilities to try to overcome a series of challenges and hardships. These leaders were able in some instanc es to use their social capital to bring "extras" into their communities and thus strengthen their own position within the community. I decided to test my observations about the changes in riverine inhabitants' social networks by collecting network data fr om a sample of inhabitants and running statistical tests to determine: 1) whether or not network variability was associated with having more information about extractive reserves and / or having more wealth, and 2) which personal characteristics best predi cted social network extensity among riverine inhabitants. In this chapter I describe the methods I used to collect network data, the statistical tests that I ran to test my hypotheses, and the results of these analyses. In addition to measuring responde nts' social capital by quantifying who they knew, and what resources they accessed from people in their networks, this survey was designed to determine what kinds of social networks (if any) had replaced the "traditional" patron dominated social networks d escribed in Chapter 2. Social Capital and Patronage Social capital has been described in several ways. The definition used in this dissertation equates social capital with the resources embedded in social relations and social networks (Bour dieu, 1986; Lin, 1982, 2001b) People use social capital to: (1) access economic resources (loans, investment tips, protected markets), (2) increase
202 their cultural and human capital through contact with experts, and (3) obtain valued credentials through affiliations with institutions (Po rtes, 1998) An important distinction that can be made in understanding a person's social capital is between "access" and "use." "Access" refers to the resources that are available in their social network, and "use" refers to the specific actions they en gage in to mobilize those resources and achieve objectives (Lin, 2001a; 2 001b) Lin (2001b) suggested that these actions to mobilize social capital can be divided into two types : 1) instrumental actions (gaining resources); and 2) expressive actions (maintaining resources). Finding a better job or house are examples of instrumental actions while personal support and the sharing of sentiments are examples of expressive actions (Lin, 2001b) When analyzing the individuals in a network and the resources they control, there are three important elements to consider: 1) th e number of individuals in the network who are prepared to help, b) the extent to which they are prepared to help (the strength of the tie), and 3) what accessible resources they control (Flap & Boxman, 2001) Two key social capital concepts underlie my analysis of the "new" social networks that have replaced "traditional" pa tron client relationships in Lbrea: (1) network tie strength and extensity, and (2) embeddedness. Tie Strength and Extensity Granovetter (1973; 1983) hypothesized that because strong ties require time to build, and usually i mply that people are similar, weak ties must therefore link people with fewer similarities, such as those of different social classes. These ties across social strata, called "bridging ties," are more likely to enable people to access information and reso urces that their strong ties would not be able to provide (Granovetter, 1973; 1983)
203 The relationships between "traditional" patrons and rubber tapper "clients" may have fluctuated between "strong" and "weak" depending on the circumstances. However, as pointed out in the tables in Chapter 3, the patrons controlled access to credit, markets, and often maintained control of the resource base. Rubber tappers and other extractivists depended on patrons who were wealthier, better connected, and had more access to vital resources. Clients and patrons may have had many similarities such as their place of origin (Northeastern Brazil), and religious beliefs, but they were differentiated by their social position and access to and control of resource s. I consider many of the "new" relationships that have emerged in the collapse of "traditional" patron client relationships to be examples of "weak" ties. Individuals in the social movement and government agencies with whom riverine people began to inte ract had much higher levels of education, were wealthier, and had much greater access to information and other key resources than the inhabitants of the riverine villages. This is an example of how "weak," "bridging" ties to individuals in other social gr oups can facilitate instrumental actions that enable people to obtain resources that are unavailable in their own social group. Lin has hypothesized that the "extensity" of network ties may be more important than their strength (Lin, 1999) Among isolated Amazonian peasants in Lbrea, the extensity of persona l networks ties was expected to be one of the most important dimensions distinguishing the access that individuals had to social capital. In addition to the "traditional" patrons who formerly dominated the social interactions of the inhabitants of riverine communities, individuals now had the opportunity to engage in "new" relationships with individuals from expanding
204 government agencies, social movement groups, religious organizations, and other groups. In fact, one possible outcome when a system based on highly multiplex patron client relationships begins to change is that clients will have to develop a more diffuse network of relationships with a greater number of patrons to obtain the resources previously provided by a single patron (Boissevain, 1977; C. Escobar, 1994; J. C. Scott & Kerkvliet, 1973; Theobald, 1983, 1992) Clients may have been in a position in which they had to increase the "extensity" of their networks in order to access the resources that had p reviously been provided by one patron. If the "extensity" of network ties was indeed more important than the "strength" of the ties, then riverine inhabitants with more ties to these "new" groups and organizations should have had more access to informatio n, wealth, and other resources. Embeddedness The concept of embeddedness is used to show where a person or resource is located within a network. In his work on "structural holes," Burt (1992; 2004) focused on the location of individuals within a network as a key indicator of social capital. Building on Granovetter's work on network ties, Burt (1992; 2004) described how people who fill "structural holes" link isolated social networks and serve as "bridging ties" that facilitate the exchange of information and resources between networks. A second application of the embeddedness concept is to focus on the location of resources within a network. Social resource theory points to wealth, power, and status as the most valued resources in many societ ies (Lin, 1982) Therefore, social capital can be measured by the amount or variety of these resources that people with whom one has network ties have at their disposal (Lin, 2001a) The variety and location of resources embedded in the personal networks of isolated riverine inhabitants of the
205 Brazilian Amazon indicate wha t kinds of social capital they have and how they use it. It is important to consider that, just because resources are embedded in an individual's network, does not mean that they will have uniform access to them. It is important to have an indicator that measures the availability of the different resources (van der Gaag & Snijders, 2005) Societies vary in how resources are distributed in stratification hierarchies, in how social networks connect different parts of hierarchies, and in the cultural and political rules governing how networks may be used to access resources" (Lin & Erickson, 2008) I would add that within societies, the political rules governing how networks may be used to access resources, and who controls these resources, change over time. One of the keys to understanding how resource control and patron client relationships had changed in Lbrea was to identify not only what kind of "new" social networks had replaced "traditional" patron client ties, but also to determine: 1) what r esources were embedded in these "new" networks, and 2) which "new" relationships the inhabitants of riverine villages depended upon to access these resources. Over the past few years several research methods have been developed to measure social capital and social networks. In order to answer the questions presented at the beginning of this chapter, I employed the position generator and resource generator data collection methods. I used a modified version of the position generator method to determine wi th what groups and organizations riverine communities had contact, the frequency of the contact, and the resources they accessed through these relationships. The flexibility of the position generator method enabled me to include questions about respondent s' understanding of the word "patron". I used a modified
206 version of the resource generator method with a small sample of individuals to further explore the relationships that riverine inhabitants used to access a specific set of resources and to determine which of the providers of these resources they considered to be "patrons." Position Generators In order to measure and analyze the relationships that a sample of riverine community inhabitants had with groups such as social movement organizations, gover nment agencies, religious entities, politicians, and merchants, I decided to use a modified version of the position generator data collection method (Lin & Dumin, 1986) The position generator method consists of asking a respondent a s eries of questions about "a sample of structural positions that are salient in a society" and their contact with individuals in those positions (Lin, 2001b) There are several ways to measure the respondents' social capital based on their answers to these questions. The first method is based on the "prestige" of the different occupations in the society. It is assumed that people in occupatio ns of higher prestige have more access to instrumentally useful resources (income, education, authority) than individuals in occupations of lower prestige (Lin & Erickson, 2008) However, people in a network that are not engaged in occupations of high prestige can offer valuable social capital for expressive actions: attention, emotional support, love, and other resources contained in their human and social capital (van der Gaag, Snijders, & Flap, 2008) This means that the position generator method has proven most useful for measuring instru mental actions of social capital, while other methods are better suited for measuring expressive actions of social capital (van der Gaag et al., 2008)
207 In a position generator interview respondents are asked questions about their relationships with individuals in a series of occupations; indices of occupational status rankings are used to weight the different occupa tions (ex. lawyers have a higher ranking than cashiers); the respondents' interactions with individuals in the different occupations are then added up; and the weighted scale provides a final score. A higher score indicates that the respondent has more in teractions with individuals engaged in occupations of high socioeconomic status and/or prestige. The method is based on the premise that individuals in occupations of higher socioeconomic status and prestige have more access to instrumental resources, and therefore the more of these individuals with whom a person interacts, the more resources they have at their disposal (Lin & Erickson, 2008) If social capital is a measure of re sources embedded in social networks, then the more people engaged in high status occupations with whom you interact, the more social capital you have at your disposal. There are several indices used to measure occupational status and prestige. The Inter national Socioeconomic Index (ISEI) is a measure of socioeconomic status and there are several measures of occupational prestige, depending on the country in question. The ISEI provides a ranking of occupations based on human resources and economic reward s, while occupational prestige rankings may better reflect a range of social rewards associated with certain occupations (Ganzeboom & Treiman, 2003; van der Gaag et al., 2008) Position Generator Measurement Seve ral deductive measures are used to determine position generator scores: 1) highest accessed prestige ; 2) range in accessed prestige ; and 3) number of different positions accessed (van der Gaag et al., 2008) Highest accessed prestige is a
208 measure of the scores of the highest ranked occupations in an individual's network; range in accessed prestige is the difference b etween the highest and lowest accessed prestige; and the number of different positions accessed is the total number of occupations in which the respondent knows someone (van der Gaag et al., 2008) Highest accessed prestige is based on the notion that the more people in high status occupations you know, the more access to resources, and thus social capital, you have at your disposal (Lin, 2001b; van der Gaag et al., 2008) Range in accessed prestige and number of different positions accessed are based on the notion that the more differentiation in relationships and resources in a network the more social capital is available (van der Gaag et al., 2008) This is a measure based on the idea that the "extensity" of the network determines the amount of access to resour ces, and thus social capital, an individual has at their disposal. Average accessed prestige and total accessed prestige are two additional measures of position generator scores (van der Gaag et al., 2008) There are also measurements for position generators developed using a more inductive approach. Using the Position Generator The nature of my research site extr emely isolated riverine villages in the Brazilian Amazon as well as the nature of the livelihoods of the inhabitants of these villages led me to adapt the position generator method to this distinct set of circumstances. Measures of occupational prestige developed for other locations simply would not have provided an accurate measure of the importance of the different actors in the social networks of riverine inhabitants in remote parts of the Amazon. Enns, et al. (2008) also develope d a modified position generator questionnaire when conducting social capital research in remote communities in British Columbia, Canada. They
209 decided that using a preformulated list of professions would not accurately measure social capital in communities where groups and organizations specific to the local context (like indigenous leaders and tribal organizations) could potentially represent important contacts within inhabitants' social networks (Enns et al., 2008) Like Enns and her colleagues, I de cided that the best way to accurately determine what inhabitants' networks looked like was to inductively build a questionnaire based on the groups and organizations that existed in Lbrea at the time the research was conducted. Producing the Position Gene rator Alter List The history of the creation of the two extractive reserves presented in Chapter 5 reveals that in spite of the relatively small size and remote location of the town of Lbrea, there were many potential contacts within the social networks o f riverine inhabitants living in the municipality. I combined my interest in the changing nature of patron client ties, with analysis of the interactions implied by the creation of extractive reserves, to develop the list of possible contacts to include i n my questionnaires. During preliminary research I received a series of reports and other documents associated with the process of creating the reserves in Lbrea from a government official who was involved in creating the reserves. Among these documents were letters sent by social movement groups in Lbrea to the Governor of Amazonas and CNPT, urging them to create the two extractive reserves. One of the criticisms of some groups opposed to extractive reserves was that they were the product of NGO and s ocial movement meddling, usually as a result of international aspirations to control the Brazilian Amazon, and did not necessarily represent the will of the people who actually lived in these areas. I began my list of position generator alters with this g roup of
210 organizations to see what kind of relationships existed between the inhabitants of the communities and the social movement groups that wrote letters demanding the creation of the extractive reserves. The following groups wrote letters in 2007 deman ding the creation of the extractive reserves in Lbrea: 1. Association of the Agroextractive Producers of the Assembly of God of the Ituxi River (APADRIT); 2. Association of the Agroextractive Producers of the Middle Purus (ATAMP); 3. National Council of Rubber T appers Middle Purus Region (CNS); 4. Pastoral Land Commission / Prelacy of Lbrea (CPT); 5. Network of the Amazon Working Group Middle Purus Region (GTA); 6. Assembly of God Lbrea (IEADLAM); 7. Peasant Womens' Movement (MMC); 8. Organization of the Indigenous Peo ples of the Middle Purus (OPIMP); 9. Prelacy of Lbrea; 10. Rural Workers' Union of Lbrea. The organizations that wrote these letters were only a subset of all of the organizations that existed in Lbrea in 2008 2009. In order to compile as complete a list as possible of the groups and organizations with which riverine inhabitants could have had contact, I conducted free lists with several social movement coordinators. The results of these free lists were a list of forty four groups and organizations compri sed of federal, state, and municipal government agencies, social movement groups, religious organizations, basic service providers, and traditional actors such as river traders. A list of the groups and organizations is presented in Appendix Table A 1. De veloping this list of groups and organizations enabled me to begin to answer the first
211 of the research questions presented at the beginning of this chapter. Who were the actors that made up the "new" social networks of riverine inhabitants in Lbrea? I f ound out later that in spite of having vetted the list with social movement leaders, there were in fact several more groups and organizations, particularly community associations, that were not included in my original list. However, the forty four alters provided a list that was representative of the groups and organizations with which riverine inhabitants in and around Lbrea could have interacted. The list of forty four potential contacts increased the length of the interviews, but I believed it was nece ssary to address my research question. Due to the number of potential contacts, I was forced to condense some groups. Instead of having seventeen different cards for the seventeen different Evangelical churches in Lbrea, I simply asked people about thei r interactions with "Pastors." Instead of asking respondents about their interactions with the nine different municipal secretaries I asked about their interactions with the general term secretrio municipal When they said that they had had contact with a municipal secretary I asked follow up questions to determine which secretary: 1) Health, 2) Education, 3) Production, etc. Given the length of these interviews I saw no other way to create the list of potential contacts. Sampling The Mdio Purus and Ituxi Extractive Reserves consisted of more than one hundred communities spread out over hundreds of kilometers. Aside from the fluvial access provided by the two river systems, there was no way to reach the communities in these reserves. The communities were not on an electric grid, and only three of the more than one hundred communities had a telephone. Several of the communities had old CB radios linked to a system in Lbrea, which was inoperable much of the time. The
212 population data gathered by socia l movement organizations was very basic and usually out of date. In order to capture the maximum of variability in the sample population, I employed a purposive sampling strategy (Bernard, 2006) My plan was to select four communities in each reserve and one outside the reserves in which to co nduct my research. I had originally planned to make livelihood strategies my main selection criteria. My plan was to select: (1) a rubber tapper community, (2) a community in which Brazil nuts were the main economic activity, and (3) a community in whic h agriculture was the principle source of income. I quickly learned that this strategy did not reflect the current socioeconomic reality of the area and that I would have to take a different approach to community selection. Agriculture was the main econ omic activity for virtually all of the communities inside (and outside) the extractive reserves. Extractive activities, particularly Brazil nut collection, still existed but were secondary and tertiary economic activities for most communities. There were no communities that depended solely upon one economic activity; as I showed in Figures 4 1, 4 2, and 4 3 in Chapter 4, individuals engaged in different activities throughout the year. Upon learning this, I decided to select communities based on: 1) geog raphic distance from urban areas; 2) participation in Extractive Reserve and social movement activities; and 3) location inside or outside the reserves. I posited that geographic distance from urban centers would limit inhabitants' interactions with socia l movement and government agency personnel, in addition to limiting their access to market options,
213 and therefore limit them to the patron client ties that characterized extractive production in Lbrea for so long. I learned that leaders from two communi ties in the Mdio Purus Extractive Reserve and two communities in the Rio Ituxi Extractive Reserve had been the driving force behind the demands to create the reserves. These leaders had actively pursued the creation of the reserves over a period of ten y ears, and I wanted to see how the social networks of inhabitants of those communities would differ from those of inhabitants of communities with a less active role in this process. Finally, I noticed that the social movement organizations and government o fficials involved in the creation of the reserves targeted the communities located in the reserves to participate in meetings and activities. By conducting interviews in some communities outside the reserves I would be able to see how (or if) the social r elations of their inhabitants differed from those of people living in the reserves. I attended a meeting in Lbrea in April 2008 in which many of the leaders of the communities in the two extractive reserves were present. I introduced myself to the grou p, explained my proposal, and inquired about the possibility of visiting their communities. This was an excellent opportunity to make contacts with individuals in the communities, and it helped me to get my interviews underway. I was also able to meet so cial movement leaders. Before the meeting I visited several communities along the Ituxi River when a boat was sent to pick up the participants who attended the meeting, and I was able to visit several communities along the Purus River when another boat wa s sent to drop participants off after the meeting.
214 My plan was to begin my social network interviews in the communities along the Ituxi, conduct the interviews in the four communities I had selected, and then move on to the Purus. I began with the commu nity of Floresta, one of those most heavily involved in the creation of the reserve, where I spent about two weeks. During this visit I also did a few interviews in Vila Vitria, the community that had pushed the hardest for the reserve on the Ituxi. On this same trip, I visited Cachoeira Vai e Volta, one of the most remote communities (310 km by river from Lbrea) in the reserve that is located on the Siriquiqui River, a tributary of the Ituxi. Unfortunately, during my visit in Floresta I contracted mal aria, which would eventually force me to change my sampling strategy for the Ituxi. After a month of recuperation and treatment, I returned to the Ituxi to try to get the rest of the interviews from the other communities to complete my sample. I contract ed malaria again and was advised by physicians that I should not visit the area again until I had completely recovered, if at all. At this point I had only fifteen or so interviews from the Ituxi, and I had hoped to get at least thirty. Residents of the Ituxi often went to Lbrea to buy supplies, visit family, etc., which allowed me to set up interviews with them when they were in town. This meant that I could no longer completely control from which communities I could obtain respondents. I had to take what I could get, which means I have a sample that includes respondents from eleven different communities in the Ituxi Extractive Reserve instead of the four on which I had originally planned. I was able to get respondents from communities involved in th e creation of the extractive reserves, others who were not involved in this process, and others from remote parts of the reserve. I was able to achieve many sampling goals, although the
215 sample was not exactly what I had originally planned. I had to change from a purposive strategy to one that included snowball sampling techniques (Goodman, 1961) in which I would ask people from the Ituxi if there were other residents in town and if they could help me set up interviews with them. I conducted thirty two position generator interviews from eleven different communities on the Ituxi. On the Purus River I was able to follow my sampling strategy for the social network interviews as I had originally planned. I was able to sele ct communities that fit my sampling criteria: 1) inside the reserve; 2) outside the reserve; 3) involved in the creation of the extractive reserve; 4) not involved in the creation of the reserve, and 5) geographically isolated. Even though I had decided t hat livelihood was not a good selection criterion, I was able to select a community on the Purus in which a significant number of men still tapped rubber. In addition to the four communities where I spent a significant amount of time conducting interviews : 1) Vila Limeira; 2) Bela Rosa; 3) Praia do Buraco; and 4) Santa Rosa, I was able to stop in several other communities to conduct interviews. On the Purus River I conducted sixty four position generator interviews from fifteen different communities. The following is a breakdown of my social network interview sample. Sample of Community Inhabitant Social Network Rating Interviews Rio Purus (Lives inside Resex) = 42 Rio Purus (Lives outside Resex but involved in Resex meetings) = 12 Rio Purus (Lives out side Resex no meetings) = 10 Rio Ituxi (Lives inside Resex) = 29 Rio Ituxi (Lives outside Resex but involved in meetings) = 3 Total number of social network interviews = 96
216 Conducting the Position Generator Interviews The following is a description of how I conducted these interviews and my observations regarding how people responded to my questions. First, I made a set of flash cards with the names and acronyms of the groups, agencies, and social movements on my position generator alter list. It was imp ortant to include both the full name of the organization and the acronym because some organizations were known more by the acronym than the full official name. I coded the cards with numbers so that I could rapidly write down how respondents grouped them. First, I introduced myself to the respondents and explained my reasons for wanting to conduct an interview with them. I let them know that I was interested in understanding the changes that had occurred in the lives of people living in the interior of Lbrea over the past fifty years and that I would like to ask them some questions about the contacts they have with a series of groups and organizations. I used my IRB script and often had to explain in more detail that the interview was not an exam and t hat I was only interested in their personal experiences. Many people were nervous when I asked to do the interviews, and this seemed to calm them. Respondent Characteristics Respondents were asked a series of basic questions to provide data that would en able me to run tests to see if certain characteristics were independent variables that explain the composition of their social networks. These questions included 1) Age, 2) Sex, 3) Religion, 4) Education, 5) Livelihood, 6) Community, 7) Years lived in the community, and 8) Number of times traveled to Lbrea per year. For respondents in communities on the western side of the municipality of Lbrea, I also asked how many times per year they went to Pauini, which was the town closest to many of those
217 communi ties. In addition I noted whether or not the individual was a community leader, the location of their community (inside or outside reserve), and the river on which their community was located. Since most of the respondents had livelihood strategies that varied from one season to another and were usually based on more than one activity, I asked them which production activity provided the most income for them over the past year. Position Generator Questions Then I used the forty four flash cards with the names of the different groups on my list of position generator contacts, and an interview script, to ask respondents questions regarding their social networks. Respondents were asked: 1) If they had heard of the group (yes/no), 2) to rate the group's pow er (powerful, regular, weak), 3) whether or not they had had contact with people from the group in their village, in the city, or anywhere else (yes/no), 4) whether or not they had spoken to someone from the group in their village (yes/no), and 5) the amou nt of assistance the group provides them (a lot, a little, none). Follow up questions were asked (particularly for questions 4 and 5) to find out how often the group visited their community and what kind of help they provided. When people said they had c ontact with a patron or river trader, I asked them who that person was, what they did, and where they lived. This constrained pile sort method of using flash cards to ask questions about relationships worked very well. However, there were some difficultie s. For example the low literacy rates of many of the respondents meant that I could not let them control the interview as much as I would have liked. I had to read the name on each card to them as I asked the question instead of giving them the cards and letting them sort them and put them into piles. This increased the length of the interviews and the burden on the
218 interviewer. In other research projects in areas with low literacy rates, pictures have been used to create flash cards that respondents ca n handle and group. This can be particularly successful when researchers are asking questions about things like fruits, fish, and animals. I thought that perhaps the logos of the organizations could be used on the cards, but this would really be testing respondents' ability to recognize logos and not necessarily measure their ability to recognize the organizations or groups. The fact that many of the groups had no logo or symbol compounded the problem. I could think of no other way to conduct the interv iew except to read each card to the respondents and put them in the piles myself, according to their answers. Question #1 Recognition, was designed to find out which of the organizations people recognized and which they did not. I simply told people tha t I would read the name on the card, and I wanted them to tell me if they had heard of the group or not. When I began the interviews I thought that respondents would not have heard of many of these organizations, and I was surprised by how many of them mo st people had heard of. People seemed to understand this question quite easily. Many people were nervous when I started these questions and then relaxed as we went along. Most of the respondents were highly conscious of their lack of formal education an d did not want to be embarrassed or feel stupid because they could not answer some of the questions. I did my best to explain that there were no "wrong" answers and that I only wanted to understand their personal experiences. I got the feeling that in so me cases respondents may have said that they had heard of an organization when they had not. In these cases I tried to ask follow up questions about the organization to determine if they had heard about it or were confusing it with something else. There were some
219 organization names that people confused with other groups and organizations, which may have affected their answers to this question. Before continuing on to the Question #2 Power, I separated out the cards of the organizations that the responden ts did not recognize. Question #2 was included to try to understand which of the groups and organizations respondents felt were the most powerful. In other studies that employ the Position Generator method "prestige" is a key variable that is used to det ermine where alters fit in a social structure relative to the position of the respondent (van der Gaag et al., 2008 ) However, since my goal was to ascertain the changing nature of patron client ties as well as understand peoples' social capital I felt that "power" would be a more appropriate metric for my study. I told the respondents that I would read the cards t o them again and that I wanted them to tell me if the organization or group was: "Powerful," "Regular," or "Weak." Ideally the scale for this question should have been "Very strong," "Strong," or "Weak" and in future studies I would not use the scale that I used here. Although I would change the scale in future studies, I feel that what made this question difficult for respondents was the diversity of organizations and groups on my list of potential contacts. For example, I noticed several times during i nterviews that respondents would change their use of "power" depending on the organization they were asked about. For example, respondents seemed to rate groups like the Army and Federal Police as powerful because they are armed and have a license to use force while for other groups they changed their rating of "powerful" to mean "good," "does their job well," or "helps us." The same was true for the "weak" rating option. During the same interview I saw respondents rate organizations as "weak" because th ey felt they were
220 "bad," "powerless," "do not do their job well," "do not have enough staff to do their job," or "do not help us." In some cases I believe that people used the "regular" category when they were not sure how to rate an organization. In oth er cases respondents told me that they had heard of the group or organization but did not have enough information about their work to rate them. In future studies, I will attempt to narrow possible interpretations of definitions. It was apparent that resp ondents rated many organizations as powerful in comparison to their personal powerlessness. In Question #3 Contact, I asked the respondents about the contact they had with the groups and organizations. I framed this question by asking them if they had s poken with members of the group or organization in town, in their community, or anywhere else. I explained that this contact could have taken place recently or in the past year or two. Depending on their level of comfort I asked follow up questions for s everal of the groups. For example, if they said they had contact with a patron, I would ask who their patron was, where he or she lived, and what he or she did for a living. With a shorter list of alters it would have been possible to ask more follow up questions, but in this case I had to make sure that the interviews were not too long and that I could finish them without taking too much of the respondents' time. Respondents had no trouble answering this question and were confident in their answers rega rding with which groups and organizations they had and had not had contact. I separated out the cards with which they said they had no contact, and moved on to Question #4 Community Contact, which was asked to determine the amount of contact respondents had with the groups and organizations in their communities. I told respondents that I would read the names on the cards again and that I wanted them to
221 tell me if they had spoken to someone from that group in their community. I clarified that in this cas e they would put the cards in the "yes" pile only if someone from the group had visited their community and they had spoken with them there. I felt that quantifying the presence of patrons, social movements, and other actors in the communities would be ke y to understanding peoples' ties and how they use them, so I decided to ask follow up questions when respondents put cards in the "yes" pile. I asked them how often per year the members of the group or organization visited their community. People respond ed both with numerical responses, "they come twice per year," and with information like "only during harvest season" or "only during political campaigns." I noted all of this information. Next I put the cards back together, leaving out only the cards th at respondents had not recognized, and went on to my final question regarding groups and organizations. Question #4 Assistance, was designed to elicit responses regarding the amount of assistance that people were receiving from the different groups and or ganizations. I informed respondents that I would read the names on the cards again, and for each one, I wanted them to tell me if the group or organization helped them "A lot," "A little," or "Not at all." Unlike the question regarding "power" people had no trouble understanding the question and appeared to answer consistently through all the cards. I felt that this question was key to answering my research question, and I decided to, once again, ask follow up questions for all of the cards that they pla ced in the "A lot" and "A little" piles. I asked them what kind of help each group provided, and I wrote their responses out as they answered.
222 Social Capital in Riverine Communities Producing the alter list that I used in the position generator questio nnaire helped me determine with whom riverine inhabitants could potentially form relationships, build social capital, and access essential resources. The data I collected using the questionnaire allowed me to address my remaining research questions. What were the resources that the actors in these "new" networks had at their disposal? How likely were these actors to provide riverine inhabitants with access to these resources? What characteristics best explain the ability of some riverine to more succe ssfully use their social networks than others? Which of the providers of these resources did the inhabitants consider to be "patrons?" Answering these questions would enable me to consider three important elements of riverine inhabitants' networks: 1) the number of contacts in the network who were prepared to help, b) the extent to which they were prepared to help (the strength of the tie), and 3) what accessible resources they controlled (Flap & Boxman, 2001) Embedded Resources As I described in Chapter 3, the riverine inhabitants of Lbrea were once engaged in the sering al rubber production system in which essential supplies and other "extras" flowed through a network of intermediaries in a credit based exchange system. They were often beholden to a seringal owner patron who maintained a stock of supplies in his barraco trading post that they could access on credit. As I described in Chapter 3, the socio economic changes that occurred in Lbrea over the past fifty years led to the collapse of this system. "New" contacts had formed and resources were embedded in and flo wing through these "new" networks.
223 A second application of the embeddedness concept is to focus on the location of resources within a network. Social resource theory points to wealth, power, and status as the most valued resources in many societies (Lin, 1982) Therefore, social capital can be measured by the amount or variety of these resources that people with whom one has network ties have at their disposal (Lin, 2001a) The variety and location of resources embedded in the personal networks of isolated riverine inhabitants of the Brazilian Amazon indicates what kinds of social capital they had and how they used it. As mentioned in Chapter 2, some researchers believe that when "traditional" patron client relationships dissolve, clients may have to form networks in which several patrons are used to access resources that were once controlled by a single patron (Boissevain, 1977; C. Escobar, 1994; J. C. Scott & Kerkvliet, 1973; Theobald, 1983, 1992) Determining which resources were embedded in the networks of riverine inhabitants would help me to understand if riverine inhabitants were in fact replacing the dyadic multiplex patron client relationships associated with the seringal system with a more diffuse set of relationships in which they accessed resources through a multitude of patrons. As I mentioned previously, during the position generator interviews I asked respondents whether each group and organization helped them: 1) A lot, 2) a little, or 3) not at all. When respondents answered "a lot" or "a little" I asked what kind of assistance they provided. This open ended follow up question enabled me to collect data about the resources that were embedded in respondents' networks and which groups and organizations provided the resources. Often respondents listed more than one kind of assistance for each group or organization. For example, one respondent
224 said that priests provided Information, Supplies, and Religious Services. In order to process this information I first grouped similar responses into thirteen categories. Cr edit Supplies Market Access Health Education Jobs Transportation Information / Advice Documents Programs Protection Religious Services Extractive Reserve Next, I added up all the responses for each category. Finally, I calculated the percentage of each r esource category that was provided by each group and organization on the position generator contact list. The results are presented Appendix Tables A 5 through A 8. In the table the types of assistance are listed on the X axis and the groups and organiza tions are listed on the Y axis. At the bottom of the table is the total number of responses for each type of assistance. In the cells I have placed the percentage of the responses associated with each group / organization. For example, of the 71 respons es in the Credit category, 23% were provided by a "patron". Embedded Resource Results The results presented here show that riverine inhabitants were building networks in which a wide variety of contacts were important for accessing resources. This would appear to support those authors (Boissevain, 1977; C. Escobar, 1994; J. C. Scott & Kerkvliet, 1973; Theobald, 1983, 1992) who observed that one of the possible outcomes when "traditional" dyadic patron client ties change is that they will be replaced by
225 networks in which a wide variety of actors provide resources that were once controlled and provided by a single patron. For example, in spite of the continued importance of Patrons, River Traders, and Shop Owners as a source of credit, the results show that IDAM, a state agency, had become a key source of credit for many inhabitants. Religious organizations provided respondents with a wide array of resources and services, as did the municipal government, which may b e an indication that these groups were more accessible to riverine inhabitants than many of the state and federal agencies. Miller's (1985) observations about the importance of municipal governments as a source of resources for inhabitants of the towns in the interior of the Amazon appear to still be valid based on the results of this analysis. The table shows that these "traditional" patrons were seen almost exclusively as the providers of credit, supplies, and ma rket access, while the other groups and organizations provided a wide variety of resources and services 1 The results also reveal the specialization of some agencies and how this impacted their propensity to provide resources and services to riverine peop les. For example, FUNAI, the federal agency in charge of indigenous affairs in Brazil was not seen as an important source of resources or services by most of the respondents. FUNAI had to prioritize its scarce resources on the provision of services to in digenous peoples and therefore could not or would not assist riverine people in most instances. The results presented in Appendix Table A 3 appear to corroborate the notion that since the departure of the "traditional" seringal owner patrons, many riverin e inhabitants 1 A data collection error caused me to gather incomplete data for the Basic Service Provider groups (Doctor, Dentist, and Teacher), which are therefore underrepresented in their provision of Education and Health in the table in Ap pendix Table A 3.
226 had increasingly turned to an ever increasing number of government agencies, religious organizations, and social movements to access resources and services. Some groups and organizations, particularly the religious organizations and municipa l government agencies, provided a wide range of resources and services. The continuous presence of these religious groups and municipal government agencies over a longer period of time may have meant that they had more multiplex interactions with riverine inhabitants. In addition, these two groups were more likely to maintain control of a wider array of resources and services than other groups, such as the social movement groups with extremely limited operating budgets. Federal agencies were more likely to be seen as a source of protection than any other resource. Several of the federal agencies were under funded field offices that had limited resources at their disposal, which may help explain this result. Although the state agency IDAM was seen as a s ignificant source of credit by respondents, it is interesting to note that River Traders, Patrons, and Shop Owners continued to be important sources of credit and supplies for riverine inhabitants. In summary, the results of the table in Appendix Table A 3 are mixed. They provide evidence that riverine inhabitants had formed relationships with a wide array of groups and organizations in Lbrea and that they were indeed obtaining resources from a variety of contacts. However, it also shows that perhaps in spite of these changes, the older forms of patron client interaction still existed. In Chapter 4 I discussed why riverine inhabitants might continue to be excluded from formal sources of credit, and this may help explain this result. Assistance and Tie Strength It is important to consider that, just because resources are embedded in an individual's network, it does not mean that they will have uniform access to them and
227 therefore, it is important to have an indicator that the measures the availability o f the different resources (van der Gaag & Snijders, 2005) Tie strength is an important measure in social capital and social network analyses and has long provided a challenge for researchers who attempt to measure it (Marsden, 2005; Marsden & Campbell, 1984) The willingness to provide assistance is often equated with the intensity, or tie strength, of a relationship (van der Gaag, 2005) Having determined what types of resources were embedded in riverine inh abitant's networks it was important to address my next research question. How likely were "new" contacts to provide riverine inhabitants with assistance? I attempted to answer this question by asking respondents to rate the amount of assistance provided by each group and organization. In table 6 1 I present their responses to this question. In the data presented in the table I have combined the Helps a lot and Helps a little responses to simplify and make two categories, Helps and Does not Help 2 This table is the aggregated set of all 96 respondents and therefore tends to obscure the differences that would appear if the interviews were separated by geographic location (which river the inhabitant lived on), and other factors. Nevertheless it provides a rough idea of how the riverine inhabitants in the sample viewed the willingness (or capability) of the groups and organizations on the alter list to provide assistance. Several patterns emerge in the table that can help us to understand the networks of the respondents in the sample. First, respondents appeared to have mixed feelings 2 Due to missing data, three of the forty four groups and organizations (IEB, MMTRL, and UEA) in the contact list were not included this analysis nor the subsequent statitstical analyses.
228 about the "traditional" group of patrons, river traders, and shop owners. Roughly half of the respondents felt that these groups were willing to provide assistance and hal f did not. Table 6 1 Likelihood of Providing Assistance Name Type Helps Does not Help Patron Traditional 43.8% 56.3% River Trader Traditional 51.6% 48.4% Shop Owner Traditional 50% 50% Prelacy Religious 44.8% 55.2% Priest Religious 65.6% 34.4% Ass embly of God Religious 35.8% 64.2% Pastor Religious 51.6% 48.4% Mayor Municipal 56.8% 43.2% Councilmen Municipal 44.8% 55.2% Municipal Sec. Municipal 49% 51% Citizen's Center Municipal 74% 26% Doctor Basic Service 88.5% 11.5% Dentist Basic Service 78.1% 21.9% Professor Basic Service 89.6% 10.4% IDAM State 78.1% 21.9% ADS State 41.7% 58.3% ITEAM State 9.4% 90.6% SEARP State 3.1% 96.9% IBAMA Federal 56.3% 43.8% ICMBio Federal 32.3% 67.7% INCRA Federal 43.2% 56.8% Federal Police Federal 21.9% 78.1% Military Police Federal 35.4% 64.6% Army Federal 29.2% 70.8% FUNASA Federal 57.3% 42.7% FUNAI Federal 10.4% 89.6% MMA Federal 34.4% 65.6% APADRIT Association 22.9% 77.1% ATAMP Association 35.1% 64.9% APACJG Association 41.7% 58.3% AS PACS Association 37.2% 62.8% CNS Social Move. 64.6% 35.4% CPT Social Move. 43.8% 56.3% GTA Social Move. 47.9% 52.1% MMC Social Move. 1% 99% OPIMP Social Move. 7.3% 92.7% Rural Union Social Move. 30.2% 69.8% Children's Pastoral Social Move. 42.7% 57. 3% CIMI Social Move. 5.2% 94.8% Grileiro Other 0% 100%
229 Respondents felt that the Catholic Prelacy and Assembly of God church organizations were less likely to provide assistance than individual priests and pastors. This was perhaps due to the more cons tant contact they had with individual priests and pastors than with the leaders of the organizations. The Citizen's Center was the municipal organization respondents felt was most likely to help them. It is interesting that this organization was in char ge of registering inhabitants for Bolsa Familia and other government programs. Bolsa Familia and other government programs, like retirement benefits, had become a key source of income for many families and it is interesting that respondents felt that the Citizen's Center was one of the organizations most likely to provide assistance. Respondents felt that the Basic Service Providers group (Doctors, Dentists, and Teachers) was the most likely to help them. Teachers spent six to eight months in the communit ies during the school year and perhaps their constant presence and the high value that respondents placed on education was part of the reason this group rated so highly. Doctors and dentists had far more sporadic contact, but also provided services respon dents deemed essential, which may also account for their high rating. IDAM was the state agency respondents felt was the most likely to provide them with assistance. IDAM had become a key source of credit (see Appendix Table A 3) and also provided technic al assistance and farming implements. ADS, the state agency in charge of providing technical assistance for extractive activities, had a much more limited staff and budget than IDAM and focused its activities on communities that were actually engaged in r ubber tapping and other extractive activities. ADS focused its attention on the communities on the Purus (where many communities were still
230 engaged in rubber tapping) and rarely visited the communities on the Ituxi (where most communities no longer tapper rubber). This in part would account for the percentage of respondents that said ADS helped them. SEARP and ITEAM did not have offices in Lbrea, which probably accounts for their low rating on this question. The Federal agencies generally rated low on this question. Some because they did not have a physical presence in Lbrea (Federal Police and MMA) and others because they had only recently arrived when the survey was conducted (ICMBio). The groups that were focused on assisting indigenous communiti es (FUNAI and FUNASA) also rated low because they focused their scarce resources on helping indigenous communities. The low rating of the community associations is probably a reflection of their inability to replace the "traditional" groups and provide c redit, supplies, and market access. Appendix Table A 3 shows that they did provide information and other resources, but the low rating for assistance perhaps reveals that riverine inhabitants expected more from them than they were able to provide. CNS r ated the highest among the social movement organizations for its propensity to provide assistance. CNS was actively involved in the process of creating the Extractive Reserves and in the implementation of the state and federal rubber subsidy program in L brea. The social movements involved in indigenous affairs (OPIMP and CIMI) were rated low, again because of their focus on indigenous communities. MMC was one of the organizations that wrote an official letter demanding the creation of the extractive re serves, but in reality had probably never really existed as a fully functioning
231 organization. GTA and CPT were also actively involved in riverine communities, although they rarely visited the Ituxi River, which may explain their low rating. Tie Strength Summary Several trends appear in the results in Table 6 1. First, the results corroborate the assertion that riverine inhabitants had formed social networks that included a wide variety of groups and organizations, including many of the "new" government agencies and social movement groups that had recently arrived in Lbrea 3 Second, these results show that respondents appeared to have stronger ties with groups that: a) spent more time with them (like the school teachers that lived in their villages), b) provided them with resources they valued highly (education, health care, government programs, credit), or c) were able to provide a wide array of resources and services (religious and municipal organizations). Third, most of the groups and organizations were rated at around 50% in terms of their willingness to provide assistance. This shows that in spite of the arrival of "new" actors, there were still many riverine inhabitants in Lbrea that were unable to access resources through these "new" sources of help. This result corroborates my observation in Chapter 4 that many riverine inhabitants were still excluded from many "formal" sources of resources and services. This inability of some riverine inhabitants to obtain resources from expanding government and civil society organizations may explain why modified forms of "traditional' patron client relationships, like those riverine people engaged in with river traders, continued to persist in Lbrea. 3 Appendix Table A 1 includes data regarding the dates that many of the groups and organizations on the list were established in Lbrea.
232 Network Variability, Network Extensity, and Social Ca pital Riverine inhabitants had differing levels of contact and tie strength with the groups and organizations in and around Lbrea. These groups and organizations provided (or had the potential to provide) a diverse array of resources and services. Howe ver, certain inhabitants of the riverine communities in Lbrea were more successful in accessing resources than others. I decided to determine what role network variability played in respondents' ability to access resources. I also decided to test what r ole certain ascribed and attained personal characteristics played in the likelihood that respondents would have more extensive social networks. Research Question #1 My first question was what role network extensity played in respondents' ability to acces s key resources (information) and (wealth). My hypothesis was that respondents with more network variability (contact with more of the groups and organization types listed in Appendix Table A 1) would be more likely to: 1) have information about the extra ctive reserves, and 2) have more wealth. In order to test this hypothesis I generated contact scores based on the data collected with the position generator questionnaires. Contact and Community Contact Scores Respondents were asked if they had contact with each of the forty four groups and organizations on the alter list. Their responses were coded Yes=1 and 0=No. The responses of each respondent were summed to determine their contact score. The minimum possible contact score was zero and the maximum was forty one. 4 4 Three organizations (UEA, MMTRL, and IEB) were not included in the final statistical analysis due to missing data.
233 The same procedure was done to determine the respondents' Community Contact Score. Their responses to the question about which of the forty one groups and organizations they had contact in their community with were coded Yes=1 and 0=No and the responses were summed to determine their community contact score. The minimum possible contact score was zero and the maximum was forty one. Weighted Contact Scores Since the groups and organizations on the contact list developed for this study d iffered in their capacity to provide access to resources, I deemed it necessary to weight the respondent's contact scores. In many other position generator studies an occupational prestige scale is used to weight the scores (van der Gaag et al., 2008) The unique nature of the study area mad standardized occupational prestige rankings inappropriate for weighting the groups with which riverine inhabitants had contact. Instead the scores were weighted using the respondents' answers to the questions in which they were asked to rate the amount of assistance each group or organization provided. Dependent Variables In a ddition to the questions designed to determine the amount of contact that riverine inhabitants in my sample had with the groups and organizations, I also gathered information for the two dependent variables that I wanted to test. The first was a series of questions to determine respondents' knowledge about the newly created extractive reserves and the second was a basic wealth indicator. Extractive Reserve Knowledge Score The a) Conselho Deliberativo (Deliberative Council) b) Plano de Utilizao (Utiliz ation Plan) c) Plano de Manejo Participativo da Resex (Resex Participatory
234 Management Plan) are three important institutional elements of an extractive reserve. They determine how the communities within the extractive reserve will manage and use natural resources. I asked respondents if they had heard of these elements to determine how well they understood the extractive reserve concept. I wanted to determine if there was a correlation between the variability of a respondent's network and their access t o information about the extractive reserves. The extractive reserve knowledge score was generated by numbering respondents' answers to these questions, 1=Yes and 0=No and then summing the scores. The scores ranged from zero to three. Wealth Indicator S cores Finally respondents were asked to respond to an eighteen question wealth indicator survey. This was done to see if there is a correlation between household wealth and social network variability. I developed this list of items with the input of a wo man from one of the communities. The following items were included on the list: 1. Roof type (metal or thatch), 2. Gas stove (Y/N), 3. Television (Y/N), 4. Radio (Y/N), 5. Forno de farinha or manioc flour cooking equipment (Y/N), 6. Rabeta Motor (Y/N), 7. Chainsaw (Y/N), 8. Bed with mattress (Y/N), 9. Mosquito netting (Y/N), 10. Hammock (Y/N), 11. Shotgun (Y/N), 12. Fishing net (Y/N), 13. Canoe, 14. Cattle, (Y/N) 15. Pigs (Y/N), 16. Ducks (Y/N), 17. Chickens (Y/N), 18. Dogs (Y/N).
235 When respondents answered "Yes" to an item like Cattle, Pigs, Dogs, Sh otgun, and Fishing Net, I asked how many they owned. I found that most of the time people were unable to tell me the number of chickens they owned. Due to the relatively small number of position generator interviews I was able to collect (96), creating a wealth index using all eighteen of my wealth indicators items was not statistically feasible. I decided to create simpler wealth indices, one of domestic consumer items, and one of production items. The domestic consumer items were: 1) Gas Stove, 2) Ra dio, and 3) Television. For these three items I had data from ninety two of the ninety six respondents. Eighty percent of these respondents had a gas stove, fifty five percent had radios, and thirty nine percent had televisions. This was designed to be a hierarchical measure of wealth. The assumption was that individuals with a radio would already have a gas stove and that individuals with a television would already have a gas stove and a radio. To generate the respondents' domestic wealth indicator sc ore the responses were number 1=Yes and 0=No and the ones were summed to determine a score between zero and three. The production item wealth indicator consisted of the 1) manioc roaster, and 2) the rabeta canoe motor. I collected data on these two item s from ninety two of the ninety six respondents. Sixty percent of the respondents had manioc roasters and seventy two percent had rabeta motors. To generate the respondents' production wealth indicator score the responses were number 1=Yes and 0=No and t he ones were summed to determine a score between zero and two.
236 I finished with three different indicators of wealth 1) Rabeta Motor, 2) Manioc Roaster, and 3) Gas Stove, Radio, and Television. The hierarchical Gas Stove, Radio, Television indicator is c alled "Wealth" in the tables that follow. Statistical Analysis Ordered Logit Models First I ran basic correlations to determine the relationships between the data. The independent variables were highly correlated, which precluded them from being used in the same regression model. The ordered logit models did not provide statistically significant results. The Extractive Reserve Knowledge score did not converge to point estimates, probably due to the small sample size and the large number of variables. C hi Squares To offset the problem of running ordered logit models with the data set I decided to run a simpler chi square analysis using SPSS Version 18. First I divided the groups and organizations based on the categories presented in Appendix Table A 1. For example, the "Traditional" group included Patron, River Trader, and Shop Owner, the "Federal" group included all of the federal agencies represented in Lbrea at the time, the "State" group included all of the state agencies, and so on. I divided the groups in this fashion to try to determine how contact with organizations from these different types of agencies was related (or not) to respondents' Wealth and Resex Knowledge Scores. Breaking down the groups and organizations in this way also enabled me to convert the overall Contact Scores, which ranged from 0 to 41, to smaller scores that could more easily be converted to ordinal values that could be used in the X 2 models. For example, the "traditional" group, consisting of Patron, River Trader, and Shop Owner had a score that ranged from 0 to 3.
237 Due to the uneven distribution of the data, I decided to collapse some value categories so that the X 2 tests would perform better. Collapsing some of the values reduced the degrees of freedom and improved the performance of the tests. Below I present the results of the X 2 's with the combined data categories. In Appendix Tables A 5 through A 8 I present the results of the X 2 's before the values were combined. Combining the value categories did not radica lly change the results. I conducted four types of chi square models using the data describing the contact between individuals in the riverine communities and the groups and organizations that I gathered using the position generator. In these models the co ntact with the groups and organizations was the independent variable and the Resex Information, Rabeta Motor, Manioc Roaster, and Wealth indicators were the dependent variables. Hypotheses H1: Respondents who had more variable social networks would have more information about the Extractive Reserves. H2: Respondents who had more variable social networks would have more wealth than those who did not. In Model #1 I ran X 2 tests in which the respondents' contact score was the independent variable. In Model #2 I ran X 2 tests in which the respondents' community contact score was the independent variable. In Model #3 I ran X 2 tests in which the respondents' contact score, weighted by the amount of assistance they said the groups and organizations provided, wa s the independent variable. In Model #4 I ran X 2 tests in which the respondents' community contact score, weighted by the amount of assistance they said the groups and organizations provided, was the independent variable. I weighted the contact scores b y simply multiplying the assistance rating by the contact score. In other words, a weighted contact variable for the municipal
238 organizations was generated with the following calculation (Weighted Municipal Contact Variable = Mayor Contact Mayor Assistan ce + Councilman Contact Councilman Assistance + Municipal Secretary Contact Municipal Secretary Assistance + Citizen's Center Contact Citizen's Center Assistance). My assumption was that the weighted scores would provide a measurement that would bet ter reflect a reality in which certain groups and organizations could provide more resources and therefore should not have the same value as organizations that provided less. Statistical Results Research Question #1 Extractive Reserve Knowledge The Chi S quare tables presented below reveal that there was a relationship between having contact with certain groups and organizations and having knowledge about the extractive reserves. Tables 6 2 and 6 3 contain the results of bivariate chi squares in which the contact and community contact scores were the independent variables and the ResexInfo, Rabeta Motor, Manioc Roaster, and Wealth Indicators were the dependent variables. Table 6 2 Network Variability and Contact Bivariate X 2 Contact Group Type ResexInf o Rabeta Motor Manioc Roaster Wealth Traditional X 2 2.446 0.647 1.782 2.399 Religious X 2 5.595 1.041 4.018 5.209 Municipal X 2 6.62 4.327 0.093 3.336 Basic Service Provider X 2 1.543 0.361 1.039 0.134 State X 2 19.359** 6.09 2.439 1.366 Federal X 2 21. 278** 0.347 1.358 2.496 Associations X 2 23.825** 5.123 2.771 7.279 Social Movements X 2 14.334* 0.293 0.766 8.966
239 Table 6 3 Network Variability and Community Contact Bivariate X 2 Community Contact Group Type ResexInfo Rabeta Motor Manioc Roaster Wealt h Traditional X 2 10.717* 4.074 2.238 12.254* Religious X 2 8.633 3.348 1.484 6.764 Municipal X 2 27.980** 4.395 3.854 4.873 Basic Service Provider X 2 10.513* 0.053 1.851 4.23 State X 2 26.362** 4.799 0.313 2.872 Federal X 2 30.556** 1.835 9.525 13.804 Associations X 2 23.091** 2.171 3.22 9.669 Social Movements X 2 26.174** 3.421 1.878 6.082 Tables 6 4 and 6 5 contain the results of bivariate chi squares in which the contact and community contact scores weighted with the assistance scores were the inde pendent variables and the ResexInfo, Rabeta Motor, Manioc Roaster, and Wealth Indicators were the dependent variables. My assumption was that weighting the contact scores with the assistance scores would provide a more accurate measure for this analysis. Table 6 4 Network Variability and Contact Weighted by Assistance Bivariate X 2 Contact Weighted by Assistance Group Type ResexInfo Rabeta Motor Manioc Roaster Wealth Traditional X 2 16.197* 0.058 6.389* 6.788 Religious X 2 3.304 3.908 1.773 4.221 Muni cipal X 2 3.929 4.003 3.429 7.27 Basic Service Provider X 2 5.188 2.451 4.889* 0.492 State X 2 24.220** 2.503 0.083 4.616 Federal X 2 19.800* 2.961 5.968 12.482 Associations X 2 18.525** 2.029 1.949 3.277 Social Movements X 2 22.940** 1.459 1.58 8.217
240 Tab le 6 5 Network and Variability and Community Contact Weighted by Assistance Bivariate X 2 Community Contact Weighted by Assistance Group Type ResexInfo Rabeta Motor Manioc Roaster Wealth Traditional X 2 7.17 1.202 2.433 5.207 Religious X 2 4.481 6.988 1 .964 8.135 Municipal X 2 10.713* 0.664 0.649 5.893 Basic Service Provider X 2 6.034 1.185 6.976 4.849 State X 2 31.888** 5.927 0.011 2.998 Federal X 2 29.756** 2.374 6.196 12.302 Associations X 2 22.692** 1.02 3.524 2.095 Social Movements X 2 30.601** 5.9 83 8.142* 8.03 Chi Square Results Resex Information and Network Variability In the X 2 results presented in Table 6 2 there is a pattern that is repeated in all of the subsequent models. Having a higher Contact Score with state and federal agencies, comm unity associations, and social movements was related to having a higher Resex Information Score. In other words, respondents who had more variable networks that included these types of groups and organizations tended to have more information about the ext ractive reserves. Since all of the community associations, most of the social movement organizations, as well as many of the federal and state agencies on the alter list were actively promoting the creation of the extractive reserves, it comes as no surp rise that respondents with more contact with these groups would have higher Resex Information Scores. There are anomalies in Models 2, 3, and 4 that I find difficult to explain. For example, in Models 2 and 3 there appears to be a relationship between h aving more
241 contact with "traditional" groups and having more information about the extractive reserves. Patrons, river traders, and shop owners were often the most threatened by the creation of the extractive reserves and many had actually spread false in formation about the reserves in the hopes that inhabitants would change their mind about creating them. The same can be said about the result for municipal organizations in Table 6 5. The municipality of Lbrea was against the creation of the extractive reserves for several years prior to this research. It is not clear why someone with more municipal contacts would have a higher Resex Information Score. Based on these results I would reject the null hypothesis for H1. There does appear to be a relati onship between having a more variable network and having more knowledge about the reserves. I would fail to reject the null hypothesis in the case of H2. These results do not show a significant relationship between the variability of network contacts and wealth. The difference between contact in general and contact in the respondents' communities seemed to increase the likelihood that they would have more knowledge about the reserves. Weighting the contact scores by the assistance provided did not appe ar to significantly change the results of the X 2 tests. Wealth and Network Extensity Another pattern established in this first model is that there appears to be little or no relationship between having contact with the groups and organizations and having m ore wealth. There are exceptions in Models 2, 3, and 4 but the general impression from these results is that having a more variable network of contacts with these groups and
242 organizations did not necessarily imply that a respondent would benefit with mate rial items. This result seems counterintuitive based on direct observations and the results of the community histories presented in Chapter 5. For example, it was clear that the inhabitants of Vila Limeira actively formed relationships with several grou ps in order to form their community association. Once they had formed their association they used their relationship with IDAM, a state agency, to obtain farming implements to process sugarcane. Rabeta motors and manioc roasters are some of the most comm on implements that have been distributed to riverine inhabitants as part of government aid programs. However, perhaps the wealth indicators used here were not an appropriate measure of wealth. Another explanation may be the timing of the research. The Mdio Purus and Ituxi extractive reserves were created in May and June 2008, but the material benefits associated with the creation of extractive reserves were not distributed until April 2010. Since these benefits were distributed after the interviews we re conducted, it is possible that the result of the wealth questions would have been different if the interviews had been conducted after the benefits had been distributed. The variability of respondents' social networks was not a good predictor of their w ealth. There did not appear to be a relationship between having contact with more groups and organizations and having more wealth. Research Question #2 Dependent Variables My second question was what role certain individual characteristics play in determi ning the extensity of a riverine inhabitants social network. In this analysis I
243 decided to use the Contact and Community Contact Scores used in Research Question #1 as dependent variables. The respondents' Recognition Score (the number of groups and orga nizations they had heard of) as well as their Membership Score were also used as dependent variables. I generated the Recognition scores by summing the number of groups and organizations that respondents said they had heard of during the position generato r questionnaire. Scores ranged from 0 to 41. Membership in groups and organizations is often considered a form of social capital (Putnam, 2000; Putnam, Leonardi, & Nanetti, 1993) and during the interviews I asked respondents if they belonged to any of the groups and organizations listed (or any others). I summed the number of groups and organizations for each respondent to determine their Membership Sc ore. Independent Variables I decided to test whether or not Age, Sex, Education (Yrschl), Religion, Living inside the Extractive Reserves (Resexyn), Community Distance from Lbrea (DistLab), and being a Community Leader were good predictors of the extensit y of riverine inhabitants' social networks. Correlations I used SPSS Version 18 to determine if any of the independent variables in my model were correlated. Age was correlated with Education (Yrschl). Religion was correlated with Living inside an Extra ctive Reserve (Resexyn) and with Community Distance from Lbrea (DistLab). Living inside an Extractive Reserve (Resexyn) was also correlated with Community Distance from Lbrea. In the following linear regression models I avoided combining correlated ind ependent variables.
244 In the following sections I present my hypotheses and statistical results generated using SPSS regarding how each of the independent variables listed above were associated with respondents' network extensity scores for 1) Recognition, 2) Membership, 3) Contact, and 4) Community Contact. Linear Regression The results for linear regression are identified as Rec#1 through #5 (Recognition), Memb#1 through #5 (Membership), Con#1 through #5 (Contact), and CC#1 through #5 (Community Contact) The independent variables used in each model are consistent for all four categories. In Score#1 the explanatory variables were Age, Sex, Religion, and Community Leader. In Score#2 the explanatory variables were Age, Sex, Lives inside Extractive Reserv e, and Community Leader. In Score#3 the explanatory variables were Sex, Education, Religion, and Community Leader. In Score#4 the explanatory variables were Sex, Education, Community Leader, and Lives inside Extractive Reserve. In Score#5 the explanator y variables were Age, Sex, Education, Community Leader, and Community Distance to Lbrea. Recognition Hypotheses for Recognition H1: Community Leaders have higher Recognition scores than non leaders. H2: Men have higher Recognition scores than women. H3: Younger respondents have higher Recognition scores than older respondents. H4: Catholics have higher Recognition scores than Evangelicals. H5: Respondents living inside the Extractive Reserves have higher Recognition scores than respondents living outs ide the Extractive Reserves.
245 H6: Respondents with more education have higher Recognition scores than respondents with less education. H7: Respondents living in communities further from the town of Lbrea have lower Recognition scores than those living cl oser to town. Recognition Results and Discussion Hypothesis #1: Community Leaders have higher Recognition scores than non leaders. Result #1: The results of the five linear regression models for Recognition reject the null hypothesis for H1. The Communi ty Leader variable was highly significant in three of the models (Rec#1, #2, and #3) and moderately significant in the remaining two models (Rec#4 and #5). There was a positive association between being a community leader and recognizing a greater diversi ty of the groups and organizations in the position generator questionnaire. As I described in Chapter 5, community leaders were increasingly assuming a "broker" position between their communities and the outside world. They were consistently invited to m eetings and were expected by other community inhabitants to have contacts with groups and organizations that could help resolve problems or bring much needed resources into the community. Therefore, it is not surprising that being a community leader was a ssociated with the respondents' ability to recognize the groups and organizations on the questionnaire. Hypothesis #2: Men have higher Recognition scores than women. Result #2: Sex was not associated with the respondents' ability to recognize the groups and organizations in the questionnaire. The five linear regression models for Recognition failed to reject the null hypothesis for H2. I had assumed that men would be more likely to be engaged in activities outside the home and would therefore be more lik ely to recognize the groups and organizations than women. The results of this analysis imply that this was not the case, at least not in the sample of interviews I collected. Women
246 appeared to have similar information about the existence of these groups and organizations as men. Hypothesis #3: Younger respondents have higher Recognition scores than older respondents. Result #3: The results of the five linear regression models for Recognition failed to reject the null hypothesis for H3. Age was mode rately significant as an explanatory variable for recognition in model Rec#5. However, it was not significant in the other two models in which I used it as an independent variable. Perhaps with a larger sample size the significance of Age as an independen t variable for Recognition would be different. However, because it was moderately significant in only one of three models I would not consider Age an explanatory variable in this analysis. Hypothesis #4: Catholics have higher Recognition scores than Ev angelicals. Result #4: Religion was moderately significant as an explanatory variable in model Rec#1 and highly significant as an explanatory variable in model Rec#3. However, the regression models failed to reject the null hypothesis because I surmised that Catholics would be more likely to recognize the groups and organizations than Evangelicals. As I mentioned in Chapter 5, Evangelicals have a reputation as being less likely to engage in social movement activities than Catholics. However, in my samp le it was the Evangelicals (coded as 1) who were more likely to recognize the groups and organizations. Therefore, although the models failed to reject the hypothesis as it is written here, religion was significant as an explanatory variable for responden ts' ability to recognize the groups and organizations. In Chapter 5 I cited several other studies that refute the notion that Evangelicals are not involved in social movement activities or
247 politics and do not engage in activities that challenge the status quo of social interactions (Freston, 1993; Ireland, 1993) These results appear to show that Evangelicals in Lbrea were actually more likely to recognize a wider array of groups and organizations than Catholics. Hypothesis #5: Respondents living inside the Extractive Reserves have higher Recognition scores than respondents living outside the Extractive Reserves. Result #5: The five linear regression models failed to reject the null hypothesis for H5. Living ins ide an extractive reserve was moderately significant as an explanatory variable for recognition in model Rec#4 but was not significant in model Rec#2. It is unclear whether or not living inside an Extractive Reserve explains the respondents' ability to re cognize the groups and organizations in the questionnaire. My assumption, based on these results, is that families inside and outside the reserves had roughly the same amount of information about the number and type of organizations operating in Lbrea. Hypothesis #6: Respondents with more education have higher Recognition scores than respondents with less education. Result #6: The linear regression models for Recognition also failed to reject the null hypothesis for H6. Education was a moderately sign ificant variable in model Rec#5, but was not significant in the other two models in which it was tested. Therefore, it is unclear what role education played in the respondents' ability to recognize the groups and organizations in the questionnaire. It ma y simply be that having a higher level of education did not mean that individuals in riverine communities would have more information about the groups and organizations operating in Lbrea.
248 Hypothesis #7: Respondents living in communities further from the town of Lbrea have lower Recognition scores than those living closer to town. Result #7: Community distance to Lbrea was correlated with Religion and Living Inside an Extractive Reserve. For this reason, I only included it as an independent variable in model Rec#5. Based on the results, this model failed to reject the null hypothesis for H7. The distance from the respondents' communities to the town of Lbrea was not associated with their ability to recognize the groups and organizations in the questio nnaire. It may be that although they lived further from town, even isolated respondents had roughly the same amount of information about the existence of the groups and organizations as respondents that lived closer to Lbrea. Recognition Summary Table 6 6 shows that of the seven independent variables tested to explain the respondents' ability to recognize the groups and organizations in the questionnaire, being a community leader was significant in all five models. In Chapter 5 I discussed Table 6 6 Reco gnition Scores Rec#1 Rec#2 Rec#3 Rec#4 Rec#5 R 2 0.178 0.135 0.208 0.175 0.199 F 4.669** 3.509* 5.569** 4.570** 4.131** Age B 0.025 0.018 0.071* Sex B 0.342 0.059 0.691 0.636 0.323 Religion B 1.634* 1.990** Yrschl B 0.206 0.195 0.318* Comm. Leader B 2.699** 2.790** 2.540** 2.202* 2.221* Resexyn B 0.889 1.613* Dist. to Lbrea B 0.002 how community leaders were becoming more and more important as brokers between their communities an d the outside world. It is therefore no surprise that they would be more likely than non leaders to recognize the groups and organizations. Religion was
249 also an important variable explaining the respondents' ability to recognize the groups and organizatio ns. In contrast to my hypothesis, Evangelicals were actually more likely than Catholics to recognize a greater diversity of groups and organizations. Evangelical communities played significant roles in the creation of the extractive reserves in Lbrea, whi ch is contrary to the notion that Evangelicals do not engage in politics or social movement activities. As discussed Chapter 5, Evangelicals in Lbrea showed a greater propensity to organize themselves and strive to improve their living standards than Cat holics. This may explain why Evangelicals were more likely to recognize the groups and organizations than Catholics. The remaining variables were either not significant or significant in only one or two of the regression models. Further research and a la rger sample size would be required to validate or refute the significance of these variables to influence the respondents' capacity to recognize the groups and organizations. Membership Hypotheses for Membership H1: Community Leaders have higher Membersh ip scores than non leaders. H2: Men have higher Membership scores than women. H3: Younger respondents have higher Membership scores than older respondents. H4: Catholics have higher Membership scores than Evangelicals. H5: Respondents living inside the Extractive Reserves have higher Membership scores than respondents living outside the Extractive Reserves. H6: Respondents with more education have higher Membership scores than respondents with less education. H7: Respondents living in communities furt her from the town of Lbrea have lower Membership scores than those living closer to town.
250 Membership Results and Discussion Hypothesis #1: Community Leaders have higher Membership scores than non leaders. Result #1: The results of the five linear regres sion models for Membership reject the null hypothesis for H1. The Community Leader variable was highly significant in all of the models (Memb#1, #2, #3, #4, and #5). There was a positive association between being a community leader and belonging to a grea ter number of groups and organizations. As I described in Chapter 5, community leaders were increasingly assuming a "broker" position between their communities and the outside world. They were consistently invited to meetings and were expected by other c ommunity inhabitants to have contacts with groups and organizations that could help resolve problems or bring much needed resources into the community. Status as a member in some cases may have strengthened a leader's tie with a group or organization and made it more likely that they would help with resources, information, etc. Being a member of certain groups and organizations may have also strengthened their position as a leader in the community, where other inhabitants expected leaders to have contacts with groups and organizations that could help them. For these reasons, it is not surprising that being a community leader was associated with the respondents' propensity to be a member in more groups and organizations. Hypothesis #2: Men have higher Mem bership scores than women. Result #2: Sex was not associated with the respondents' propensity to be a member in groups and organizations. All five linear regression models failed to reject the null hypothesis for H2. I had assumed that men would be more likely to be engaged in activities outside the home and would therefore be more likely to be members of groups and organizations than women. The results of this analysis imply that this was not the case, at least not in
251 the sample of interviews I collecte d. Men and women appeared to have similar tendency to be members of groups and organizations. These results seem to indicate that women may have had more opportunities for social engagement outside the home than in the past. In my trips to the communiti es I noticed that women were highly involved in church groups, meetings, and other activities designed to bring improvements to their communities. In addition, the head of the IBAMA office in Lbrea was a woman, as was the head of the office of the Nation al Council of Rubber Tappers. These women were just as likely to interact with women as they were with men on trips to the interior, and made sure that women were included in meetings and other activities. Perhaps the increasing opportunities for women i n government and social movement organizations had also led to more opportunities for women in riverine communities to participate in groups and organizations than in the past. Hypothesis #3: Younger respondents have higher Membership scores than older r espondents. Result #3: The results of the five linear regression models failed to reject the null hypothesis for H3. Age was not significant as an explanatory variable for Membership in any of the regression models. A respondent's age was not associated with their propensity to be a member in groups and organizations. Groups and organizations appeared to have members from a fairly wide range of age groups and perhaps there are other factors that better explain why people decided to participate in groups and organizations. Hypothesis #4: Catholics have higher Membership scores than Evangelicals. Result #4: Religion was highly significant as an explanatory variable in models Memb#1 and Memb#3. However, the regression models failed to reject the null h ypothesis
252 because I surmised that Catholics would be more likely to recognize the groups and organizations than Evangelicals. As I mentioned in Chapter 5, Evangelicals have a reputation as being less likely to engage in social movement activities than Cat holics. However, in my sample it was the Evangelicals (coded as 1) who were more likely to be members of groups and organizations. Therefore, although the models failed to reject the hypothesis as it is written here, religion was significant as an explan atory variable for respondents' propensity to be members of groups and organizations. Hypothesis #5: Respondents living inside the Extractive Reserves have higher Membership scores than respondents living outside the Extractive Reserves. Result #5: The f ive linear regression models rejected the null hypothesis for H5. Living inside an extractive reserve was highly significant as an explanatory variable for recognition in model Memb#2 and Memb#4. My assumption, based on these results, is that families in side the reserves were: a) more likely to have community leaders that encouraged them to participate in groups (such as community associations) to better their lives (such as the pastor who organized communities on the Ituxi River) or, b) were more likely to have opportunities to engage in group activities because government and social movement organizations that were interested in creating the Extractive Reserves had prioritized their communities as places they would work. Hypothesis #6: Respondents with more education have higher Membership scores than respondents with less education. Result #6: The linear regression models for Membership also failed to reject the null hypothesis for H6. Education was a moderately significant variable in model Memb#5, but was not significant in the other
253 two models in which it was tested. Therefore, it is unclear what role education played in the respondents' propensity to be members of groups and organizations. Hypothesis #7: Respondents living in communities further from the town of Lbrea have lower Membership scores than those living closer to town. Result #7: Community Distance to Lbrea was correlated with Religion and Living Inside an Extractive Reserve. For this reason, I only included it as an independent va riable in model Memb#5. Based on the results, this model rejected the null hypothesis for H7. The distance from the respondents' communities to the town of Lbrea was associated with their propensity to be members of groups and organizations. As I will show later, community distance was also a significant variable in explaining the amount of contact that respondents had with the groups and organizations in their communities. Social movement groups and even government agencies faced serious resource cons traints that made it difficult for them to visit the more remote communities, which may help explain this result. Fewer visits to their communities by the groups and organizations may have meant fewer opportunities for community inhabitants in remote comm unities to participate as members in the groups and organizations. Membership Summary Table 6 7 Membership Scores Memb#1 Memb#2 Memb#3 Memb#4 Memb#5 R 2 0.249 0.235 0.276 0.282 0.281 F 7.115** 6.919** 8.118** 8.461** 6.499** Age B 0.003 0.002 0.014 Sex B 0.227 0.152 0.246 0.225 0.234 Religion B 0.618** 0.710** Yrschl B 0.054 0.054 0.075* Comm. Leader B 0.982** 0.910** 0.921** 0.735** 0.774** Resex B 0.231** 0.823** Dist. to Lbrea B 0.002**
254 Table 6 7 shows that of the seven independent variables tested to explain the respondents' propensity to be members of groups and organizations, being a community leader was significant in all five models. In Chapter 5 I discussed how comm unity leaders were becoming more and more important as brokers between their communities and the outside world. It is therefore no surprise that they would be more likely than non leaders to be members of groups and organizations. Being a member of groups and organizations may have strengthened their position as leaders in the community by raising their prestige and enabling them to bring resources into the community. Religion was also an important variable explaining the respondents' propensity to belon g to groups and organizations. In contrast to my hypothesis, Evangelicals were actually more likely than Catholics to be members of groups and organizations. In Chapter 5 I discussed how Evangelical communities played a significant role in the creation of the extractive reserves in Lbrea, which is contrary to the notion that Evangelicals do not engage in politics or social movement activities. Evangelicals in Lbrea showed a greater propensity to organize themselves and strive to improve their living sta ndards than Catholics. The result of these Membership regressions seems to corroborate these observations. Living inside the extractive reserve was also significant for explaining the respondents' propensity to belong to groups and organizations. This may be because they: a) had leaders who were more active in encouraging them to participate in groups and organizations, or b) they had more opportunities because their communities were
255 prioritized as work areas for government agencies and social movements interested in creating extractive reserves. Distance was also a significant variable for explaining respondents' propensity to belong to groups and organizations. Respondents living closer to the town of Lbrea were easier to visit and therefore may ha ve had more opportunities to become members in these groups and organizations. Age, Sex, and Education were either not significant or significant in only one of the regression models. Further research and a larger sample size would be required to validat e or refute the significance of these variables to influence the respondents' propensity to become members of groups and organizations. Contact Hypotheses for Contact H1 : Community Leaders have higher Contact scores than non leaders. H2: Men have higher Contact scores than women. H3: Younger respondents have higher Contact scores than older respondents. H4: Catholics have higher Contact scores than Evangelicals. H5: Respondents living inside the Extractive Reserves have higher Contact scores than resp ondents living outside the Extractive Reserves. H6: Respondents with more education have higher Contact scores than respondents with less education. H7: Respondents living in communities further from the town of Lbrea have lower Contact scores than thos e living closer to town. Contact Results and Discussion Hypothesis #1: Community Leaders have higher Contact scores than non leaders. Result #1: The results of the five linear regression models for Membership
256 reject the null hypothesis for H1. The Commun ity Leader variable was highly significant in all of the models (Con#1, #2, #3, #4, and #5). There was a positive association between being a community leader and having contact with a greater number of groups and organizations. As I described in Chapter 5, community leaders were increasingly assuming a "broker" position between their communities and the outside world. They were consistently invited to meetings and were expected by other community inhabitants to have contacts with groups and organization s that could help resolve problems or bring much needed resources into the community. The more contacts a leader had with outside groups and organizations, the more likely he or she would be able to bring benefits to the community. Having contact with a wide array of groups and organizations may have also strengthened their position as a leader in the community, where other inhabitants expected leaders to have contacts with groups and organizations that could help them. For these reasons, it is not surpr ising that being a community leader was associated with the respondents' tendency to have contact with a wider array of groups and organizations. Hypothesis #2: Men have higher Contact scores than women. Result #2: Sex was not associated with the respond ents' tendency to have contact with groups and organizations. All five linear regression models failed to reject the null hypothesis for H2. I had assumed that men would be more likely to be engaged in activities outside the home and would therefore be mo re likely to contact groups and organizations than women. The results of this analysis imply that this was not the case, at least not in the sample of interviews I collected. Men and women appeared to have similar tendency to contact groups and organizat ions.
257 These results seem to indicate that women may have more opportunities for social engagement outside the home than in the past. During interviews I found that the needs of their children often inspired women to make contacts outside of their commun ities. One woman I interviewed who lived in an isolated community on a tributary of the Ituxi River told me that she had knocked on the door of virtually every organization she could find in Lbrea that might possibly help her acquire medical assistance f or one of her children. It was apparent that the often dire economic circumstances of riverine families inspired both men and women to make contact with as wide an array of groups and organizations as possible in the hopes of obtaining assistance. This may have also been a reflection of their need replace a single patron who provided an array of resources with a multitude of actors who provided a smaller range of resources and services. Hypothesis #3: Younger respondents have higher Contact scores than older respondents. Result #3: Age was not significant as an explanatory variable for Contact in any of the three regression models in which it was used. The results of the linear regression models failed to reject the null hypothesis for H3. A responden t's age was not associated with the amount of contact they had with groups and organizations. Hypothesis #4: Catholics have higher Contact scores than Evangelicals. Result #4: Religion was moderately significant as an explanatory variable in both models in which it was used (Con#1 and Con#3). However, the regression models failed to reject the null hypothesis because I surmised that Catholics would be more likely to have contact with the groups and organizations than Evangelicals. As I mentioned in Cha pter 5, Evangelicals have a reputation as being less likely to engage in social movement
258 activities than Catholics. In addition, the Catholic Church in Lbrea had a long history of social movement organization (Christian Base Communities, Rural Workers' U nion, etc.). However, in my sample it was the Evangelicals (coded as 1) who were more likely to have contact with a wider array of groups and organizations. Therefore, although the models failed to reject the hypothesis as it is written here, religion wa s significant as an explanatory variable for respondents' propensity to contact a wide array of groups and organizations. Hypothesis #5: Respondents living inside the Extractive Reserves have higher Contact scores than respondents living outside the Extra ctive Reserves. Result #5: The linear regression models rejected the null hypothesis for H5. Living inside an extractive reserve was moderately significant as an explanatory variable for contact in model Con#2 and highly significant in Con#4. My assumpt ion, based on these results, is that families inside the reserves were a) more likely to have community leaders that sought out contact with groups and organizations to bring much needed assistance to their communities and/or b) were more likely to be cont acted by groups and organizations that have opportunities to engage in group activities because government and social movement organizations interested in creating the extractive reserves had prioritized their communities as places they would work. I beli eve that the answer is a combination of the two. As I described in Chapter 5, community leaders actively sought out groups and organizations to try to resolve land tenure and resource conflicts. However, once the process of creating the extractive reserv es was in motion, the government agencies and social movements interested in creating these reserves targeted the communities
259 inside the proposed reserves as priority work areas and visited them more regularly than communities outside the reserves. Hypot hesis #6: Respondents with more education have higher Contact scores than respondents with less education. Result #6: The linear regression models also failed to reject the null hypothesis for H6. Education was not significant in any of the three models in which it was tested. Hypothesis #7: Respondents living in communities further from the town of Lbrea have lower Contact scores than those living closer to town. Result #7: Community Distance to Lbrea was correlated with Religion and Living Inside a n Extractive Reserve. For this reason, I only included it as an independent variable in model Memb#5. Based on the results, this model failed to reject the null hypothesis for H7. The distance from the respondents' communities to the town of Lbrea was n ot associated with the amount of contact they had with groups and organizations. This is an interesting finding considering that community distance was a significant variable in explaining the propensity of respondents to be members of groups and organiza tions and the amount of contact that they had with the groups and organizations in their communities. Perhaps the example of the female respondent from the Ituxi once again helps to explain this phenomenon. Economic crisis, resource conflicts, and other p roblems inspired many respondents to go to the city and seek out the groups and organizations on the questionnaire for assistance. On the other hand, while several social movement and even government agencies were interested in achieving their goals of cr eating extractive reserves and therefore prioritized the communities inside the reserves for their visits, they were not driven by the same degree of need to visit the
260 communities as the inhabitants of the communities were to seek them out. Perhaps this e xplains why distance was a deciding factor in determining whether or not groups and organizations visited the communities, but was less significant in explaining whether or not inhabitants had contact with the groups and organizations. Contact Summary Tab le 6 8 Contact Scores Con#1 Con#2 Con#3 Con#4 Con#5 R 2 0.198 0.194 0.203 0.237 0.179 F 5.298** 5.418** 5.405** 6.682** 3.613** Age B 0.072 0.044 0.087 Sex B 0.779 0.153 1.744 1.491 0.597 Religion B 2.691* 3.246* Yrschl B 0.012 0.008 0.076 Comm. Leader B 5.742** 5.732** 6.014** 4.841** 5.725** Resex B 3.862* 4.979** Dist. to Lbrea B 0.003 Table 6 8 shows that of the seven independent variables tested to explain the responden ts' propensity to be members of groups and organizations, being a community leader was highly significant in all five models. In Chapter 5 I discussed how community leaders were becoming more and more important as brokers between their communities and the outside world. It is therefore no surprise that they would be more likely than non leaders to be in contact with groups and organizations. Having contact with a wide array of groups and organizations may have strengthened their position as leaders in the community by enabling them to bring resources into the community. Religion was also a moderately significant in explaining the respondents' propensity to have contact with groups and organizations. In contrast to my hypothesis, Evangelicals were actually more likely than Catholics to have contact with groups and organizations. In Chapter 5 I discussed how Evangelical communities played a
261 significant role in the creation of the extratctive reserves in Lbrea, which is contrary to the notion that Evangelic als do not engage in politics or social movement activities. Evangelicals in Lbrea showed a greater propensity to organize themselves and strive to improve their living standards than Catholics. The result of these Membership regressions seems to corrob orate these observations. Living inside the extractive reserve was also significant for explaining the respondents' tendency to have contact with groups and organizations. This may be because they: a) had leaders who were more active in organizing them and bringing them into contact with groups and organizations, or b) they had more opportunities because their communities were prioritized as work areas for government agencies and social movements interested in creating extractive reserves. Distance was not a significant variable for explaining respondents' propensity to have contact with groups and organizations. Although respondents living closer to the town of Lbrea were easier to visit and therefore may have had more opportunities to become members in these groups and organizations, a series of factors, including economic crisis and resource conflict, inspired riverine inhabitants to seek out groups and organizations in Lbrea regardless of the distance of their community from town. Age, Sex, and Education were not significant in any of the models. Further research and a larger sample size would be required to validate or refute the significance of these variables to influence the respondents' propensity to become members of groups and organization s.
262 Community Contact Hypotheses for Community Contact H1: Community Leaders have higher Community Contact scores than non leaders. H2: Men have higher Community Contact scores than women. H3: Younger respondents have higher Community Contact scores tha n older respondents. H4: Catholics have higher Community Contact scores than Evangelicals. H5: Respondents living inside the Extractive Reserves have higher Community Contact scores than respondents living outside the Extractive Reserves. H6: Respondent s with more education have higher Community Contact scores than respondents with less education. H7: Respondents living in communities further from the town of Lbrea have lower Community Contact scores than those living closer to town. Community Contact Results and Discussion Hypothesis #1: Community Leaders have higher Community Contact scores than non leaders. Result #1: The results of the five linear regression models reject the null hypothesis for H1. The Community Leader variable was highly signifi cant in all of the models (CC#1, #2, #3, #4, and #5). Respondents who were community leaders were more likely to have contact within their communities with a wider array of groups and organizations than respondents who were non leaders. This result agai n corroborates my observations that community leaders were increasingly assuming a "broker" position between their communities and the outside world. When local politicians, social movement groups, and other outsiders would visit these communities they wo uld generally seek out the community leader when they wanted to organize meetings or other activities in the community. This protocol
263 appeared to strengthen the position of many community leaders and invariably gave them many more opportunities to make co ntact with the groups and organizations in their communities than non leaders. Since community leaders were consistently invited to meetings outside the community they were more likely to have rapport with the individuals representing the groups and organ izations, increasing the likelihood that these individuals would seek them out when they visited the communities. Community leaders not only served as guides for individuals from the groups and organizations, who were generally not familiar with the river ine villages, but also validated the presence of these outsiders in the eyes of other community members, who might be suspicious of their presence. For these reasons, it would be surprising if being a community leader was not associated with the responden ts' tendency to have contact with a wider array of groups and organizations in the communities. Hypothesis #2: Men have higher Community Contact scores than women. Result #2: Sex was not associated with the respondents' tendency to have contact with grou ps and organizations. All five linear regression models failed to reject the null hypothesis for H2. I had assumed that men would be more likely to be considered the head of the household and would therefore be more likely to interact with the groups and organizations during their visits to the communities. The results of this analysis imply that this was not the case. Men and women appeared to have a similar tendency to interact with the groups and organizations when they arrived in the communities. On e possibility is that the extremely small size of the communities (most were twenty households or fewer) and the fact that they tended to be clustered together, meant that it was difficult to visit the community and not interact with almost everyone in the village.
264 However, during the meetings I attended in riverine communities I noticed a significant participation by women in community meetings and other group activities involving groups and organizations from outside the community. As previously mention ed, at the time of this research the head of the IBAMA office in Lbrea was a woman, as was the head of the office of the National Council of Rubber Tappers. These women made a point of including women in meetings and other activities. Perhaps the incre asing opportunities for women in government and social movement organizations have led to more opportunities for women in riverine communities to have contact in their communities with groups and organizations than in the past. Another possibility is that the geographic clustering of households that had once been dispersed in family rubber production areas on the seringal known as colocaes may mean that it is more likely that everyone will have contact with groups and organizations passing by than ever b efore. In the past a women may have stayed at home in the remote colocao while her husband traveled to a meeting. Now, with the clustering of the communities, many women can participate in meetings without being forced to travel far from their homes an d the domestic tasks in which they spend a great deal of time engaged. Therefore, I believe that women may have had more opportunities to participate in meetings and other activities outside the home than in the past. Hypothesis #3: Younger respondents have higher Community Contact scores than older respondents. Result #3: Age was not significant as an explanatory variable for Community Contact in any of the three regression models in which it was used. The results of the linear regression models failed to reject the null hypothesis for H3. It
265 appears that personal interest or some other motivation was more likely than age to determine whether or not individuals interacted with outside groups and organizations visiting their communities. Hypothesis #4: Catholics have higher Community Contact scores than Evangelicals. Result #4: Religion was not significant as an explanatory variable in both models in which it was used (CCC#1 and CC#3). The regression models failed to reject the null hypothesis. This result was surprising considering that Religion was a significant variable for explaining Recognition, Membership, and Contact. Perhaps the small size of the villages meant that once the groups and organizations arrived, Catholics and Evangelicals were j ust as likely to interact with them. It could also mean that many of the groups and organizations were just as likely to visit Catholic communities as they were to visit Evangelical communities. Hypothesis #5: Respondents living inside the Extractive R eserves have higher Community Contact scores than respondents living outside the Extractive Reserves. Result #5: The linear regression models rejected the null hypothesis for H5. Living inside an extractive reserve was highly significant as an explanator y variable for Community Contact in model CC#2 and highly significant in CC#4. My assumption, based on these results, is that families inside the reserves were a) more likely to have community leaders that sought out contact with groups and organizations to bring much needed assistance to their communities and/or b) were more likely to be contacted by groups and organizations that have opportunities to engage in group activities because government and social movement organizations interested in creating t he extractive reserves had prioritized their communities as places
266 they would work. I believe that the answer is a combination of the two. As I described in Chapter 5, community leaders actively sought out groups and organizations to try to resolve land tenure and resource conflicts. However, once the process of creating the extractive reserves was in motion, the government agencies and social movements interested in creating these reserves targeted the communities inside the proposed reserves as priorit y work areas and visited them more regularly than communities outside the reserves. Hypothesis #6: Respondents with more education have higher Community Contact scores than respondents with less education. Result #6: The linear regression models also fa iled to reject the null hypothesis for H6. Education was not significant in any of the three models in which it was tested. Hypothesis #7: Respondents living in communities further from the town of Lbrea have lower Community Contact scores than those li ving closer to town. Result #7: Community Distance to Lbrea was correlated with Religion and Living Inside an Extractive Reserve. For this reason, I only included it as an independent variable in model CC#5. Based on the results, this model rejects the null hypothesis for H7. The distance from the respondents' communities to the town of Lbrea was highly significant as an explanation for number of groups and organizations respondents had contact with in their communities. Community distance was also a significant variable in explaining the propensity of respondents to be members of groups and organizations. Due to the tremendous distances between the town of Lbrea and the villages where this research was conducted I am not surprised that inhabitants in remote areas had less contact with the groups and organizations in their communities than respondents living closer to
267 town. As I mentioned previously, social movement groups and government agencies all faced serious budget constraints that made it dif ficult for them to purchase the diesel fuel and other supplies that trips into the interior of Lbrea required. The average distance of the from Lbrea of the 26 villages 5 in which I conducted interviews was 204.7 kilometers, implying high fuel costs that many groups and organizations could not cover on a regular basis. Community Contact Summary Table 6 9 Community Contact Scores CC#1 CC#2 CC#3 CC#4 CC#5 R 2 0.150 0.235 0.179 0.308 0.259 F 3.698** 6.762** 4.510** 9.360** 5.652** Age B 0.024 0.006 0.064 Sex B 0.495 0.149 0.943 1.922 1.277 Religion B 1.825 2.470 Yrschl B 0.140 0.346 0.302 Comm. Leader B 6.134** 5.621** 6.161** 4.337* 5.322** Resex B 5.206** 6.602** Dist. to Lbr ea B 0.012** Table 6 9 shows that of the seven independent variables tested to explain the respondents' contact with groups and organizations in their communities, being a community leader was highly significant in all five models. In Chapter 5 I discussed how community leaders were becoming more and more important as brokers between their communities and the outside world. It is therefore no surprise that they would be more likely than non leaders to be in contact with groups and organizations Having contact with a wide array of groups and organizations may have strengthened their position as 5 These distances were measured using a GPS unit and include the meanders of the rivers. I did not have the distance for one of the communities in which I conducted interviews, so the average may actually be a bit higher than the number presented here. For the complete table of community distances fro m Lbrea please the table in Appendix Table A 9.
268 leaders in the community by enabling them to bring resources into the community. When individuals from social movements and government agencies visited the communities they usually made it a point to contact community leaders first upon arriving, thus ensuring that they would have more contact than non leaders. Community leaders often acted as guides for outsiders visiting their remote communities and w ere the ones invited to participate in meetings in the city, both of which assured that community leaders would have more contact than non leaders. Religion was also a moderately significant in explaining the respondents' propensity to have contact with groups and organizations in their communities. In contrast to my hypothesis, Evangelicals were actually more likely than Catholics to have contact with groups and organizations. In Chapter 5 I discussed how Evangelical communities played a significant ro le in the creation of the extractive reserves in Lbrea, which is contrary to the notion that Evangelicals do not engage in politics or social movement activities. Evangelicals in Lbrea showed a greater propensity to organize themselves and strive to imp rove their living standards than Catholics. The result of these Community Contact regressions seems to corroborate these observations. Living inside the extractive reserve was also significant for explaining the respondents' tendency to have contact wit h groups and organizations. This may be because they: a) had leaders who were more active in organizing them and bringing them into contact with groups and organizations, or b) they had more opportunities because their communities were prioritized as work areas for government agencies and social movements interested in creating extractive reserves. Community leaders sought
269 out government agencies and social movements to get the extractive reserves created, and many of these same social movements and gover nment agencies prioritized communities inside the reserves when they made trips into the interior. The mutual interest of creating extractive reserves that some community leaders and some social movements and government agencies had made it more likely th at reserve residents would have contact with these groups and organizations than non residents. Distance was a highly significant variable for explaining respondents' propensity to have contact with groups and organizations in their communities. Social movement groups, religious organizations, and government agencies in Lbrea all face budget constraints that limited their ability to travel to remote villages along the Purus and Ituxi Rivers. For this reason, it is not surprising that the inhabitants of the more remote communities received fewer visits than communities closer to town. Age, Sex, and Education were not significant in any of the models. Further research and a larger sample size would be required to validate or refute the significance of th ese variables to influence the respondents' propensity to become members of groups and organizations. Who are the Patrons? One of the key research questions of this doctoral study was to analyze the changes that have taken place in the patron client social relations that became institutionalized in many parts of the Amazon during the rubber boom. In Chapter 3 I described the evolution of patron client relationships in Lbrea that were institutionalized as part of the seringal production unit. I discussed the expectations of "generalized exchange" between patrons and clients that developed and how they changed over time in response to the socioeconomic changes that affected extractive industries in the Amazon. In Chapter 3 I described the economic decline that led many
270 of the "traditional" seringal owners to abandon the interior, in many instances leaving their former clients to form new relationships in order to access vital resources. In Chapter 4 I showed how seasonal production cycles, exclusion from "formal" credit, and a lack of basic services put many of the inhabitants of riverine communities in Lbrea in a precarious position that often necessitated the formation of relationships that would enable them to access "informal" sources of credit. The "informal" nature of these arrangements, without formal contractual obligations, required other forms of trust, potentially based on patron client relationships. In Chapter 5 I discussed how the inhabitants of riverine communities began to build relation ships with government agencies and social movements to try to improve their lives in the wake of the collapse of the old seringal system. In Chapter 2 I summarized some of the extensive body of literature regarding patron client relationships, in essence the etic perspective of what these relationships mean. However, this analysis would be incomplete without the perspective of those who lives had long been enmeshed in these complex and often contentious relationships. How did the riverine inhabitants of communities in Lbrea understand patron client relationships? How had their understanding of these relationships changed in the wake of the socioeconomic changes that had impacted their lives so profoundly in the past thirty years? In this section I pre sent the results of my attempt to uncover the emic understanding of the patron client relationships that existed in Lbrea in 2008 2009 based on extensive interviews with patrons, clients, government officials, social movement participants, and of course t he riverine inhabitants.
271 The Patro in the Eyes of Community Inhabitants During the nearly two years that I spent in Lbrea doing the field research for this dissertation I conducted interviews in several different formats and engaged in countless informa l conversations to try to better understand the local understanding of patron client ties. In the following section I consolidate excerpts from these interviews and conversations to present these local understandings of the concept to the reader. Since th e word patro had so many meanings I decided to ask direct questions that would enable me to find out who my interview respondents considered to be patrons. It is worth noting that when I asked respondents which of the groups and organizations they recogn ized and which they did not, all of the respondents recognized the patro ." This confirms that this word is widely used in Lbrea and is not simply an esoteric term associated with academia or with the increasingly forgotten rubber boom period. However in spite of respondents' universal recognition of the term patro ," there were different ways of understanding its meaning. Considering the history of the Brazilian Amazon and the extensive literature regarding the rubber boom, which seems to dominate much of the historical texts about the region, it is easy to assume that the word patro is the equivalent of seringalista or rubber estate owner. However, socioeconomic changes in the past thirty years appear to have altered the way in which many people used and understood the word in Lbrea. Understanding the Term Patro in Lbrea As I conducted the ninety six position generator interviews in which I asked respondents five questions about groups and organizations with which they might interact, I trie d as often as possible to ask follow up questions that would help me to
272 understand how these interactions were taking place. For example, when people responded that they had contact with the patro ," I would ask them who their patron was, what they did f or a living, and other questions regarding their interactions with that person. These were open questions and respondents were not given options from which to choose. I have combined the answers that were similar in the tables below, to show different wa ys in which people understood the concept of the patro ." In the following tables I present a breakdown of respondents' answers to my questions regarding their interactions with patrons. Power Twenty nine out of ninety six, or 30%, of respondents provide d qualitative responses to my follow up questions after asking them to rate the power of the patrons. Table 6 1 0 Who are the patrons? (Power) # of Respondents % of Responses Type of Response 11 38% Patron no longer exists 11 38% Response related to pat ron's power 4 14% Patron buys, sells, or provides credit 3 10% Patron is a river trader or shop owner 29 100% Total In Table 6 10 I present the follow up answers of those who gave responses regarding the amount of power held by the patron. They gener ally rated it as being low compared to what it once was. Those who said the patron no longer existed were generally older respondents for whom the term meant seringalista or some other term associated with the seringal rubber production system. During in terviews I noticed that buying, selling, and exchange seemed to be a common characteristic in determining whether or not someone was considered a patron. In this case if we combine the
273 responses that have this characteristic (Buys, sells, provides credit and patron equals river trader or shop owner) we see that 24% of the responses share this feature. Contact Forty six of the ninety six respondents, or 48%, provided follow up comments when asked about their contact with patrons. A summary of these respo nses is presented in Table 6 11 Table 6 1 1 Who are the patrons? (Contact) # of Respondents % of Responses Type of Response 30 65% Patron is same as river trader or shop owner 6 13% Patron no longer exists 4 9% Patron buys, sells or provides credit 2 4% Patron is president of ASPACS association 2 4% Patron is someone who claims the land they occupy 1 2% Patron is a city councilperson 1 2% Unrelated answer 46 100% Total Most of the respondents who answered the follow up questions to the amount of contact they had with patrons said that for them the word is synonymous with river trader ( regato ) or shop owner ( comerciante ). This seems to confirm that with the collapse of the seringal system and the elimination of the barraco or trading post as a source of supplies, respondents had formed relationships with river traders and store owners who bought their agricultural and extractivist production in exchange for supplies. Those who said the patron no longer existed were once again older residents who associated the word with the seringalistas and rubber barons that had abandoned the area. Two respondents named the president of the ASPACS association as their patro It is worth noting here that of the four associations respondents were asked abo ut (APADRIT, APACJG, ASPACS, and AMIMP) ASPACS is the only one that had
274 the business relationships, market access, and working capital to buy and sell with riverine inhabitants. The other associations had focused on community organizing, accessing governm ent projects, and helping members obtain government benefits. Two respondents said the patron they had contact with was someone who claimed that the land they lived on belonged to them as part of an old seringal property and in one case was still demandin g a monthly rent from the inhabitants. One person said their patron was a city councilperson that had provided travel expenses and money during an illness. Again if we combine the responses that were related to buying, selling, and exchange we see that for 78% of the people who provided follow up answers a patron is someone from whom they buy and sell. Community Contact Thirty three respondents, 34% of the total sample, provided follow up responses related to the question about whether or not they had contact with the forty four groups and organizations in their communities. Table 6 12 is a summary of these responses. Table 6 1 2 Who are the patrons? (Community Contact) # of Respondents % of Responses Type of Response 26 79% Patron is same as river tra der or shop owner 3 9% Patron buys, sells or provides credit 2 6% Patron no longer exists 1 3% Patron is president of ASPACS association 1 3% Patron is someone who claims land they occupy 33 100% Total The majority of those who provided follow up re sponses said that the patro was the same as river trader or shop owner. A couple said the patron no longer exists, again referring to the seringalistas that had abandoned the area.
275 Once again if we combine the answers associated with buying, selling, a nd credit we can see that this was a defining characteristic of being a patron for 91% of those that provided follow up answers. It was apparent that the way most of the inhabitants understood the word patro was still greatly influenced by the interact ions that had taken place in the barraco trading posts that were so characteristic of the forest products' economy that dominated in the region for so long. Inhabitants appeared to have a much more specific understanding of the type of interactions that constituted being a patron than the academic literature on the subject. The ability to control and provide access to supplies was the dominant response for what constituted a patron. Resource Generator Interview Results In addition to the ninety six pos ition generator interviews, I did six resource generator interviews with community inhabitants to try to determine how they used and understood the word patro ." The resource generator is a data collection instrument in which respondents are asked to pro vide information about whom they go to in order to obtain a list of items (van der Gaag & Snijders, 2005) Their responses to these questions are used to determine their social capital (van der Gaag & Snijders, 2005) I developed a modified version of this tool, not to measure respondents' social capital, but instead to determine what kinds of people they believed were "patrons." First, respondents were asked from whom they would borrow seven different items: 1) chainsaw, 2) outboard motor, 3) one liter of fuel, 4) fifty liters of fuel, 5) one kilo of manioc flour, 6) fifty liters of manioc flour, and 7) supplies to harvest Brazil nuts. For each of the seven questions respon dents were asked: 1) what they would give in exchange for borrowing the item, 2) what their relationship was to the lender (friend, family, acquaintance), 3) what the lender did for a living, and 4) if the person they listed
276 could not or would not loan the item, if there was someone else they would ask. The seven questions were then repeated but respondents were asked from whom they would buy these items instead of from whom they would borrow them. Four follow up questions were asked for each of the "buy ing" questions also. Next, respondents were asked to whom they would sell certain items. 1) 100 liters of manioc flour, 2) 100 kilos of Brazil nuts, 3) 100 liters of copaiba oil, and 4) 100 kilos of rubber. For each of these questions they were asked fo ur follow up questions: 1) what would you receive in exchange for the item, 2) what their relationship was to the buyer (friend, family, acquaintance), 3) what the buyer did for a living, and 4) if there was someone else they would sell to if that person d id not want to purchase the item. Respondents were next asked five questions about other types of assistance they might receive. They were asked to whom they would go: 1) if they needed help with documents to access government programs like retirement be nefits, Bolsa Familia etc., 2) if they needed help to resolve a conflict, 3) if they needed information about employment opportunities, 4) if they needed assistance with legal matters, and 5) if they needed information about the prices of the items they b uy and sell. Once again, for each of these questions they were asked the four follow up questions they had been asked for the "borrowing," "buying," and "selling" questions. Once they had finished answering these questions and had provided a list of nam es of the people they would borrow from, buy from, sell to, and receive certain kinds of assistance from I went down the list of the names and asked for each name if they would call that person a patro or not. The results of this questionnaire are presen ted in Appendix Table A 4.
277 The results of the table are similar to those from the follow up questions of the position generator questionnaire. In the column on the far right respondents overwhelmingly considered "patrons" to be people that: 1) provided credit, 2) bought large quantities of the items they produced (like manioc flour), or 3) sold them the tools and supplies they needed. In most cases patrons were considered to be acquaintances. It was less common that they were considered friends and hig hly unlikely that they would be relatives. This seems to agree with the notion that patrons were usually from different social groups than their clients and the patron client relationship represented a "bridging tie." As in the case with the position gen erator interviews, buying, selling, and credit seemed to determine whether or not someone was considered a patron. Individuals who provided legal advice, employment information, help with paperwork, and help resolving conflicts were generally not consider ed to be patrons. The results of this section show the divide between emic and etic understandings of the word patron. The etic understanding presented in Chapter 2 encompasses a great many situations and types of interaction that may constitute a patro n client relationship. In Lbrea the term was narrowly used to refer to interactions involving buying, selling, and providing credit. In spite of the many changes that had occurred in the social networks of riverine inhabitants since the collapse of the seringal system, their exclusion from formal sources of credit, dependence on seasonal agro extractive production activities, and other factors mean that "patrons" are still very much a part of their daily existence. These findings also show that many of the groups and organizations that have replaced the "traditional" patrons as sources of key resources
278 are not called "patrons" by the riverine inhabitants themselves. In this sense the long history of the cultura do barraco seems to have shaped the way people use and understand the term patron. The etic version includes a wide array of exchanges and relationship types, while the emic understanding of the word in Lbrea is confined to certain types of market exchanges. Chapter Summary This chapter des cribes the contacts to whom riverine inhabitants potentially could go to in order to access resources and services. In Lbrea this included a mix of more than forty groups and organizations that included community associations, government agencies, social movements, and religious organizations. This chapter provides a quantitative analysis of the interactions between these groups and organizations and riverine communities. This chapter also describes the resources these groups and organizations controll ed and the perceptions of riverine inhabitants regarding their propensity to provide these resources and services. The data show that although most of the groups and organizations were rated rather low in this regard, riverine inhabitants were accessing r esources from a wide range of these groups and organizations. This result corroborates my assertion that multi hub social networks comprised of a multitude of actors had replaced the dyadic patron client relationships that had become entrenched during the rubber boom. This chapter shows how variability in riverine inhabitants' networks was associated with their access to information about extractive reserves, but did not explain why some inhabitants were wealthier than others. The characteristics that b est explained network extensity among riverine inhabitants in Lbrea were also explored in this chapter.
279 Community leaders, Evangelicals, and extractive reserve inhabitants were more likely to have extensive social networks. This result confirms my obser vation that community leaders were playing an increasingly important role in the social structures that were replacing the patronage pyramids established during the rubber boom. It also shows that in spite of the long history of mobilization by the Cathol ic Church in Lbrea, Evangelicals were in fact more engaged in many social organizing activities than Catholics. Finally, it shows the two way nature of these "new" networks, in which community leaders were forming relationships in the effort to bring imp rovements to their communities, but were also being sought out by government agencies and social movements interested in creating extractive reserves, or types of development work. This chapter also reveals the very specific characteristics that constit ute being a "patron" in the eyes of riverine inhabitants in Lbrea. Buying, selling, and controlling credit and supplies were what constituted being a patron in the local conceptualization of patronage. River traders and storeowners were the actors that had taken over these activities in the wake of the collapse of the seringal system.
280 CHAPTER 7 DISCUSSION: POST SERINGAL PATRONAGE The case studies in Chapter 5 show how social structures in Lbrea, based on the patron client relationships that were institutionalized during the rubber boom, changed dramatically between the 1940s and the 2000s. As we can see in Figure 7 1 1 below, supplies once came to the seringal through the owner's contacts with merchant houses, who in turn often had ties to large b anks. This is an example of a patronage pyramid, in which a patron uses his influence to maintain control of a series of clients (Martin, 2009) Figure 7 1 Seringal Patronage Triangle Seringal owner patrons were able to keep their tapper clients by providing supplies on credit on an exc lusive basis. In the event that the tapper sold rubber to someone else, such as a river trader, the patron threaten to invoke sanctions, such as restricting credit or evicting the tapper and his family from the seringal Priests, government 1 This figure was inspired by the patronage pyramid figures in (Martin, 2009)
281 agents, polit icians, and others respected the seringal hierarchy and would seek out the patro upon arriving at the seringal Economic collapse, particularly the decline of the rubber industry, made it difficult for seringal owners to provide supplies on credit and t hey were unable to maintain their land tenure claims. Patrons could no longer maintain exclusive exchange with their clients because they could not provide credit and could not threaten to evict them from the land. They were no longer able to enforce the "one patron" rule, which is considered a key element of maintaining patronage pyramids (Martin, 2009) A new multi hub structure began to emerge in which community inhabitants began to seek out a myriad of actors to obtain resources and services. Figure 7 2 below shows my impression of how soci al networks were changing into more multi hub structure in which community leaders began to assume key roles as brokers between the village inhabitants and the outside world. Figure 7 2 Post Seringal Multi Hub Network
282 In Chapter 2 of this dissertation I summarized some of the extensive body of literature regarding patron client relationships, particularly the elements that encourage the formation of these relationships, such as inequality. In Chapter 3 I described the evolution and institutionalization of patronage in the rubber economy of Lbrea. I showed how the arrangements between patrons and clients changed over time based on economic conditions and the ability of patrons to control and provide access to scarce resources. I showed how the arrival of "new" actors, particularly government agencies and social movements, provided alternative options for accessing resources that did not exist in the past. In Chapter 4 I provided information about the conditions that still excluded many riverine inhabi tants in Lbrea from participating in the types of relationships that were replacing "traditional" patronage. In Chapter 5 I presented three case studies revealing how three riverine communities in Lbrea responded to the changes in social structures by a ctively forming relationships with brokers and intermediaries who could help them after the departure of seringal patrons. In Chapter 6 I presented the results of survey designed to determine which of the "new" actors were providing resources and what the perceptions of riverine inhabitants were regarding their propensity to provide assistance. The results of this survey show how riverine inhabitants' networks changed to multi hub networks in which they no longer depended on a single patron. Significant c hanges in Lbrea's socioeconomic realities have led to changes in the social networks of inhabitants of the interior of the municipality. The following socio economic changes may have created the conditions necessary to change patron client ties developed on the seringal system: 1) economic collapse of extractivist industries
283 and the seringal production system; 2) rural/urban migration in which patrons and clients abandon the interior; 3) the rise of social movements that challenge the inequality of patron client ties and unequal land distribution; 4) the establishment of land reform / conservation units (Terra Indigena, Reserva Extrativista, Reserva de Desenvolvimento Sustentavel, etc.) designed in part to break patron client ties and ensure the continuati on of "traditional" livelihoods; 5) the creation of government programs to provide credit to small rural producers as well as social assistance programs, like Bolsa Familia ; 6) the establishment of government agencies and social movement organizations that actively worked with communities to set up associations designed to provide them with an alternative to traditional patron client ties; 7) the shift in focus of government agencies that led to government funding for programs, like the rubber subsidy prog ram, to be channeled through community associations instead of through patrons; 8) access to personal modes of transportation, particularly the 5.5 HP rabeta motors, enabled community members to reach cities, like Lbrea, to sell their products and buy mer chandise without having to row for days at time; 9) improvements in communication, such as two way, transistor radios, and even satellite dishes facilitated communication between communities and enabled people receive national and international news; 10) a return to democratic government in Brazil after 21 years of military dictatorships and articles in the 1988 constitution that specifically guaranteed the rights of indigenous peoples and other minority groups provided legal recourse that did not exist at the height of the seringal system. In spite of the collapse of the seringal system, the emergence of social movement organizations, and migration to urban areas there was still the potential for patron client
284 ties to exist because: 1) people still needed supplies in their remote communities; 2) there were still some people in positions of power and many people in positions of powerlessness, 3) some individuals in the society have more access to information, loans, jobs, capital, etc., than others, 4) the s ociety in many parts of the interior of the Brazilian Amazon was still quite hierarchical with little room for social mobility; 5) people lacked education, and illiteracy was a significant barrier to accessing government programs, health care, employment, etc.; 6) the associations, social movement activities, and government programs were not completely successful in helping people to access markets, transport their goods, etc. which left many with no choice but unequal trade relationships; 7) many people st ill lacked even the most basic citizenship documents, many of which are required to participate in government programs, obtain loans, etc.; 8) most of the inhabitants of the Purus and Ituxi Rivers did not have land title to areas they use and live on, whi ch in addition to providing security against being forced off the land, are often essential as collateral in the process of obtaining loans. Although the patron client systems that characterized the seringal system had disappeared for the most part, othe r kinds of patronage were alive and well. Political patronage in local government and during election cycles was ubiquitous and included everything from small favors in exchange for loyalty to the purchase of votes for money, food, and equipment. As com munity and association leaders gained more power as the key brokers between community members and the rest of society, the potential that these relationships will develop into patron client arrangements also exists. As Martin (2009) has pointed out, although a multi hub network may be more egalitarian, it may take
285 more effort to access resources through many patrons than through one Finding one patron who can provide the same resources and services may be more efficient, which may explain why patronage pyramids can persist or resu rface (Martin, 2009) Patronage in the B razilian Amazon has survived previous socioeconomic upheaval and it remains to been seen whether or not the changes of the past thirty years will lead to its final demise. The events that have transpired in Lbrea chronicled in this dissertation were not isolated incidents, similar processes were occurring throughout the Brazilian Amazon, in other parts of Brazil, other Latin American countries, and other regions of the world. Many of the larger forces that were leading to changes in social structures in Lbrea (rural economic transitions, rural/urban migration, land use change, and resource conflict) were also occurring throughout the Amazon and in other developing, natural resource rich regions of the world. Although a great number of the inhabitants o f many of these areas joined the migration flow to urban areas, in many of these areas, just as in Lbrea, others had decided to stay in rural areas (or return from the cities). Those who stayed often faced the same types of challenges facing riverine inh abitants in Lbrea, which often included economic transitions that also produced transitions from a set of established social interactions to new social systems that inhabitants had to negotiate. In the Brazilian Amazon, and other parts of Brazil, this ha s often implied a transition from social systems based on multiplex dyadic patron client relationships to new social arrangements in which resources and services are obtained from a multitude of actors.
286 Throughout the Brazilian Amazon, and in other regio ns of Brazil, many of the individuals who were negotiating these social transitions gathered together, sometimes with the help of other groups, organized themselves, and formed social networks they hoped would help them improve their living standards, reso lve resource use conflicts, and achieve other goals. The emergence of community leaders and formation of networks with governmental, religious, and social movement groups and organizations described here was not unique to Lbrea and was in fact occurring in many areas throughout the region, and throughout Brazil (A. W. B. d. Almeida, 2008) In some cases the result of these community organization and networking activities were laws at the municipal, state, and even federal levels that (at least on paper) es tablished land tenure security and use rights for many of these communities, like the riverine and indigenous peoples in Lbrea, that long occupied these areas (A. W. B. d. Almeida, 2008) In other cases, as seen in Chapter 5 of this dissertation, "new" soc ial networks enabled communities to secure land tenure stability in the form of extractive reserves. In many cases the process of creating these areas involved struggles and the formation of social networks similar to those described in this dissertation (M. W. B. d. Almeida, 1992; Sant'Ana Jnior, Pereira, Alves, & Pereira, 2009) Similar processes were taking place in neighboring South American countries and throughout the region (A. Escobar, 2008; Little, 2001) The socioeconomic changes that riverine inhabitants were struggling to overcome by forming "new" social networks to replace more entrenched types of patron client relations are often attributed to the forces of globalization and are not unique to this area. The responses of riverine communities to these social and
287 economic forces may be helpful for understanding how isolated communities in remote areas in other parts of the world may respond to similar circumstances.
288 APPENDIX ORGANIZATION INFORMA TION
289 Table A 1. Organization Information Name Type Mem. Org. (Y/N) Office Lbre a (Y/N) Year Founded in Brazil Year Established in Lbrea River Trader Traditional No Yes NA NA Patron Traditional No Yes NA NA Shop Owner Traditional No Yes NA 1871 Prelacy Religious No Yes No data 1925 Priest Religious No Yes NA 1873 IEADLAM Religi ous No Yes 1911 No data Pastor Religious No Yes No data No data Mayor Municipal No Yes No data 1886 Councilmen Municipal No Yes No data 1886 Municipal Sec. Municipal No Yes No data No data Citizen's Center Municipal No Yes 2006 2006 Doctor Basic Ser vice No Yes NA No data Dentist Basic Service No Yes NA No data Professor Basic Service No Yes NA No data IDAM State No Yes 1996 1996 ADS State No Yes 2007 2007 ITEAM State No No 2003 N/A SEARP State No No 2007 N/A UEA State Yes Yes 2001 2008 I BAMA Federal No Yes 1989 No data ICMBio Federal No Yes 2007 2009 INCRA Federal No Yes 1970 1979 Federal Police Federal No No 1964 N/A Military Police Federal No Yes 1834 1887 Army Federal No Yes 1822 1999 FUNASA Federal No Yes 1999 1999 FUNAI F ederal No Yes 1967 1981
290 Table A 1. Continued Name Type Membership (Y/N) Office Lbrea (Y/N) Year Founded in Brazil Year Established in Lbrea MMA Federal No No 1992 N/A APADRIT Association Yes Yes 1997 1997 ATAMP Association Yes Yes 2005 2005 APACJG Association Yes Yes 2003 2003 ASPACS Association Yes Yes 1997 1997 AMIMP Association Yes Yes 2003 2003 CNS Soc. Mov. No Yes 1985 2004 CPT Soc. Mov. No Yes 1975 No data GTA Soc. Mov. Yes Yes 1992 1999 MMC Soc. Mov. Yes No NA N/A OPIMP Soc. Mov. No Yes 1995 1995 Rural Union Soc. Mov. Yes Yes 1979 1979 Children's Pastoral Soc. Mov. No Yes 1995 1993 CIMI Soc. Mov. No Yes 1972 1978 IEB Soc. Mov. No Yes 1998 2008 MMTRL Soc. Mov. Yes Yes 1997 1997 Grileiro Other No No NA NA
291 Table A 2. Organization Assistance Name Assistance they said they provide River Trader No information. Patron No information. Shop Owner Sell goods to people. Prelacy Guide youth in the community, ammeliorate social problems, a ssist with housing, try to help with health care. Priest Help people with housing program, education, youth training, food, clothing. IEADLAM Take collections to help people with problems. Pastor Help with some donations. Try to avoid giving people things. Mayor No information. Councilmen Help communities with basic services, assistance with travel expenses, medicine, bills, Municipal Sec. (Sec. of Environment) Provide documents to cut timber for homes. IPAAM does documents fo r commercial timber. Citizen's Center Provide documents and access to social programs (Bolsa Familia). Doctor Provide health care. Dentist Provide dental care. Professor Provide education and study materials. IDAM Provide agricult ural credit, implements, livestock management, seeds. ADS Provide technical assistance, documents, receipts, implements for extractivism. ITEAM Legalization of land titles with municipal cartorios. SEARP Communication broker between local soc.move. And other gov. agencies. Documents, find funding, etc. UEA Provide education. IBAMA Implement environmental legislation prevent environmental crimes. ICMBio No information. INCRA Agricultural credits and documentation. Federal Police No information. Military Police Enforce laws, monitor elections. Army Monitor elections, train young people FUNASA Provide health care, Mainly to indigenous but also some to riverine people. FUNAI Provide production ma terials, help with documents, provide access to social programs, help with production chains. MMA Technical Assistance, Documents, Program Financing APADRIT Help with documents, transportation, clothing, food.
292 Table. A 2 Continued Name Ass istance they said they provide ATAMP Help resolve land and resource conflicts. Broker between communities and government agencies and social movements. APACJG Help people with documents, governmet benefits, crop improvements, rubber subsidy progr am, provide rubber tappping implements. ASPACS Buy peoples' production (Brazil nuts, agricultural products, andiroba, etc.) AMIMP Help with organization in indigenous communities. CNS Help resolve land and natural resource conflicts. Health issues, provide information, link people to other organizations. CPT Help resolve land and natural resource conflicts. Lake preservation and volunteer environmental monitor programs. GTA Helps with organization, logistical support, help finding funds for association projects. MMC No information. OPIMP Help indigenous communities with problems and with implements and supplies. Rural Union Help with documents and receipts. Documents that help provide access to government programs. Chi ldren's Pastoral Provide food and "natural" medicine for children, information about pre natal care, child care, hygiene. CIMI Help indigenous people get health care, education, resolve land issue and conflicts, monitor government Indian agencies. IE B Strengthen local organizations, community associations. MMTRL Help with documents. Grileiro
293 Table A 3 Resources Provided by Groups and Organizations in Lbrea Name Credit Supplies Markets Health Educ. Jobs Transport Info. /Advice Docs. Programs Protection Religion Resex Patron 23% 8% 11% 1% River Trader 8% 21% 14% Shop Owner 15% 13% 6% Prelacy 1% 3% 8% 8% 5% 2% 15% 2% Priest 1% 3% 3% 12% 11% 5% 31% 5% Assem. of God 1% 4% 1% 4% 1% 23% Pastor 1% 3% 1% 8% 3% 31% Mayor 9% 7% 14% 67% 16% 9% Councilperson 4% 12% 12% 8% 1% 2% 2% Municipal Sec. 2% 7% 50% 33% 4% 1% Citizen's Cent er 3% 1% 4% 4% 2% 40% 10% Doctor 3% Dentist 2% Professor 32% IDAM 31% 8% 5% 19% ADS 1% 11% 3% 5% 7% ITEAM 4% 2 % SEARP IBAMA 4% 6% 4% 1% 26% 7% ICMBio 1% 9% 9% 12% INCRA 1% 7% 5% 5% 6% 14% Federal Police 2% 13% Military Police 20% Army 1% 7% 2% 11% FUNASA 1% 32% FUNAI 5% 4% MMA 3% 2% 2% 7%
294 Table A 3. Continued Name Credit Supplies Markets Health Educ. Jobs Transport Info./Advice Docs. Programs Protection Religion Resex ATAMP 8% 7% 3% 4% 12% APACJG 6% 1% 8% 4% 4% 4% 5% ASPACS 6% 3% 22% 1% 8% 2% APADRIT 1% 4% 1% CNS 1% 2% 11% 12% 2% 10% 19% CPT 2% 6% 4% 4% 3% 4% 12% GTA 14% 2% 10% 2% MMC OPIMP 2% Rural Union 23% 3% 2% Child's Past. 12% 11% 2% CIMI 2% Grileiro Totals (N) 71 121 36 95 28 3 25 89 57 203 54 26 43
295 Table A 4 Who are the Patrons? Patron Client Resource Generator Lbrea, AM, Brazil 2009 N=6 Do you know someone who would % "Yes" If "yes" what is your relationship with them? Relative Friend Acquaintance Neighbor Are they a patron? (%Y) 1. Loan you a chainsaw 50 33 67 0 0 0 2. Loan you a boat motor 100 16.6 16.6 16.6 50 0 3. Loan you 1lit of fuel 100 16.6 33 33 16.6 0 4. Loan you 50lit of fuel 83 0 16.6 50 16.6 16.6 5. Loan you 1kil of manioc flour 100 33 33 0 33 0 6. Loan you 50k of manioc flour 100 16.6 33 16.6 33 0 7. Loan you credit for Brazil nut supplies 83 0 40 60 0 100 8. Sell you a chainsaw 100 0 50 50 0 100 9. Sell you a boat motor 100 0 33.5 66.5 0 83 1 0. Sell you 1lit of fuel 100 0 33 50 17 67 11. Sell you 50lit of fuel 100 0 33.5 66.5 0 83 14. Buy 100kil of manioc flour from you 100 0 33.5 66.5 0 67 15. Buy 100kil of Brazil nuts from you 100 0 33.5 66.5 0 100 16. Buy 100lit copaiba/andiroba from yo u 100 0 17 83 0 100 17. Buy 100kil of rubber from you 100 0 17 66 17 67 18. Help you with paperwork 67 0 0 75 25 0 19. Help you resolve problem/quarrel 83 0 0 100 0 0 20. Provide you employment information 33 0 50 0 50 0 21. Give you legal advice 83 0 0 80 20 0
296 Figure A 1 Map of Lbrea
297 Figure A 2 Map of Deforestation
298 Figure A 3 Map of Conservation Units Lbrea
299 Below are the tables of the bivariate X 2 analyses conducted before I combined the categories that did not have h igh enough values in some of the cells. Table A 5 Network Variability and Contact Bivariate X 2 Contact Group Type ResexInfo Rabeta Motor Manioc Roaster Wealth Traditional X 2 4.798 .662 2.495 5.284 Religious X 2 11.781 3.317 5.220 10.907 Municipal X 2 18.601 13.343 1.055 8.667 Basic Service Provider X 2 3.626 1.576 2.324 12.142 State X 2 22.354* 6.134 3.727 12.954 Federal X 2 29.668 6.274 8.582 24.589 Associations X 2 29.653** 5.132 2.771 14.351 Social Movements X 2 36.815* 3.328 4.299 23.919 Table A 6 Network Variability and Community Contact Bivariate X 2 Community Contact Group Type ResexInfo Rabeta Motor Manioc Roaster Wealth Traditional X 2 17.679* 4.157 2.505 15.582 Religious X 2 14.582 3.938 1.502 11.968 Municipal X 2 29.838** 4.395 3.854 6 .948 Basic Service Provider X 2 15.544 .217 5.190 7.630 State X 2 29.372** 4.909 1.471 9.841 Federal X 2 38.273* 2.014 11.670 23.242 Associations X 2 33.205** 2.171 3.729 20.403 Social Movements X 2 36.510* 3.503 3.548 16.405
300 Table A 7 Network Varia bility and Contact Weighted by Assistance Bivariate X 2 Contact Weighted by Assistance Group Type ResexInfo Rabeta Motor Manioc Roaster Wealth Traditional X 2 20.667* .058 6.389 13.778 Religious X 2 5.739 4.451 2.317 7.912 Municipal X 2 10.188 4.749 5.5 59 13.540 Basic Service Provider X 2 7.252 3.193 6.559 9.494 State X 2 28.399** 2.738 3.449 7.156 Federal X 2 26.741 3.818 7.480 23.267 Associations X 2 26.308* 10.088* 4.262 16.501 Social Movements X 2 30.100* 1.930 4.346 18.453 Table A 8 Network Variab ility and Community Contact Weighted by Assistance Bivariate X 2 Community Contact Weighted by Assistance Group Type ResexInfo Rabeta Motor Manioc Roaster Wealth Traditional X 2 21.358* 1.655 3.665 12.412 Religious X 2 8.506 7.865 2.078 13.209 Municip al X 2 15.250 1.629 1.958 10.308 Basic Service Provider X 2 17.091* 1.241 7.211 10.113 State X 2 35.983** 5.966 1.151 8.554 Federal X 2 38.666* 3.930 8.469 26.104 Associations X 2 23.622** 4.260 3.527 9.670 Social Movements X 2 43.664** 6.462 12.270* 22.522
301 Table A 9 Village Distances (Research Sites) From To River Distance Kilometers Lbrea Praia de Lbrea Purus 4.7 Boca do Ituxi Purus 12 Goiaba Ituxi 196.7 Palmapi Ituxi 207.4 Cabeudo Ituxi 210.8 Floresta Ituxi 220.2 Jarusale n Punicici No Data Volta do Bucho Ituxi 225.3 Praia Alta Ituxi 238 Pouso do Camaleo Siriquiqui 253.5 Cachoeira Vai e Volta Siriquiqui 310.2 Vila Vitria Ituxi 314.7 Nova Esperana Ituxi 327.4 Maciari Purus 14.6 Cassian Pu rus 21.4 Jucuri Purus 26.9 Buraco Purus 44.9 Santa Rosa Purus 54.3 Jurucu Purus 73.4 Bananal Purus 86.6 Santa Cndida Purus 238 Vila Acim Purus 289.1 Vrzea Grande Purus 355.1 Bela Rosa Purus 408.3 Ajuricaba Purus 4 76.7 Vila Limeira Seruini 507.1
302 LIST OF REFERENCES Adriance, M. C. (1995). Promised Land: Base Christian Communities and the Struggle for the Amazon Albany, NY: State University of New York Press. Akers, C. E. (1914). The Rubbe r Industry in Brazil and the Orient London: Methuen & Co. Ltd. Allegretti, M. (2002). A Construo Social de Polticas Ambientais: Chico Mendes e o Movimento dos Seringueiros. Unpublished Tese de Doutorado, Universidade de Braslia, Braslia, Brasil. Alme ida, A. W. B. d. (2008). Terras de Quilombo, Terras Indgenas, "Babauais Livres", "Castanhais do Povo", Faxinais e Fundos de Pasto: Terras Tradicionalmente Ocupadas Manaus: PGSCA UFAM. Almeida, M. W. B. d. (1992). Rubber Tappers of the Upper Juru River, Brazil: The Making of a Forest Peasant Economy. University of Cambridge, Cambridge. Aramburu, M. (1994). Aviamento, modernidade e ps modernidade no interior amaznico. Revista Brasileira de Cincias Sociais, 9 (25), 82 99. Barham, B. L., & Coomes, O. T. ( 1996). Prosperity's Promise: The Amazon Rubber Boom and Distorted Economic Development (Vol. 34). Boulder, CO: Westview Press. Bearfield, D. A. (2009). What is Patronage? A Critical Reexamination. Public Administration Review, 69 (1), 64 76. Benatti, J. H., Surgik, A. C. S., Treccani, G. D., McGrath, D. G., & da Gama, A. S. P. (2005). A Questo Fundiria e o Manejo dos Recursos Naturais da Vrzea: Anlise para a Elaborao de Novos Modelos Jurdicos Manaus, AM: Edies Ibama / ProVrzea. Benchimol, S. (1999 ). Amaznia: Formao Social e Cultural Manaus: Editora Valer / Editora da Universidade do Amazonas. Bernard, H. R. (2006). Research Methods in Anthropology (4 ed.). Oxford, UK: AltaMira. Berryman, P. (1987). Liberation Theology: Essential Facts about the Revolutionary Movement in Latin America and Beyond Bloomington, IN: Meyer Stone Books. Bittencourt, A. C. R. (1918). O Municpio da Lbrea: Noticia sobre a sua origem e desenvolvimento e sobre o rio Purs Manos Amazonas: Cesar, Cavalcanti, & Ca. Boff L. (1985). Church: Charism and Power Liberation Theology and the Institutional Church (J. W. Diercksmeier, Trans.). New York: Crossroad. Boissevain, J. (1977). When the saints go marching out: Reflections on the decline of patronage in Malta. In E. Gel lner & J. Waterbury (Eds.), Patrons and Clients in Mediterranean Societies (pp. 81 96). London: Duckworth. Bourdieu, P. (1986). The Forms of Capital. In J. Richardson (Ed.), Handbook of Theory and Research for the Sociology of Education Westport, CT: Gree nwood Press. Braga, W. S. M., Brasil, L. M., de Souza, R. A. B., de Melo, M. S., Rosas, M. D. G., Castilho, M. d. C., et al. (2004). Prevalncia da infeco pelos vrus da hepatite B (VHB) e da hepatite Delta (VHD) em Lbrea, Rio Purus, Estado do Amazonas. Epidemiologia e Servios de Sade, 13 (1), 35 46.
303 Brooke, J. (1995, July 9, 1995). Rubber Trees Being Planted in Brazil, the Source The New York Times, p. 114. Brown, C. B., & Lidstone, W. (1878). Fifteen Thousand Miles on the Amazon and its Tributaries London: Edward Stanford. Bruneau, T. C. (1986). Brazil: The Catholic Church and Basic Christian Communities. In D. H. Levine (Ed.), Religion and Political Conflict in Latin America (pp. 106 123). Chapel Hill and London: The University of North Carolina Pre ss. Burt, R. S. (1992). Structural Holes Cambridge, MA: Harvard University Press. Burt, R. S. (2004). Structural Holes and Good Ideas. American Journal of Sociology, 110 (2), 349 399. Cabral, A. L. (1984 (1949)). Dez Anos no Amazonas (1897 1907) (2 edio ed.). Braslia, DF: Senado Federal. Cardoso, E. (2009). A Brief History of Trade Policies in Brazil: From ISI, Export Promotion and Import Liberalization to Multilateral and Regional Agreements, The Political Economy of Trade Policy in the BRICS New Orle ans, LA. Chambers, R., Longhurst, R., & Pacey, A. (Eds.). (1981). Seasonal Dimensions to Rural Poverty Totowa, New Jersey: Allanheld, Osmun & Co. Publishers, Inc. Chandless, W. (1866). Ascent of the River Purs. Journal of the Royal Geographical Society o f London, 36 86 118. Chesnut, R. A. (1997). Born Again in Brazil: The Pentecostal Boom and the Pathogens of Poverty New Brunswick, New Jersey, and London: Rutgers University Press. CIMI Quem Somos. (2004). Retrieved January 12, 2010, from http://www.cimi.org.br/?system=news&action=read&id247&eid=224 Coppen, J. J. W. (1995). Non Wood Forest Products: Gums, resins and latexes of plant origin Rome, Italy: FAO Food and Agric ulture Organization of the United Nations. Costa, C. (1940). A Conquista do Deserto Ocidental (Subsidios para a Historia do Territorio do Acre) So Paulo: Companhia Editora Nacional. Costa, F. P. (2005). Seringueiros, Patres e a Justia no Acre Federal, 1904 1918 Rio Branco, Acre, Brasil: Editora da Universidade Federal do Acre EDUFAC. Cruz, O. (1913). Relatrio sobre as Condies Mdico Sanitrias do Vale do Amazonas Rio de Janeiro: Typ. do Jornal do Commercio, de Rodrigues & C. da Costa, A. M., Ramos, J. L., & de A. Cunha, M. V. (2007). Preservao de Lagos: Uma Experincia Ribeirinha Manaus: Comisso Pastoral da Terra Regional Amazonas CPT/AM. da Cunha, E. (2003a). Amaznia: Um paraso perdido Manaus: Editora Valer / Governo do Estado de Amazona s / Editora da Universidade Federal do Amazonas. da Cunha, E. (2003b). Entre os seringais. In C. Gallindo (Ed.), Canudos e Outros Temas (pp. 165 168). Braslia: Senado Federal. da Cunha, E. (2003c). O Povoamento e a navegabilidade do rio Purus. In C. Galli ndo (Ed.), Canudos e Outros Temas (pp. 169 188). Braslia: Senado Federal. da Gama e Abreu, J. C. (1895). As regies amazonicas: estudos chorographicos dos estados do Gram Par e Amazonas Lisboa: Imprensa de Libanio da Silva. DaMatta, R. A. (1990). Carnav ais, Malandros e Heris: Para uma Sociologia do Dilema Brasileiro Rio de Janeiro: Editora Guanabera Koogan S.A.
304 Davis, S. H. (1977). Victims of the Miracle: Development and the Indians of Brazil Cambridge: Cambridge University Press. de Melo, M. D. (1994 ). Do Serto Cearense s Barrancas do Acre (1 edio ed.). Manaus: Editora Calderaro Ltda. Dean, W. (1987). Brazil and the Struggle for Rubber: A Study in Environmental History New York: NY: Cambridge University Press. Diagnstico Scio econmico e Ca dastro Empresarial de Lbrea (1996).). Manaus: Program Informao Servio de Apoio s Micro e Pequenas Empresas do Amazonas SEBRAE/AM. Ehrenreich, P. (1929). Viagem no rios Amazonas e Purs. Revista do Museu Paulista, XVI 279 312. Eisenstadt, S. N., & Lemarchand, R. (Eds.). (1981). Political Clientelism, Patronage and Development Beverly Hills / London: SAGE Publications. Eisenstadt, S. N., & Roniger, L. (1980). Patron Client Relations as a Model of Structuring Social Exchange. Comparative Studies in Society and History, 22 (1), 42 77. Eisenstadt, S. N., & Roniger, L. (1984). Patrons, Clients, and Friends: Interpersonal Relations and the Structure of Trust in Society London: Cambridge University Press. Enns, S., Malinick, T., & Matthews, R. (2008). It' s Not Only Who You Know, It's Also Where They Are: Using the Position Generator to Investigate the Structure of Access to Embedded Resources. In N. Lin & B. H. Erickson (Eds.), Social Capital: An International Research Program (pp. 255 281). New York: Oxfo rd University Press. Escobar, A. (2008). Territories of Difference Durham and London: Duke University Press. Escobar, C. (1994). Clientelism and Social Protest: Peasant Politics in Northern .*/*01+23"4%"53")*%+,&-"!"63"#$%&' Ayata (Eds.), Democracy, Clien telism, and Civil Society (pp. 65 85). Boulder and London: Lynne Rienner Publishers. Ferrarini, S. A. (1981). Lbrea: 1881 Ontem Hoje 1981 Manaus: Imprensa Oficial do Estado do Amazonas. Ferrarini, S. A. (2009). Rio Purus: Histria Cultura Ecologia So Paulo: FTD. Flap, H., & Boxman, E. (2001). Getting Started: The Influence of Social Capital on the Start of the Occupational Career. In N. Lin, K. Cook & R. S. Burt (Eds.), Social Capital: Theory and Research (pp. 159 181). New York: Aldine de Gruyter Foster, G. M. (1961). The Dyadic Contract: A Model for the Social Structure of a Mexican Peasant Village. American Anthropologist, New Series, 63 (6), 1173 1192. Foster, G. M. (1963). The Dyadic Contract in Tzintzuntzan, II: Patron Client Relationship. Am erican Anthropologist, New Series 65 (6), 1280 1294. Frank, Z., & Musacchio, A. (2006). Brazil and the International Rubber Trade, 1870 1930. In S. Topik, C. Marichal & Z. Frank (Eds.), From Silver to Cocaine: Latin American Commodity Chains and the Buildin g of the World Economy, 1500 2000 (pp. 271 299). Durham, NC: Duke University Press.
305 Freston, P. (1993). Brother Votes for Brother: The New Politics of Protestantism in Brazil. In V. Garrard Burnett & D. Stoll (Eds.), Rethinking Protestantism in Latin Ameri ca (pp. 66 110). Philadelphia: Temple University Press. Ganzeboom, H. B. G., & Treiman, D. J. (2003). Three Internationally Standardised Measures for Comparative Research on Occupational Status. In J. H. P. Hoffmeyer Zlotnik & C. Wolf (Eds.), Advances in C ross National Comparison: A European Working Book for Demographic and Socio Economic Variables (pp. 159 193). New York: Kluwer Academic/Plenum Publishers. Gay, R. (1998). Rethinking Clientelism: Demands, Discourses and Practices in Contemporary Brazil. Eur opean Review of Latin American and Caribbean Studies, 65 7 24. Gondim, J. (2001). Atravs do Amazonas: Impresses de viagens realizadas em 1921 (2 edio ed.). Manaus: Edies Governo do Estado do Amazonas / Secretaria de Estado da Cultura, Turismo e Des porto. Goodman, L. A. (1961). Snowball Sampling. Annals of Mathematical Statistics, 32 148 170. Gough, K. (1981). Rural Society in Southeast India Cambridge: Cambridge University Press. Goulding, M., Barthem, R., & Ferreira, E. (2003). The Smithsonian At las of the Amazon Washington, D.C.: Smithsonian Books. Granovetter, M. S. (1973). The Strength of Weak Ties. The American Journal of Sociology, 78 (6), 1360 1380. Granovetter, M. S. (1983). The Strength of Weak Ties: A Network Theory Revisited. Sociologica l Theory, 1 201 233. Greenfield, S. M. (1979). Patron Client Exchanges in Southeastern Minas Gerais. In M. L. Margolis & W. E. Carter (Eds.), Brazil Anthropological Perspectives: Essays in Honor of Charles Wagley (pp. 362 378). New York: Columbia Univer sity Press. Guasti, L. (1981). Clientelism in Decline: A Peruvian Regional Study. In S. N. Eisenstadt & R. Lemarchand (Eds.), Political Clientelism, Patronage and Development (pp. 217 248). Beverly Hills and London: SAGE Publications. #$%&' Ayata, A. (1994). Clientelism: Premodern, Modern, Postmodern. In L. Roniger & 63"#$%&' Ayata (Eds.), Democracy, Clientelism, and Civil Society (pp. 19 28). Boulder and London: Lynne Rienner Publishers. Hemming, J. (1978). Red Gold: The Conquest of th e Brazilian Indians, 1500 1760 Cambridge, MA: Harvard University Press. Hemming, J. (1987). Amazon Frontier: The Defeat of the Brazilian Indians Cambridge, MA: Harvard University Press. Hemming, J. (2003). Die If You Must: Brazilian Indians in the Twenti eth Century London: Macmillan. Hoelz, J. J., & Martinez, A. A. (1972). A Cultura da Seringueira no Estado de So Paulo, I Seminrio Nacional da Seringueira Cuiab, Mato Grosso: Ministrio da Indstria e do Comrcio Superintendncia da Borracha. Ireland R. (1993). The Crentes of Campo Alegre and the Religious Construction of Brazilian Politics. In V. Garrard Burnett & D. Stoll (Eds.), Rethinking Protestantism in Latin America (pp. 45 65). Philadelphia: Temple University Press.
306 Jackson, J. (2008). The Th ief at the End of the World: Rubber, Power, and the Seeds of Empire New York: Penguin Books. Jaramillo, H. (1902). Breve noticia sobre os valles dos rios Purs e Madeira: affluentes do amazonas : estradas de ferro projectadas Madeira e Mamor no Matto Gro sso Central do Amazonas no rio Purs Manos: Livraria e Typ. de M. Silva Kroemer, G. (1985). Cuxiuara O Purus dos Indgenas: Ensaio Etno Histrico e Etnogrfico sobre os ndios do Mdio Purus So Paulo: Edies Loyola. La Rue, C. D. (1926). The Hevea Rubber Tree in the Amazon Valley Washington, D.C.: Government Printing Office United States Department of Agriculture Labre, A. R. P. (1872). Rio Purs Noticia Maranho: Typ. do Paiz. Imp. M.F.V. Pires. Land, C. H. (1983). Political Clientelism in Political Studies: Retrospect and Prospects. International Political Science Review, 4 435 545. Lange, A. (1911). The Rubber Workers of the Amazon. Bulletin of the American Geographical Society, 43 (1), 33 36. Lange, A. (1912). In the Amazon Jungle New Yo rk: G.P. Putnam's Sons. Lange, A. (1914). The Lower Amazon New York: The Knickerbocker Press. Lemarchand, R. (1981). Comparative Political Clientelism: Structure, Process and Optic. In S. N. Eisenstadt & R. Lemarchand (Eds.), Political Clientelism, Patron age and Development (pp. 7 32). Beverly Hills and London: SAGE Publications. Lin, N. (1982). Social Resources and Instrumental Action. In P. V. Marsden & N. Lin (Eds.), Social Structure and Network Analysis (pp. 131 145). Beverly Hills, CA: Sage. Lin, N. ( 1999). Social Networks and Status Attainment. Annual Review of Sociology, 25 467 487. Lin, N. (2001a). Building a Network Theory of Social Capital. In N. Lin, K. S. Cook & R. S. Burt (Eds.), Social Capital: Theory and Research (pp. 3 29). New York, NY: Al dine de Gruyter. Lin, N. (2001b). Social Capital: A Theory of Social Structure and Action New York, NY: Cambridge University Press. Lin, N., & Dumin, M. (1986). Access to Occupations Through Social Ties. Social Networks, 8 365 385. Lin, N., & Erickson, B H. (2008). Theory, Measurement, and the Research Enterprise on Social Capital. In N. Lin & B. H. Erickson (Eds.), Social Capital: An International Research Program (pp. 1 24). New York: Oxford University Press. Little, P. E. (2001). Amazonia: Territorial Struggles on Perrennial Frontiers Baltimore, MD: The John Hopkins University Press. Loureiro, A. J. S. (1981). A "Gazeta do Purus"; scenas de uma pocha (Sena Madureira 1918 1924) Manuas: Imprensa Oficial do Estado do Amazonas. Loureiro, A. J. S. (200 7). Histria da Navegao no Amazonas Manaus: Lorena Ltda. Luxardo, L. (1973). Purus: Histria de Ontem Estrias de Hoje Belm, Par, Brasil: GRAFISA Cia. Grfica e Editora Globo. Marsden, P. V. (2005). Recent Developments in Network Measurement. In P. J. Carrington, J. Scott & S. Wasserman (Eds.), Models and Methods in Social Network Analysis (pp. 8 30). New York, NY: Cambridge University Press.
307 Marsden, P. V., & Campbell, K. E. (1984). Measuring Tie Strength. Social Forces, 63 (2), 482 501. Martin, J L. (2009). Social Structures Princeton, NJ: Princeton University Press. Martinello, P. (2004). A "Batalha da Borracha" na Segunda Guerra Mundial Rio Branco: EDUFAC. Mathews, E. D. (1879). Up the Amazon and Madeira Rivers, through Bolivia and Peru Lond on: Sampson Low, Marston, Searle & Rivington. Medeiros, J. F., Py Daniel, V., Barbosa, U. C., & Izzo, T. J. (2009). Mansonella ozzardi in Brazil: prevalence of infection in riverine communities in the Purus region, in the state of Amazonas. Memrias do Ins tituto Oswaldo Cruz, 104 (1), 74 80. Medeiros, J. F., Py Daniel, V., Barbosa, U. C., & Ogawa, G. M. (2008). Current profile of Mansonella ozzardi (Nematoda: Onchocercidae) in communities along the Ituxi river, Lbrea municipality, Amazonas, Brazil. Memrias do Instituto Oswaldo Cruz, 103 (4), 409 411. Miller, D. (1985). Replacement of traditional elites: an Amazon case study. In J. Hemming (Ed.), Change in the Amazon Basin Vol.2: The frontier after a decade of destruction (pp. 158 171). Manchester: Manchester University Press. Mintz, S. W., & Wolf, E. R. (1950). An Analysis of Ritual Co Parenthood (Compadrazgo). Southwestern Journal of Anthropology, 6 (4), 341 368. O nascimento da CPT (2010).). Goinia, Gois: Comisso Pastoral da Terra (CPT) Secretaria Nac ional. Pantoja, M. C. (2004). Os Milton: Cem Anos de Histria nos Seringais Recife: Fundao Joaquim Nabuco. Parquia Nossa Senhora de Nazar Lbrea AM 125 Anos Evangelizando (2003).). Lbrea, AM: Parquia Nossa Senhora de Nazar Patron client rel ations. (2000). In A. G. Johnson (Ed.), The Blackwell Dictionary of Sociology: A User's Guide to Sociological Language (Second Edition ed., pp. 224). Malden, MA: Blackwell Publishers. Pearson, H. C. (1911). The Rubber Country of the Amazon New York: The I ndia Rubber World. Peret, J. A. (1985). Amazonas: Histria, Gente, e Costumes Braslia: Senado Federal Centro Grfico. Portes, A. (1998). Social Capital: Its Origins and Applications in Modern Sociology. Annual Review of Sociology, 24 1 24. PROBOR II S egundo Programa de Incentivo Produo de Borracha Natural: Fundamentos Objetivos Normas Operativas. (1978). In M. d. I. e. d. C. S. d. Borracha (Ed.). Programa de Incentivo Produo de Borracha Vegetal Normas Operativas Aplicveis ao Financiamen to de Projetos PROBOR (1972). In M. d. I. e. d. C. S. d. Borracha (Ed.) (Vol. Decreto lei n 7"839:9;"<&"8=3=38>=9?3 Putnam, R. D. (2000). Bowling Alone: The Collapse and Revival of American Community New York, NY: Simon & Schuster. Putnam, R. D., Leonard i, R., & Nanetti, R. Y. (1993). Making Democracy Work: Civic Traditions in Modern Italy Princeton: Princeton University Press. Reis, A. C. F. (1953). O Seringal e o Seringueiro Rio de Janeiro: Ministrio da Agricultura / Servio de Informao Agrcola.
308 R eis, A. C. F. (1998). Histria do Amazonas (3 ed.). Belo Horizonte: Editora Itatiaia Limitada. Ribeiro, D. (1996). Os ndios e a civilizao: a integrao das populaes indgenas no Brasil moderno (3. reimpresso ed.). So Paulo: Companhia das Letras. Ri cardo, F. (Ed.). (2004). Terras Indgenas & Unidades de Conservao da natureza: o desafio das sobreposies So Paulo: Instituto Socioambiental. Roniger, L. (1990). Hierarchy and Trust in Modern Mexico and Brazil New York: Praeger. Roniger, L. (1994a). Conclusions: The Transformation of Clientelism and Civil Society. 4%"53")*%+,&-"!"63"#$%&' Ayata (Eds.), Democracy, Clientelism, and Civil Society (pp. 207 214). Boulder and London: Lynne Rienner Publishers. Roniger, L. (1994b). The Comparative Study of Cl ientelism and the Changing Nature of .+@+/"A*B+&CD"+%"CE&".*%C&0F*-2-D"G*-/<3"4%"53")*%+,&-"!"63"#$%&' Ayata (Eds.), Democracy, Clientelism, and Civil Society (pp. 1 18). Boulder and London: Lynne Rienner Publishers. )*%+,&-;"53;"!"#$%&' Ayata, A. (Eds.). (1994). Democracy, Clientelism, and Civil Society Boulder and London: Lynne Rienner Publishers. Sant'Ana Jnior, H. A. d., Pereira, M. d. J. F., Alves, E. d. J. P., & Pereira, C. R. A. (Eds.). (2009). Ecos dos conflitos socioambientais: a RESEX de Tau Mi rim So Lus: EDUFMA. Santos, F. S. D. d. (2000). Ainda a "cultura do barraco" nos seringais da Amaznia. Histria Oral Revista da Associao Brasileira de Histria Oral, 3 (Junho), 69 89. Santos, F. S. D. d., & Muaze, M. d. A. F. (2002). Tradies em M ovimento: Uma Etnohistria da Sade e da Doena nos Vales dos Rios Acre e Purus Braslia: Paralelo 15. Santos, R. (1980). Histria Econmica da Amaznia (1800 1920) So Paulo, Brazil: T.A. Queiroz. Schmink, M., & Wood, C. H. (1992). Contested Frontiers i n Amazonia New York: Columbia University Press. Schurz, W. L., Hargis, O. D., Marbut, C. F., & Manifold, C. B. (1925). Rubber Production in the Amazon Valley Washington, D.C.: Department of Commerce / Government Printing Office. Scott, J. (1977). Patrona ge or exploitation? In E. Gellner & J. Waterbury (Eds.), Patrons and Clients in Mediterranean Societies (pp. 21 39). London: Duckworth. Scott, J. C. (1985). Weapons of the Weak: Everyday Forms of Peasant Resistance New Haven, CT: Yale University Press. Sc ott, J. C. (1990). Domination and the Arts of Resistance: Hidden Transcripts New Haven and London: Yale University Press. Scott, J. C., & Kerkvliet, B. J. (1973). How Traditional Rural Patrons Lose Legitimacy. Cultures et dveloppment 501 540. Secreto, M V. (2007). Soldados da Borracha: Trabalhadores entre o serto e a Amaznia no governo Vargas So Paulo: Editora Fundao Perseu Abramo. Shelley, A. J., & Coscarn, S. (2001). Simuliid Blackflies (Diptera: Simuliidae) and Ceratopogonid Midges (Diptera: C eratopogonidae) as Vectors of Mansonella
309 ozzardi (Nematoda: Onchocercidae) in Northern Argentina. Memrias do Instituto Oswaldo Cruz, 96 (4), 451 458. Silva, M. C. (1994). Peasants, Patrons, and the State in Northern Portugal. In L. )*%+,&-"!"63"#$%&' Ayata (Eds.), Democracy, Clientelism, and Civil Society (pp. 29 47). Boulder and London: Lynne Rienner Publishers. Smith, H. H. (1879). Brazil the Amazons and the Coast New York: Charles Scribner's Sons. Smith, N. J. H. (1999). The Amazon River Forest: A Natural History of Plants, Animals, and People New York, NY: Oxford University Press. Soares, F. V., Ribas, R. P., & Osrio, R. G. (2010). Evaluating the Impact of Brazil's Bolsa Famlia: Cash Transfer Programs in Comparative Perspective. Latin American Research Review, 45 (2), 173 190. Sorauf, F. J. (1960). The Silent Revolution in Patronage. Public Administration Review, 20 (1), 28 34. Steere, J. B. (1903). Narrative of a Visit to Indian Tribes of the Purus River, Brazil. In (pp. 359 393). Washington, D.C.: Smithsonian Institution / United States National Museum / Government Printing Office. Stepan, N. L. (2003). "The Only Serious Terror in These Regions": M alaria Control in the Brazilian Amazon. In D. Armus (Ed.), Disease in the History of Modern Latin America: From Malaria to AIDS (pp. 25 50). Durham and London: Duke University Press. Stoll, D. (1990). Is Latin America Turning Protestant? The Politics of Ev angelical Growth Berkeley: University of California Press. Stoll, D. (1993). Introduction: Rethinking Protestantism in Latin America. In V. Garrard Burnett & D. Stoll (Eds.), Rethinking Protestantism in Latin America (pp. 1 19). Philadelphia: Temple Unive rsity Press. Tastevin, C. (1943). The Middle Amazon: Its People and Geography (Eleven Articles by Constant Tastevin) (T. S. I. o. t. Americas, Trans.). Washington, D.C.: Coordinator of Inter American Affairs Office of Emergency Management. Teixeira, C. C (2009). Servido Humana na Selva: O Aviamento e o Barraco na Sociedade do Seringal Manaus: Editora Valer / Edua. Theobald, R. (1983). The Decline of Patron Client Relations in Developed Societies. Archives europennes de sociologie, 24 (1), 136 147. The obald, R. (1992). On the Survival of Patronage in Developed Societies. Archives europennes de sociologie, 33 183 191. Townsend, J. (1985). Seasonality and capitalist penetration in the Amazon. In J. Hemming (Ed.), Change in the Amazon Basin Vol.2: The fr ontier after a decade of destruction (pp. 140 157). Manchester: Manchester University Press. van der Gaag, M. (2005). Measurement of Individual Social Capital Amsterdam: F&N Boekservices. van der Gaag, M., & Snijders, T. A. B. (2005). The Resource Genera tor: social capital quantification with concrete items. Social Networks, 27 1 29. van der Gaag, M., Snijders, T. A. B., & Flap, H. (2008). Position Generator Measures and Their Relationship to Other Social Capital Measures. In N. Lin & B. H. Erickson (Eds .), Social Capital: An International Research Program (pp. 27 48). New York: Oxford University Press.
310 Verbrugge, L. M. (1979). Multiplexity in Adult Friendships. Social Forces, 57 (4), 1286 1309. Vercruijsse, E. (1984). The Penetration of Capitalism A Wes t African Case Study London: Zed Press. Wagley, C. (1977). Welcome of Tears: The Tapirap Indians of Central Brazil Prospect Heights, IL: Waveland Press, Inc. Waterbury, J. (1977). An attempt to put patrons and clients in their place. In E. Gellner & J. Waterbury (Eds.), Patrons and Clients in Mediterranean Societies (pp. 329 342). London: Duckworth. Weingrod, A. (1968). Patrons, Patronage, and Political Parties. Comparative Studies in Society and History, 10 (4), 377 400. Weingrod, A. (1977). Patronage an d power. In E. Gellner & J. Waterbury (Eds.), Patrons and Clients in Mediterranean Societies (pp. 41 51). London: Duckworth. Weinstein, B. (1983). The Amazon Rubber Boom 1850 1920 Stanford, California: Stanford University Press. Wolff, C. S. (1999). Mulhe res da Floresta: uma histria Alto Juru, Acre (1890 1945) So Paulo: Editora Hucitec. Woodroffe, J. F. (1914). The Upper Reaches of the Amazon London: Methuen & Co. Ltd. Yungjohann, J. C. (1989). White Gold: The diary of a rubber cutter in the Amazon 19 06 1916. In G. T. Prance (Ed.). Oracle, AZ: Synergetic Press, Inc.
311 BIOGRAPHICAL SKETCH My passion for natural resource management stems from family hikes in the canyon country of western Colorado. The beauty of these places inspired me to study nat ural resource management at Colorado State University. After graduation I served in the Peace Corps in Guatemala. As I worked with a community to develop an ecotourism project and improve the management of their forest reserve I became inspired to pursue a career working in Latin America. I love learning languages, being exposed to different cultures, and working with communities as they strive to improve their lives. My research in the Natural Resources and Sustainable Development Master's program at American University and the United Nations University for Peace in Costa Rica focused on community participation in the Mesoamerican Biological Corridor. An internship in Nicaragua assisting with the development of a regional forest management strategy re inforced my desire to work with communities. In spite of my field experiences, I felt I needed more academic training to fulfill my career goals. C ompleting a PhD in Interdisciplinary Ecology at the University of Florida and my dissertation research exp erience i n Brazil's Amazonas state has prepare d me to work as a university professor. I intend to work on projects that focus on communities and natural resource management in Latin America. I plan on a career conducting research and facilitating informa tion flow between donor agencies, policy makers, and, most importantly, local resource managers and users.