HUNTING AND HOUSEHOLD IN PDS SÃƒ O SALVADOR, ACRE, BRAZIL By ERIC MINZENBERG A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2005
Copyright 2005 by Eric Minzenberg
to all of my teachers
iv ACKNOWLEDGMENTS What seemed as an eternity when I firs t entered Gainesville several years ago is finally coming to closure. This dissertati on could not have been completed without the love, guidance, patience, and support of numero us people. First, I thank my family for their continual help throughout my graduate school journey. My mother, Jean, although she may not have always understood wh at I was studying (and why), provided unconditional love and emotiona l support. Now she is hoping that I can market this degree. My brother, Mike, was the firs t person who suggested I pursue the PhD. Needless to say, I took his advice. Mike al so helped me think through my research questions, methodology, and data analysis, in the process making my writing clearer and more succinct than it otherwise would have b een without his input. My enjoyment with finishing this dissertation was enhanced by La ura Bird who provided tenderness, comfort, and love as this process reached conclusion. I regret that my father did not live long enough for me to present him with a copy of this work. I was blessed with a talented and tho ughtful supervising committee consisting of my chair, Marianne Schmink, and Michael Heckenberger, Brenda Chalfin, and Stephen Perz. I am the latest in a long line of st udents that Marianne Schmink has mentored throughout her distinguished car eer. I could not have asked for a more responsive and insightful chair that constantly pushed me to be the best student and writer that I could be. Michael Heckenberger continually provide d poignant insights that helped me think of my work within th e broader regional pe rspective of Amazonian studies. Brenda
v ChalfinÂ’s careful observations of the theoretic al aspects of my work made my analysis cleaner and stronger. I could not have begun to undertake the statistical analysis used throughout this dissertation wit hout the guidance of Stephen Perz. Throughout my time at the University of Florida, the four members of my committee mentioned above provided countless hours of time and energy, always in a friendly and positive manner, towards improving my doctoral studies. In addition to my supervisory committee, th ere were several other professors that played a role in shaping my thinking. H. Russell Bernard was instrumental in helping me define my research questions and prepare a coherent research de sign and methodology. It is largely due to him that my research proposal achieved funding. I first garnered an appreciation for who is the Â“peasantÂ”, and what he/she does from Leslie Anderson. Tom Davies, Jr., from the days of my masterÂ’s work at San Diego State University, was a strong influence in the early years of my graduate school training. I regularly traded storie s and ideas concerning our work, with my colleagues Richard Wallace, Valerio Gomes, Samant ha Stone, Jackie Vadjunec, and David Salisbury. Their comments and suggestions duri ng fieldwork and on earlier drafts of this manuscript were very helpful. The Brazilian NGO, PESACRE, was very accommodating in providing logistical support in Acre. I especially thank VangelÃ¢ Nascimento, and (Cazuza) Eduardo Borges. Their friendshi p helped me survive the loneliness of fieldwork in isolated areas. Research a ssistants, MÃ¡rcia and MarinÃªs collected data diligently and provided a smoother integration within the communities where I did fieldwork than otherwise would have been po ssible. I would never have been able to reach my field sites without the boat operato rs Jesuita and Novo. We suffered together
vi the arduous all-day boat trips in the cold ra in and blinding sun entering and leaving PDS SÃ£o Salvador. The families of JosÃ© Franci sco and Dora in the community of Rio Azul, and Seu Paolo and Dona Nenem in Boa Vista were gracious hosts during fieldwork. Not once did they complain that my insisten t questions were a burden to them. I also thank the National Science Founda tion and the Working Forests in the Tropics program at the University of Flor ida for funding this research project. Finally, I thank all of the families of PDS SÃ£o Salvador who allowed me to intrude into their lives.
vii TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iv LIST OF TABLES.............................................................................................................ix LIST OF FIGURES.............................................................................................................x ABSTRACT....................................................................................................................... xi CHAPTER 1 INTRODUCTION........................................................................................................1 Animals and Tropical Forests.......................................................................................4 The Peasant Household.................................................................................................6 Amazonian Peasantry...................................................................................................8 Hunting and the Peasantry..........................................................................................13 Diagram of Peasant Hunting.......................................................................................20 Intra-household Hypotheses on Hunting....................................................................22 Inter-household Hypotheses on Hunting....................................................................25 Dissertation Layout.....................................................................................................26 2 THE HISTORY OF PDS SÃƒO SAL VADOR AND THE RESEARCH SITE..........29 The History of Rubber in the Development of the State of Acre...............................32 Post World War II Amazonian Development..........................................................36 Rubber in the Development of PDS SÃ£o Salvador.....................................................40 Post-rubber Political-economy in PDS SÃ£o Salvador.................................................45 The State of Acre........................................................................................................61 The Municipality of MÃ¢ncio Lima.............................................................................67 The Research Site PDS SÃ£o Salvador.......................................................................69 Seasonal Calendar of Resource Use...........................................................................75 Travel in PDS SÃ£o Salvador.......................................................................................80 3 METHODS.................................................................................................................87 The Household............................................................................................................89 House and Household in Movement...........................................................................91 Fieldwork....................................................................................................................96
viii Methodological Considerations..................................................................................97 Household Surveys.....................................................................................................98 Productive and Reproductive Household Survey.....................................................103 Time Allocation Surveys..........................................................................................104 Hunting Outings........................................................................................................109 Participant Observation............................................................................................110 4 HUNTING WITHIN THE HOUSEHOLD (INTRA-HOUSEHOLD).....................112 The Hunt in PDS SÃ£o Salvador................................................................................113 Hunting Techniques..................................................................................................122 Material Realities of the Household and Hunting....................................................126 Household Gender Relationships and Hunting.........................................................134 Gender Hypotheses and Hunting Practice................................................................141 Emotional Commitme nt to Hunting.........................................................................144 Seasonality and Hunting...........................................................................................147 Modeling Intra-household Effect s of Hunting Frequency........................................151 Conclusion................................................................................................................155 5 HUNTING ACROSS HOUSEHOLDS (INTER-HOUSEHOLD)...........................162 Hunting Across Households.....................................................................................162 Meat Exchange.........................................................................................................173 Inter-household Hypotheses on Hunting..................................................................180 Modeling Inter-household Hunting Relationships....................................................183 Modeling Intra-household with Inte r-household Effects on Hunting.......................185 Conclusion................................................................................................................188 6 CONCLUSION.........................................................................................................190 Summary of Key Findings and Conclusions............................................................192 Implications of Findings for Anthropology..............................................................205 Implications of Findings for Conservation and Development..................................209 Implications of Findings for Gender Studies............................................................213 APPENDIX A HOUSEHOLD SURVEY.........................................................................................215 B PRODUCTIVE/REPRODUCTIVE HOUSEHOLD SURVEY...............................220 C TIME ALLOCATION..............................................................................................222 D HUNTING OUTINGS..............................................................................................223 LIST OF REFERENCES.................................................................................................224 BIOGRAPHICAL SKETCH...........................................................................................240
ix LIST OF TABLES Table page 2-1 Community Hunting Regulat ions in PDS SÃ£o Salvador..........................................59 3-1 Field Research Schedule..........................................................................................97 3-2 Household Sampling..............................................................................................102 4-1 Households that Enjoy Hunting.............................................................................115 4-2 Age that Hunters Began to Hunt............................................................................117 4-3 Hunting Frequency across Communities...............................................................118 4-4 Spot Sampling of Adult Activities by Community................................................120 4-5 Domesticated Animals by Household....................................................................131 4-6 Mean, Standard Deviation of Domesticated Animals/Household.........................131 4-7 Productive and Reproductive H ousehold Activities by Gender.............................136 4-8 Spot Sampling of Household Activities by Gender...............................................139 4-9 Household Activities as a Function of Rain or Sun...............................................150 4-10 T-test for Significance of Intra-hous ehold Material & Gender Variables..............151 4-11 Multivariate Regression Modeling of In tra-household Influence on Frequency of Hunting..............................................................................................................153 5-1 Frequency of Hunting per Household by Community...........................................168 5-2 Hunting Success/Household...................................................................................173 5-3 Meat Sharing by Relation to Giving Household....................................................177 5-4 Sharing of Wild Game by Gender within Hunting Households.............................178 5-5 Multivariate Regression Modeling of In ter-household Influence on Frequency of Hunting...................................................................................................................184 5-6 Multivariate Regression Models (I ntra-household and Inter-household)..............186
x LIST OF FIGURES Figure page 1 Diagram of Peasant Hunting in PDS SÃ£o Salvador..................................................21 2-1 Map of PDS SÃ£o Salvador........................................................................................53 2-2 Map of Acre.............................................................................................................71 2-3 The Moa River Cut through the Heart of PDS SÃ£o Salvador...................................75 2-4 Seasonal Calendar of Activities...............................................................................76 3-1 Research Assistant Interviewing Re sident of the Community Boa Vista................96 4-1 Frequency of Hunting.............................................................................................113 4-2 Hunter from Rio Azul with Tucano ( Ramphastos vitellinus) and Nambu Galinha ( Tinamus guttatus )..................................................................................................124 4-3 Number of Children Living in a Household...........................................................129 5-1 Number of Hunting Outings per Month.................................................................167 5-2 Method of Hunting pe r Household by Community...............................................169 5-3 Hunting Success per Household by Community....................................................170 5-4 Size of Animals Hunted per Household by Community........................................171 5-5 Woman Cleaning Paca ( Agouti paca ) Hunted by Her Husband............................179
xi Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy HUNTING AND HOUSEHOLD IN PDS SÃƒ O SALVADOR, ACRE, BRAZIL By Eric Minzenberg December 2005 Chair: Marianne Schmink Major Department: Anthropology Hunting is an integral part of the fore st-based activities of peasant populations throughout the tropical forest s in the Amazon region, yet th e motivations of hunting by peasant households are not well represented in the literature. Th is study addresses the need to understand peasant hunting behavior a nd strategy in Amazonia as part of wider forest management strategies by peasant hous eholds. The principal objectives of the study are 1) to understand the influence of household organization and decision-making on peasant hunting strategy a nd behavior, and 2) to understand the effects of interhousehold relationships on hunting practices. Although hunting frequency has declined recently in PDS SÃ£o Salvador, the overwhelming majority of households sampled in this study continue to hunt wild game. Men are the hunters, but womenÂ’ s participation in insuring th at the reproductive needs of the household are met, in the cooking of wild game, and the partitioning of game meat to other households in the community, is critical to the overall hunti ng performance of the household.
xii Multiple regression analysis indicates that the most salient influences on hunting output in PDS SÃ£o Salvador are a househol dÂ’s desire to continue with the hunting lifestyle and the social bonds that are produced with the performance of the hunt. Peasant households that are located at the periphery of the market, as occurs in PDS SÃ£o Salvador, refrain from furthe r market integration, in part , because of the non-market exchanges that are created and maintained between households with the hunt. Hunting, integrated within a forest culture, as an activity of fundamental importance within the repertoire of the householdÂ’s livelihood stra tegy, is thus vital to the construction of peasant livelihoods that have some agency in terms of their engagement with the market system. The findings of this study are mixed in te rms of their contribution to conservation and development in the region, although the domestication of livestock was shown to lessen the frequency of hunting. Small animal projects could provide an alternative source of meat for families, while also potentially raising the status of women of the household who are currently invested w ith the raising of these animals.
1 CHAPTER 1 INTRODUCTION This project investigates the practice of hunting within peasant households and communities in the Brazilian Amazonian State of Acre. The principal objectives of the study are 1) to understand the influence of household organization and decision-making on peasant hunting strategy a nd behavior, and 2) to understand the effects of interhousehold relationships on hunting practices. Additionally, this project considers the historical trajectory of resource use by residents in PDS SÃ£o Salvador and the implications of hunting for other forest based ac tivities and the sustainability of the forest resource base within the settlement. In th e state of Acre, an ar ea known for its Â“forest government,Â” this study aims to link the impor tance of hunting in th e rural livelihoods of peasant households and communities to wider is sues of forest management. Past research on hunting in Amazonia ha s focused on the effects of hunting on wildlife populations, or hunting behavior and strategy by indigenous groups, but little is known about the social component of intrahousehold and inter-household influences on hunting by peasant groups in the re gion. Hunting is an integral part of the forest-based activities of peasant populations throughout th e tropical forests in the Amazon region, yet the motivations of hunting by peasant households are not well represented in the literature. This study addresses the need to understand peasant hunting behavior and strategy in Amazonia as part of the wider forest management strategies of peasant households.
2 This study links theories of the peasan t economy within the household, and across peasant households in the co mmunity, with the literatu re on hunting strategy and behavior by indigenous groups in Amazonia. This project presumes that hunting by peasants is not strictly utilitarian in th e provision of protein an d calories for peasant households, but takes on symbolic and social func tions as well. The symbolic function is seen in an emotional commitment to hunting by men as a way to initiate their sons and nephews into the ways of manhood. The decisi on to hunt, and how much, is not made by men on their own, but is part of a larger e ffort, a negotiation at th e household level, to allocate labor and other resources. Understa nding this negotiation is essential for understanding peasant hunting behavior. Furt hermore, the hunt takes on social functions across households in the community that help to join households together, thus insuring the survival of all members of the community. The current government in the state of Acre, popularly known as the Â“governo da florestaÂ” (forest government), is an important ally in the search for sustainable resource policies, and has incorporated a deve lopment philosophy that seeks ecosystem conservation, social equity, and cultural di versity (Kainer et al. 2003). Innovative approaches of conservation and development policy in the state of Acre have brought together concerned scientists and unive rsities, state agencies, non-governmental organizations (NGOs), and local peoples to ach ieve socially and ecologically sustainable resource policies. As the state of Acre ha s taken the lead in the implementation of novel development policies in Amazonia, this res earch will provide useful information to increase the capacity of development planne rs to implement sustainable social and ecological resource policies.
3 Development planners must include local pe oples in the management of wildlife in their efforts to protect game species from ove ruse, as Â“hunting issues cannot be separated from land useÂ” (Wadley et al. 1997: 263), and th erefore, they must understand the social dynamics of hunting and the use of wild game in the livelihood systems of peasant communities in Amazonia. An understanding of the social dynamics of the practice of hunting and the use of wild game in the liv elihood systems of peasant communities in Amazonia will help conservation and development organizations, and the state of Acre, in their attempts to provid e projects that account for bot h the cultural and biological diversity of the region. This study builds upon the work of a Brazi lian NGO from the State of Acre, PESACRE, that has worked with the comm unities of PDS SÃ£o Salvador since 1998. The idea for this research began as a dialo gue between a PESACRE biologist, VÃ¢ngela Nascimento, and me during my initial visit to the area in the summer of 2002. I came to western Acre in that year to investigat e the possibility of conducting research on a proposed non-timber forest product (NTFP) pr oject in the JuruÃ¡ Valley. As the NTFP project was having some difficulty taking off, I learned in conversation with VÃ¢ngela and other PESACRE employees living in the town of MÃ¢ncio Lima that hunting by the residents that lived within the sustainable development settlement SÃ£o Salvador (Projeto Desenvolvimento SustentÃ¡vel (PDS) SÃ£o Salvador on the Moa River), was a prime source of local conflict, and an ar ea of concern for PESACRE. An earlier study of wildlif e populations within the ar ea concluded that some wildlife species had become locally extinct as a result of hunting pre ssure by residents of the area (Fragoso et al. 2002). It was sugge sted that a research project designed to
4 investigate the hunting practices and motivations of the local reside nts of the area could help PESACRE in its attempt to maintain viable populations of wildlife in PDS SÃ£o Salvador while also accounting for the impor tance of wildlife with in the livelihoods and lifestyles of the residents in the region. My interests in the field of anthropology are critically centered on gender relations within the household, and wider community, in the broader context of natural re source use in the lowland tropics of Latin America. Therefore, this research topic combines an interest stated by PESACRE in hunting and the management of wildlife in PDS SÃ£o Salvad or with my own interests in the dynamics of household relations and relationships. Animals and Tropical Forests Meijaard et al. (2005) descri be several important ways in which animals influence the tropical forest ecosystem. First, animal s graze and browse on ve getation in tropical forests, affecting the structur e and composition of the forest. Many animals that live in tropical forests feed on fruits that fall to the forest floor and Â“are involved in seed dispersal, seed predation, and the structur ing of tropical forestsÂ” (Redford 1992: 418). Rodents frequently bury seeds, ultimately ha ving a role in influencing the process of germination and eventual seedling establishmen t of certain plant species. Bodmer (1991, 1994) has described how tapirs play an importa nt role in the dispersal and production of Mauritia flexuosa by spitting out seeds of the fruit from this palm. Other studies have also demonstrated the importance of animal sp ecies in the predation and dispersal of tree seeds and thus the structuring of tropical forests (Peres 1991; Wunderle 1997; Galetti et al. 2003). Animals also have various secondary effects on the environment through Â“habitat engineeringÂ” (Meijaard et al . 2005: 5). This includes anim als excavating holes, creating
5 pools, and turning and (thus) aerating the soil and litter layer on the fo rest floor. In the Amazon region, the paca ( Agouti paca ) and different species of armadillo ( Cabassous sp. , Dsaypus sp. , and Priodontes mazimus ) burrow into downed logs and create holes in the forest floor in search of their insect prey and as a means to escape predators, including man. Larger animals compact soils, whic h can locally increase erosion. White-lipped peccaries ( Tayassu pecari ) in neotropical forests can run in packs of from 50 to 300 individuals (Emmons 1999) followi ng well-worn feeding trails that are frequently revisited by the pack. These feed ing trails through th e forest become mudladen, compacted and devoid of vegetation. Sometimes hunters in PDS SÃ£o Salvador will sling a hammock between tw o trees in the forest alongside one of these trails, lying in wait for a pack of peccaries to pass. Often larger mammals that are close to sett lement areas are the first species to show significant population decline as a result of subsistence hunting (Redford and Robinson 1987; Robinson and Bennett 2000; Meijaard et al . 2005). Larger-bodied animals occur at lower densities than do smaller-bodied an imals (Robinson and Redford 1986) and hunters usually take the largest species in the forest (Redford 1992). Often the largest animals in tropical forests are frugivores (Redford 1992). One study conducted in the Peruvian Amazon indicated that forest fruits compri se 81% of the diet of red-brocket deer ( Mazama americana ), 66% for white-lipped peccary ( Tayassu pecari ), and 59% for collard peccary ( Tayassu tajacu ) (Bodmer 1989). All three of these species are commonly hunted in PDS SÃ£o Salvador. Because these species play an important role in the structuring of tropical fore sts through seed dispersal of fo rest fruits, the removal of
6 these animals through hunting could change the ecological dynamics of the tropical forest environment. Redford (1992) noted that in ar eas of tropical forest that ar e still largely intact (as is PDS SÃ£o Salvador), hunting remains the greatest threat to tropical diversity. Several authors write that hunting poses a greater threat to large tropical forest mammals than does the loss of habita t through timber harvesti ng (Bennett et al. 2002; Linkie et al. 2003; Wilkie and Carpenter 1999). Some species of animals may have become locally extinct in PDS SÃ£o Salvador as a result of hunting by th e peasantry (Fragoso et al. 2002); five of these species are on the International Uni on for Conservation of Nature and Natural Resources (IUCN) Red List of vulnerable, threatened, and extinct species (IUCN 2004). Peters (1996) cautioned that because the dist ribution and regenerati on of forest tree populations are frequently cont rolled by the action of seed dispersal agents of which many are animals, the failure to conserve vi able breeding populations of tropical forest animals could potentially be an irreversible management error. The loss of animal species in PDS SÃ£o Salvador could have negative effects for the sustainable functioning and structure of the forest ecosystem, and subseque ntly further impoverish local inhabitants in the area that rely on hunted wild game as an important source of food. The Peasant Household In the study of rural households, Chay anovÂ’s (1986) cl assic Russian study stated that the peasant economy has its own internal logic that is not dictated by capitalist relations of production, but rath er that the goal of the peasant household is simple reproduction. Household consumption takes precedent over house hold production, and is dependent on the life cycle of the house hold. Chayanov (1986: 78) writes, Â“The volume of a familyÂ’s activity depends entirely on the number of consumers and not at all
7 on number of workers.Â” As the number of i ndividuals in the house hold increases through the birth of more children, the productive efforts of th e household (hunting, fishing, agriculture) increase to accommodate the increas e in the number of mouths to feed. If the peasant household cannot meet its cust omary consumptive needs, the collective members of the household Â“tight en their beltsÂ” and consump tion momentarily decreases, or household labor is allocated to increase productive output to m eet the increase in consumptive needs of the household. Production often fluctuates throughout the year for the rural peasant household, and as much of th e productive output of the rural peasantry is consumed within the household, as pr oduction momentarily decreases, consumption correspondingly decreases for household members. James Scott (1976) added to the theory of the peasant economy by proposing the idea of the moral economy of the peasant that entails the social right to subsistence, and the norm of reciprocity across households with in the community. Reciprocity of goods, services, and labor across peas ant households acts as a no rmative social and economic leveling mechanism within the community. Ha rris (2000) writes that in Amazonia, kin desire to reside in clusters of households w ithin close proximity to one another, and that the reciprocal distribution of fish and wild game between households creates a sense of community and affirms the value of these rela tionships. Links between households are often maintained by women in the community who distribute the products of fish and game derived from the productive efforts of men. The recipro cal exchange of goods (meat, fish) and services (labor) in Kekchi Maya communities of Belize is similarly more frequent among household clusters that ar e densely grouped (Wilk 1997). Household
8 clusters that engage in a r eciprocal system of exchange amongst households can cope with diversified subsistence systems better than can is olated households. The peasantry diversify their productive efforts by engaging in hunting, fishing, agriculture, domestic livestoc k production, foraging, and wage labor. Domestic livestock production (pigs, cattle) is an important stock of potential income that can be sold in times of need by rural peasant households. Chibnik (1994) says that ribereÃ±o families (the peasantry living along rivers in tropical forests in Peru) prefer to sell their pigs to raise cash for medical emergencies rather th an becoming encumbered with debt from a bank loan. As Gudeman and Rivera (1990: 86) explain for the peasantry in Colombia, Â“Hogs are true Â‘piggy banks.Â’Â” In many areas in Brazil, the rural peas antry view cattle as an investment, and as such they are rarely slaughtered for household consumption, but are sold (and traded) to provide for other goods and services as the need arises within the household. Smaller domesticated animals incl uding chickens and ducks are raised for consumption by rural households in Brazil. The rural peasantry fundam entally diversifies their livelihood strategies to minimize risk, to adapt in order to survive, and as a means to create a brighter future for themselves and succeeding generations of household and community members (Ellis 2000). The Amazonian peasant combines mark et with non-market transactions in the diversity of its liveli hood strategies. At the periphery, market activities tend to be less frequent as the maintenance of the house hold economy and its members takes precedent over increased integration with the market system (Gudeman and Rivera 1990). Amazonian Peasantry Peasant reproductive and producti ve strategies (including hunting) in Amazonia are critically dependent on, and created in conj unction with, the seasonal nature of the
9 environments that they inhabit (Chibnik 1994; Harris 1998, 2000). In tropical forest environments there are generally two seasons Â–when it is dry and when it is rainingÂ–that affect the productive strategies employed by peasant households, and the allocation of household labor.1 Hunting is particularly important for the families living in the tropical forests of Amazonia during the rainy season when agricultural produc tion is minimal. Beckerman (1994) says in most cases in th e Amazonian environment, hunting is driven by fishing, as people hunt more when fishing re sources are harder to obtain (in the rainy season when the forests are flooded). Furthe rmore, the activity of fishing by women who reside in the household critically influences the frequency of hunting by the men of the household. If women within the household ar e involved in fishing, and the household has a steady intake of food (and protein) from th e catch of fish, then men in the household do not need to hunt to provide food for the tabl e. As mentioned above, there are, though, other compelling reasons that men engage in the practice of hunting than solely as a function of the utilitarian necessity of providing food for the family. Historically the peasantry in Amazoni a (often referred to in Brazil as caboclos ) has been blamed for everything from the destruc tion of the tropical forest environments of the Amazon, to the death and extinction of indigenous groups and communities in the region. Nugent (1993: 32, 45) says that ca boclos suffer because they were created in conjunction with the eradication of Amaz onian indigenous peoples. Many in the developmental and environmental commun ity saw caboclos as contaminating the traditional indigenous societies that had mana ged to maintain their Â“purityÂ” from the 1 During my visit to Acre in 2002, one elder resident of the community of Boa Vista in PDS SÃ£o Salvador joked that the dry season is when it rains every day and the rainy season is when is rains all day. Wet years and drought years have tremendous influence on the yearly agricultural cycle.
10 ravages of Western civilization. This overlooks the fact that no cultural or social group in history has remained beyond the influence and touch of neighboring and/or distant ethnic and racial groups (See Wolf 1982). Even the indigenous groups that have been Â“discoveredÂ” in the past few decades in th e remote corners of the Amazon are located where they are today and have developed over th e past few centuries largely as a result of the attempt to flee to the interior, leaving in their tracks the advancing priests, colonists, escaped slaves, conquistadors, and capitalis ts. Furthermore, culture and society and practice are not static, but in a constant pr ocess of co-evolution and adaptation based on the constraints and opportunities that present themselves to groups of people throughout history. It has become incr easingly difficult to ignore th e patterns and practices of resource use by caboclos in Amazonia, who co mprise the overwhelming majority of the population in the region, and whose use, and abuse, of natural resources in the Amazon basin has tremendous impact on the present and future ecological and social/cultural sustainability in these tr opical forests. The word caboclo has traditionally been used as a pejorative term to describe the rural inhabitants of the Amazon region in Brazil.2 Caboclos are generally described as lazy, dirty, uneducated, uncivilized folk. Nissy (1966: 159) described the caboclo as living Â“placidly, seemingly without ambitionÂ” an d they were not to be trusted. The label caboclo is applied in Brazil in much the sa me way as Â“redneckÂ” or Â“country-bumpkinÂ” is used in the United States. No one desires the label of caboclo with its historical negative connotations. The caboclo is also defined in terms of what he/she is notÂ–not indigenous, 2 The word caboclo is also frequently used by caboclos as a term of affection amongst themselves. In nearly every one of the ten communities in PDS SÃ£o Salv ador, there is at least one individual who is known by the nickname Caboclo.
11 not nationalÂ–rather than in positive term s of what actually constitutes the caboclo (Nugent 1993). Throughout the modern deve lopment of the Amazon, the culture of the caboclo has become invisible because their livelihoods and priorities have not been accounted for in the paradigm of deve lopment (Wagley 1973; Ross 1978b; Parker 1985; Nugent 1993). The Amazon caboclo has borrowed much from indigenous culture in the region and has developed production systems that ar e adapted to the phy sical and ecological environments of the tropical forests of the Amazon (Wagley 1973; Parker 1985). Generally the caboclo has striven for self-s ufficiency, and the caboclo family usually lives in isolated single-family dwellings. Ro ss (1978b) believes that the extractive rubber economy beginning in the nineteenth century in Brazil created the Amazon peasant caboclo. The caboclo economy is a combinati on of basic subsistence activities combined with some marketing of surplu s production in local and regi onal market centers. Caboclo production systems are characterized by thei r lack of technical sophistication and minimal use of credit or other inputs. Hunti ng, fishing, the gathering of forest resources, and small-scale agriculture centered on the pl anting of manioc orga nize the subsistence sector of caboclo life. Horticulture is of prime importance for some caboclo households, whereas other caboclo households rely on the extraction of non-timber forest products that dominate their socio-economic fra mework. Rubber, Brazil nuts, fish, farinha (manioc flour), forest fruits, and wood enco mpass the variety of the products produced and frequently sold by the caboclo household in the marketplace. There exist many similarities between th e livelihood strategies and cosmology of the Amazonian caboclo and indigenous inhabi tants in the region. Both groups rely on
12 non-market transactions, including collective labor arrangements and the sharing of wild game and fish, as an integral part of th e cultural-economic formation of the household and community. Both groups also frame thei r productive relationships in response to the consumptive needs and desires of the household rather than what is experienced in the purely capitalist system based on the maximiza tion of production and the reinvestment of capital to increase future produc tive (and consumptive) desires. Both groups have also utilized their exit-option (Hyden 1980), mean ing migration, as a means to resist the totalizing force of the capitalist system and the political goals of the State. Migration within the territorial patrimony and across nation states is a continual form of peasant resistance to the larger forms of dominati on that rain down upon the peasantry from the capitalist State. In terms of hunting, caboclos have borrowed much from indigenous culture. Indigenous Amazonians have an elaborate set of ritualized practice and belief incorporated within the practice of hunting and peopleÂ’s relationship with wild animals, and the rural peasantry in Brazil has corres pondingly adopted many of the practices and belief systems (Wagley 1976; Parker 1985 ; Smith 1996; Melo 2000; da Cunha and Almeida 2002). Many of the animals of the fore st that are classified as taboo to hunting and human consumption by Brazilian Indians ar e also taboo for the rural peasantry living within their midst. These taboo animals can temporarily cause the hunter to be unsuccessful in hunting wild game and furthe r cause bodily pain if they are eaten by humans. The polluting effects of women in reference to the hunt and wild game are recognized in indigenous and peasant soci ety. In many areas, pregnant and/or
13 menstruating women are removed from both hunt er and game meat after a successful hunt. Additionally, some animals are endowed w ith human characteristics, interacting with human beings beyond simply a predator Â–prey relationship. The classic example of this throughout the Amazon is with the boto encantado , or enchanted river dolphin (Wagley 1976; Slater 1994; Smith 1996; Me lo 2000). Not only are these animals taboo to hunting, but also they are said to oc casionally take human form, enticing humans into entering their world. Furthermore, there are a set of mythological creatures of the forest present in Amazonian folklore that protect wild game from humans, shaping the relationship of hunters with their prey in both indigenous and peasant communities in the Amazon region (Wagley 1976; Smith 1996; Melo 2000). Increasingly, though, as both of these groups of peoples, indigenous and peasant, are engaging in closer relations with the market, these cultural beliefs and pr actices are modified and many have been lost as a result (Smith 1996). Also, the decenter ing of the forest as a critical space of the construction of culture and liv elihoods (often accompanied with a move from rural areas to the urban environment), as much of the peasantry shifts from the reliance on the extraction of forest products to swidden ag riculture combined with the raising of livestock, has contributed to a break of tradit ion with the younger gene ration and a loss in much of the symbolic relationship of man with his/her environment. Hunting and the Peasantry Most of the literature on hunting in Amazonia has investigated the hunting practices of indigenous peopl es (Siskind 1973; Gross 1975; Ha wkes et al. 1982; Hurtado et al. 1985; Stearman 1989, 1991; Al vard 1995; Alvard et al. 1996) in spite of the fact that non-indigenous peasant populations ma ke up the overwhelming majority of the
14 population in Amazonia. Redford and Robi nsonÂ’s study (1987) is a notable exception that compares the effects of hunting on wild life populations by indigenous and colonists in the neotropics. Little is known, though, of the social dynamics of hunting practices, hunting strategy, and the importance of hunting in the rural livelihoods of peasants in Amazonia. Nugent (1993) attributes this to the scant attention paid by both researchers and the State to the forest management prac tices of peasant Amaz onians, historically Â“invisibleÂ” because of their non-exotic nature. Hunting is an integral part of the fore st-based activities of peasant populations throughout the tropical forests of the world, a Â“subsidy from natureÂ” (Hecht et. al 1988), and without it, Â“many other so-called sustaina ble activities such as rubber tapping, would not take placeÂ” (Redford 1992: 421). Crossculturally hunting has been important for rural households and communities for several reasons. The first, and often most compelling, reason for peasant populations th roughout the world is the hunting of wildlife as a source of food for the hous ehold. In many poor peasant households (especially those lacking domesticated anim als or the means to purchase meat), wild game serves as the most important, and cheap est, source of locally available protein. Meat provided from the hunting of wild game is an important source of food security for many rural households throughout the world (Bennett 2002; Rao and McGowan 2002). It has been estimated that the value of wild game harvested from the Amazon basin exceeds US$175 million per year (Tratado de Coope racion Amazonia 1995, cited in Rao and McGowan 2002). Some researchers believe that the hunting of wild game for food (rather than habitat loss) is the most important threat to the conserva tion of biodiversity in
15 the tropics worldwide in the next couple of decades (Robinson and Bennett 1996; Wilkie and Carpenter 1999). Hunting also holds symbolic gender signi ficance in rural communities, both in the developed and underdeveloped world. With in the household ec onomy, adolescent boys are taught by their fathers how to hunt and are socialized into hun ting as an important attribute of becoming a man (Stedman and He rbelein 2001). This symbolic function is seen in the dedication to hunting by elder men in the household as a way to initiate their sons and nephews into the ways of manhood. Â“Fathers who participate in hunting view the activity positively and would like to pass it down to their offspring, and the traditional Â‘malenessÂ’ of hunting makes sons the most reas onable targets of soci alizationÂ” (Stedman and Herbelein 2000: 606). This dedication to hunting varies with the presence or absence of adolescent boys and young men in the househ old, if the household is male-headed or female-headed, and if kin is living in the community. In her studies of the YuquÃ in Bolivia, Stearman (1989; 1991) reports that the prestige earned by hunting prowess and the social standing of men in the community derived from hunting takes precedent over the pursuit of fish by men, even though the later is a better strategy to optimize protein acquisition. In the United States, Kellert and Berry (1987) write that gender is one of the most important factors in determining at titudes and behavior towards wildlife and hunting. Simply stated, the act of the hunt is defined as an activity reserved mostly for men, and men tend to have more positive atti tudes towards wildlife and hunting than do women. Some authors state that subsistence re source strategies are based on the sexual division of labor that strives to account for offspring survival (Hurtado et al. 1985;
16 Jochim 1988). Women, as the primary care takers of children, engage in resource procurement activities that are risk averse as a means of enhancing the survival of their offspring. WomenÂ’s activities are performed in close proximity to their children (and the household), allowing for conti nuous and uninterrupted childc are. Men, who are not encumbered with childcare and many other domestic duties, specia lize in household food provision. Therefore, men are free to hunt, but women do not. Women pursue foods that are sub-optimal, supplying the reliable portion of the diet (gathering and agriculture), while men pursue resources of greater effici ency (higher protein content such as wild game), but lower reliability (Jochim 1988). Women stay closer to the settlement, avoiding energy demanding task s that require much travel, and emphasize reliable production to assure a stable energy and nutrient intake. A persistent feminist criti que of household gender relations is related to womenÂ’s subordination through the sexual division of labor. Bouquet (1984: 155) writes of rural England that it is Â“Â…through th e sexual division of labor within the farm household, whereby women are responsible for reproduc ing the conditions necessary for men to engage in specialist producti on.Â” The focal point of wome nÂ’s productive activities is seen as based on their special role in th e reproduction of the (household and community) labor force. Male domination develops around the need to control reproduction (BenerÃa 1979; Goldelier 1986) and Â“It is in the reproductive sphere of the household that the primary relations of subordination/dominati on between the sexes ar e locatedÂ” (BenerÃa 1979: 209-210). Women are seen as relegate d primarily to the domestic realm of the household with the household re productive functions of moth ering, caring, and nurturing, whereas men are free to pursue productive ende avors that accrue highe r individual status.
17 As prestige is often displayed in peasant and indigenous communities through hunting prowess (Stearman 1989, 1991) the most prolif ic hunters are awarded with reproductive benefits in the community including decrease d infanticide of their offspring, increased access to extramarital affairs, and increas ed access to women in the community as potential wives (Kaplan and Hill 1985; Thiel 1994). The resource procurement strategies and activities of women are not afforded si milar status in the household or larger community. Many individuals also hunt because they en joy it. For some, hunting is a form of recreation as an opportunity to escape the pressures of cr eating a livelihood, or as a means to find solitude within the forest environment (Kensinger 1981; Fine 2000). Some households are more emotionally committed to hunting as a result of the household membersÂ’ desire to continue with the prac tice of hunting as an integral part of the livelihood strategy of the household. These househol ds may continue to hunt even if they have other readily available s ources of food and protein as a substitute for the meat of wild game. Poor rural households devote a lot of atte ntion to the maintenance of kin networks in the community (Ellis 2000). These households may be deficient in stocks of economic and political capital, but the social capital th at they create thro ugh kin living in other households in the community is vital to their survival. The exchange of meat from wild game across households is an important social mechanism that binds households together within the community. It has been suggested that men hunt large animals and share the meat with others in the community primarily for the social benefits received in the exchange (Hawkes 1993; Kent 1993). The reci procal sharing of meat from the hunt
18 across households fits within the moral economy of obliga tions in the community in peasant and indigenous societies (Scott 1976). This non-market form of peasant and indigenous exchange helps to insure the su rvival of community members (Ellis 1993), and thus the community as a whole (Anderson 1994). Harris (2000) notes that, in the Brazilia n Amazon, it is often the women of the household who redistribute the meat from w ild game brought into the household by their husbands and sons to other households (usu ally kin related) in the community. The reciprocal distribution of meat from wild game remains an important mechanism in the creation of social cohesion between househol ds in indigenous and peasant communities in the tropics, although the dist ribution of wild game is c ontext specific and depends on such things as the costs of storage and the defense of wild game from others (Hegmon 1991; Hawkes 1993), the size of the family (Aspelin 1979; Kaplan and Hill 1985; Stearman 1989), and the obligations to kin a nd other social norms of sharing between households in the community (Aspelin 1979; Kensinger 1983; Speth 1988). Meat sharing in Amazonia across households is normally kin based. Thus, though not participating in the act of hunting, as do men, women in the household and community are important actors in the system of hunti ng through their function in meat exchange across households within the community, and in the preparation and cooking of the game meat that is brought into the household. Wild game carries special status in ma ny peasant and indigenous communities in Amazonia. Women in Sharanahua society in eastern Peru are said to Â“cry for meatÂ”, sending their men out to hunt and withholding sexual favors in the exchange for meat (Siskind 1973). The YuquÃ of Bolivia have a word to designate Â“meat hungerÂ”
19 (Stearman 1989) as do the Achuar in Ecuador (Descola 1996). In many rural communities worldwide, Â“meat and its acquisi tion tend to have a value disproportionate to their economic and nutritional contribution compared to plant and fish resourcesÂ” (Kent 1989: 7). In PDS SÃ£o Salvador, households desire to have at least a small portion of meat (or fish) with every meal. Wome n in the household though, are often the most obvious targets of inequalities in food sharing. Women receive smaller shares of meat, and taboos often prohibit women from consum ing meat at certain times throughout the year (e.g. during menstruation and/or pregnancy) (Speth 1988). Within the peasant households of PDS SÃ£o Salvador , at meal time, men choose the meat and fish they wish to consume prior to women choosing the food they will eat.3 Some authors (Jorgenson 2000; Lee 2000; Bennett 2002) have suggested that the domestication of animals by th e household is a viable altern ative to the consumption of meat from wild game. Furthermore, the av ailability of fishing resources lessens the reliance on hunted wild game as a source of protein (Ross 1978a, 1987; Stearman 1989). It has been stated that governments can al so work towards reducing the need of meat from wild animals for rural peoples by providi ng alternative sources of protein including domesticated animals, reared fish, or plan t products (Meijaard et al. 2005: 183). The increased income of households has also been suggested to decrease the consumption of wild game in Latin America as households freq uently substitute the meat from wild game for meat of domesticated animals (St earman and Redford 1995; Bennett and Robinson 2000). As incomes increase, many households replace wild foods with foods they purchase with money. 3 In addition, in some of the households I stayed w ith, men and adolescent boys ate at the dinner table, while women and younger children ate their meals sitting on the floor.
20 Ironically, though, wild game often holds th e double connotation of stigma for the lower class diet, yet a status symbol for th e elite gourmet (Wilk 1997). The middle class, that does not want to be lumped together with the lower class, identifies their superiority with the ability to pay for items (including meat); therefore the consumption of wild game signifies poverty and the inability to pa y for market items. The elite show their superiority from the middle class by paying hi gh prices for foods th at the middle class do not eat. An example of this occurs with th e consumption of squab (pigeon) in expensive restaurants in the United States. Pigeons ar e raised for home consumption by some poor households, yet also comprise part of the diet of the elite. Diagram of Peasant Hunting Decision-making on resource use and liveli hood strategy in all households is based on a combination of factors or iginating from within the household, as well as beyond the individual household. Households do not exis t in isolation from one another, as cultural and social norms of appropriate behavior are enacted in the relationships between households within the community. Households are also critically affected by their relationship with the market. A distinguish ing feature of peasant households, though, is their partial, or incomplete, integration with the market (Ellis 1993) where production is often simultaneously prepared for the mark et and home consumption. Furthermore, peasant households located at the periphery of the market (as are those in PDS SÃ£o Salvador) show less market in tegration than peasant households that are encaptured by the market system (Gudeman and Rivera 1990) . Finally, the natura l environment plays a role in the use of resources by peasant households. Hunting by the peasantry reflects combined processes and relationships that occur within the household and outside of the individual household. As hunting is integral to
21 the livelihoods of peasant households, this study measures hunting out put at the level of the household. Figure 1, below, diagrams the factors that influence hunting by the peasant household in PDS SÃ£o Salvador. Figure 1: Diagram of Peasant Hunting in PDS SÃ£o Salvador Within the peasant household, there are thr ee sets of variable s that critically influence the output of hunting: material, ge nder, and symbolic. Poor peasant households often rely on wild game to supply a larg e portion of the protein intake for household members. As such, hunting is critical in providing food security for the household. Hunting is also defined as a male activity within peasant households and communities. The process of boys becoming men in the commun ity is integrally tied to their learning the ways of hunting. Finally, the livelihood strategy of some households reflects an emotional commitment by household members to persist with hunting as a way of life.
22 This reflects the desire of individual househol ds to continue to hunt because they enjoy this activity, maintaining this practice in their repertoire of livelihood practices. Across households, hunting plays a part in the creation and maintenance of social bonds. This is clearly seen in the recipr ocal exchange of wild game. Community obligations are met through th e exchange of meat amongst households as households are joined together with the perfor mance of meat exchange. The reciprocal exchange of wild game also insures that all households in th e community have some meat to eat. Meat exchange is a non-market transaction that is an essential aspect of the peasant socioeconomic system. As stated above, market integration is in complete in peasant households (indicated in Figure 1 with a dashed arrow), especially so at the periphery of the market. In PDS SÃ£o Salvador, market influence on hunting output is minimal in comparison to other factors. The natural environment plays a ro le mostly in the timing of hunting, and with the techniques used by hunters in the pursu it of wild game in the settlement. The seasonal nature of the tropical forest envir onment, evenly divided between a wet and dry season, influences resource use and livelihood strategy, and how hunters search for wild game. Intra-household Hypotheses on Hunting In this study, I argue that the hunting practices and be liefs of households in PDS SÃ£o Salvador can be measured and tested by investigating the influence of the combination of material conditions of the hous ehold and gender relationships within the household, and social obligations and relations hips across households in the community. As wild game is a critical source of protei n and calories for reside nts of the area, the number of individuals residing within the household, if a househol d has domesticated
23 animals or not, the availability of outside sources of income for household members, and if women of the household fish or not, shoul d be factors in determining how hunting is practiced. Therefore, I hypothesize that: H1: Households with lots of children hunt more often than households with fewer children. H2: Households that lack domesticated an imals hunt more often than households that have domesticated animals. H3: Households that lack an outside sour ce of income (wages, retirement checks) hunt more often than households that ha ve an outside source of income. Hypothesis 1 is based on the idea that the more mouths to feed within the household, the greater the effort needed to s ecure food to feed household members. This is based on the Chayanovian thesis that household consumption drives household production. Hypotheses 2 and 3 re flect the substitution of wild game by other sources of meat. Households that raise animals do not have as great a necessity to enter the forest in the pursuit of wild game as do households that do not ente rtain this activity. Also, households that have a secure source of inco me may be able to pur chase meat instead of having to hunt to obtain meat for the household.4 In addition to the material realities of th e household, gender is a key variable in the practice of the hunt in peasant households. First, hunting is fr equently defined as a male role and responsibility, throughout most peasan t and indigenous communities worldwide. Second, the emotional commitment of the hous ehold tied with the practice of hunting 4 Wilk (1997:156) writes that the substitute effect of rural households purchasing domestically raised meat instead of relying on the meat of wild game is not determined solely by economic considerations, but also by the status assigned to wild foods and food purchased in the market economy. The consumption of wild game by rural peoples is often stigmatized as charact eristic of lower class diets. Thus, the ability to purchase market produced foods (in this case domesticated meat) signifies increased status for the consumers of these products. These are demand driven responses by rural households as poor rural peoples replace wild foods with foods that th ey purchase with cash.
24 reflects the socialization process of b oys becoming men within the household and community. Just as women are defined as th e domestic caretakers of the household and instilled with the majority of the responsibil ity for childcare, hunti ng encompasses part of the definition of adult male status, with me n invested in primarily productive pursuits (including hunting) within th e household. Third, gender ro les and responsibilities of household members critically affect the outpu t of hunting by the household. It is argued here that the decision to hunt, by whom, and when, does not reside with the individual hunter alone, but is based on the combined productive and reproducti ve efforts of all household members. Therefore, I hypothesize that: H4: Male-headed households hunt more often than female headed households. H5: Households that have adult males and hunting age adolescent boys hunt more often than households that do not have adult ma les and hunting age adolescent boys. H6: Households in which women do not fi sh hunt more often than households where women do fish. Hypotheses 4, 5, and 6 result from gender de fined roles of men and women in the peasant household. Hunting practice of househol d members is, in part, the result of the coordination of household activities by the men and women that are members of the household. Hypothesis 6 reflects a combination of the substitu tion effect described above (H2, H3), but in this case the exchange of fi sh for wild game instead of purchased meat for wild game, and the productive and reproduc tive responsibilities of women within the household. The practice of hunting within the house hold is also dependent on the emotional commitment of the members of the househol d to hunting. This cu ltural component of hunting reflects the desires of household memb ers to continue with the practice of
25 hunting as an integral part of the collective liv elihood strategy of the household. This partially is a result of the householdÂ’s gende r dynamics, as hunting is associated with male status within the hous ehold and community, and men ma y feel compelled to hunt due to cultural norms, yet it is also depe ndent on whether hunters within the household enjoy hunting and wish to continue this practice. Therefor e, I hypothesize that: H7: Households that are more emotionally committed to hunting hunt more often than households that are less emotionally committed to hunting. Inter-household Hypotheses on Hunting The practice of hunting is not solely expl ained by intra-household processes, but also reflects how relationships across households within the co mmunity come into play in the dynamics of hunting. Community norms, rights, and responsibilities shape how hunting is practiced by househol ds. Frequently, Â“the intern al composition and division of labour [sic] within productiv e householdsÂ…are largely de termined by the external relations of households to each other and to other social groupsÂ” (Friedman 1980: 159). Social obligations tied to r eciprocity between households are important in maintaining healthy and happy communities. Peasant households in the Amazon have been characterized by the clustering of familial units in the creation of community (Harris 2000). These reciprocal bonds across house holds, though, are stronger with households tied together by kin relations than with thos e households lacking such bonds. Therefore, I hypothesize that: H8: Households that are grouped together in clusters within the community hunt more often than households that are not grouped together in cl usters within the community. H9: Households that have family living w ithin the community hunt more often than households that do not have family living within the community.
26 Hypotheses 8 and 9 are influenced by th e sharing of resour ces across peasant households in the Amazon. In addition, we w ould expect to find that blood ties between households are stronger than ot her forms of inter-household b onds within the community. Dissertation Layout In the following chapter I begin with a presentation of the history of PDS SÃ£o Salvador. A historical political ecologi cal framework is used to understand the transformation of resource use in the area a nd its effect on household livelihoods. I will first trace the importance of rubber in the de velopment of the state of Acre and its subsequent influence on the social-econo mic framework of peasant households and livelihoods in PDS SÃ£o Salvador. The layout of the ten communities in the area reflects the history of rubber exploitati on in Acre and the history of the development of PDS SÃ£o Salvador as will be explained in the next chapter. Finally, the transformation from livelihoods based on extraction of forest pr oducts (rubber) to ones based on small-scale agriculture (farinha) is explained. This is an important shift in socio-cultural practices of resource use by local residents, with potentially important co nsequences for the resource base of the forest environment. Following the historical description of the area I will introduce the current state of affairs in Acre, and the research site, PDS SÃ£ o Salvador. The state of Acre and PDS SÃ£o Salvador are described in terms of their bi ophysical and geographica l characteristics. The seasonal nature of resour ce use in PDS SÃ£o Salvador is discussed, an important constraint in the timing of hun ting in the area. The chapter concludes with a discussion on the importance of travel and the gender specific natu re of hunting travel. Chapter 3 presents the methodology (bot h research design and methods) used throughout this study. Quantitative methods were used to test res earch hypotheses, and
27 qualitative methods were used to describe hunting practices and as supporting material towards hypotheses testing. The research de sign and methods used in this project investigated both the effects of intrahousehold and interhousehold (between households) relationships on the practi ce of hunting in the area. Chapter 4 will examine hunting within the household of the peasantry in the study site, specifically, how the mate rial realities of the house hold, gender relationships of household members (intra-household), and the emotional commitment of household members to hunting critically affect th e practice of the hunt . The roles and responsibilities of men and women in the househ old are examined in light of their effects on the practice of the hunt. The timing of re source use and the seasonality of resource use are also reviewed. A de scription of hunting practice is given and the intra-household hypotheses enumerated above are tested. Chapter 5 focuses on relationships amongs t households in the community (interhousehold) and how this influences hunting in the seringal. The practice of the hunt is one of the key ways in which households are tied together across the community, insuring the survival of all households in the community as a whole, and acting to create a sense of community amongst the often widely dispersed households of the community. The practice of group hunting involving memb ers of different households within the community and the exchange of wild game between households is discussed. Kin relations across households play a key role in the nature of group hunts and meat exchange across households. Inter-household hypothe ses are also tested in this chapter.
28 Finally, chapter 6 presents my summary of findings and conclusions. I will also discuss the implications of this proj ect for anthropology, for conservation and development in the area, and the implications of the study findings for gender studies.
29 CHAPTER 2 THE HISTORY OF PDS SÃƒO SAL VADOR AND THE RESEARCH SITE This chapter begins by tracing the historical development of the state of Acre since the late nineteenth century and that of PDS SÃ£o Salvador since the 1930s. A political ecological approach is used to view th e transformation of resource use by local inhabitants. Changes in subsistence strategi es by local resource users over time can lead to significant modifications of the local enviro nment. This understanding is essential in the implementation of effective environmental policies that respect both the natural environment and the livelihood strategies of rural peoples whose daily sustenance is dependent upon the use of natural resources. Communities in PDS SÃ£o Salvador have undergone dramatic, often rapid, change in socio-political structures that have transformed household livelihoods. Though still marginalized from the benefits of the capit alist system, residents of PDS SÃ£o Salvador maintain active engagement with the market sy stem. Whereas for the majority of the past century, the livelihoods of the peasantry we re based on the extractive rubber economy, by the 1990s swidden agriculture centered on the production of farinha for market, combined with the raising of livestock, dominated the household economy. Household livelihood strategies have passed from the cultu re of rubber to the culture of farinha in this part of western Acre.
30 Since its incorporation into Brazil in the early 1900s1, Acre has existed on the margins of Brazil and held litt le importance in the political -economic development of the country until the end of the nineteenth centu ry when rubber suddenly became a valuable commodity on the world market. The devel opment of the rubber economy has defined Acre more than any other Amazonian state in Brazil. Bakx (1988) stated that Acre was colonized by Brazilians with the sole intent to extract latex from the forest to produce rubber. After the collapse of the rubber economy in Brazil by the 1920s, Acre returned to its historical state of isolati on from the hearts and minds of the nation for several decades. The 1970s witnessed the infusion of cattle interests in the stat e of Acre (and the rest of Amazonia) and the emergence of the rubber tapper ( seringueiro ) movement spearheaded by Chico Mendes and other leaders. The in ternational conservation and development community embraced the seringueiros and their resistance to the conversion of tropical forestlands to pasture, even if their fellow countrymen were less likely to do so. Even today very few people in Brazil have ever vent ured to Acre and further, most have little desire to. If Acre as a whole is isolated from the re st of the country, then western Acre where PDS SÃ£o Salvador sits, is even further removed from the nation. The political sensibilities of the se ringueiro movement within Acre ha ve been largely limited to the eastern portion of the state, resulting in increased nationa l and international attention directed towards the East at the expense of the West. The lack of infrastructure, critically a paved highway connecting the West w ith eastern Acre, has severely limited communication and supplies, and separated individuals and families from the benefits of 1 The area that today comprises the state of Acre was ceded by Bolivia to Brazil with the Treaty of PetrÃ³polis in 1903 (Nissy 1966).
31 development projects concentrated in the eastern portion of the state. As of 2005, western Acre had only 150km of paved roads (Campbell 2005). The research site, PDS SÃ£o Salvador, is located on the western fringe of Brazil bordering the countries of Peru and Bolivia. Th e state of Acre, a sma ll state in terms of land base and population of the stat e of Acre, is isolated from the rest of the country. In the very western portion of the state, about as far west as one can go and still be within the territorial limits of Brazil, sits PDS SÃ£ o Salvador. I will de scribe where PDS SÃ£o Salvador is located as well as the biophysical characteristics of the settlement, along with a brief discussion of the infr astructure development (or lack thereof) in the area. Following, comes a description of the yearly calendar of household resource procurement activities. Livelihood strategies are significan tly affected by climatic conditions in the seringal. Finally, the importance of travel for th e residents in PDS SÃ£o Salvador is mentioned. Travel in time and through space is a highly gender regulated activity in the seringal. Hunting travel is an importa nt mechanism in creating alliances across households within the community. Men, who are the hunters, are afforded the advantages that the hunter-tra veler creates in the construc tion of these socio-political bonds, whereas women, who are limited in their traveling, are thus limited in their participation with the hunt and the co rresponding creation of bonds with other households. Women, though, are afforded other opportunities in the creation and maintenance of social networks across house holds. In terms of hunting, the reciprocal sharing of wild game between households in the community is often performed by women of the household.
32 The History of Rubber in the Development of the State of Acre In order to understand the history of PDS SÃ£o Salvador and the socio-cultural practices and beliefs, and resour ce practices of its residents today, it is first important to understand the importance of the history of rubb er in the socio-cultural milieu of Acre. Throughout the past century, the extractive fo rest economy centered on the production of rubber has dominated socio-polit ical and economic structures and livelihoods throughout the state of Acre. This legacy continues in the stateÂ’s political system with the current Â“forest governmentÂ” of Gover nor Jorge Viana that seeks to prevent conversion of forestland through the protection and support of extractive forest activ ities (other than rubber). Charles Goodyear discovered the proce ss of vulcanization in 1839 that allowed latex to maintain its consistency despite changes in temperature (Dean 1987). This process transformed rubber in to a valued commodity on th e worldwide market and the scramble for Amazonian rubber took off be ginning in the 1850s. In the late 1800s, rubber was exported from Brazil to supply the bicycle craze that had swept Europe. By the early twentieth century, th e production of FordÂ’s Model-T redirected rubber exports from Brazil to the burgeoning market for tires in the United States. The Amazonian rubber economy was based on the aviamento system of supply, credit, and control of peasant labor (Dean 1987; Barham and Coomes 1994; Coomes and Barham 1994). In this system seringueiros (rubber tappers) many of them poor, landless migrants from the drought-plagued Northeastern sertÃ£o , existed in a merciless system of debt slavery to the local barracÃ£o (trading post) usually oper ated by the local owner of the rubber estate concession ( dono de seringal ). Each seringueiro lived within his colocaÃ§Ã£o , the expanse of land within the seringal th at contained the rubber trees that he
33 worked and was expected to tap for latex, in order to pay the annual rent to the dono de seringal for use of the piece of land. The seringueiros would transport their latex to the local barracÃ£o where they would be extended credit and supplies of basic necessities at vastly inflated prices, payabl e with the next installment of latex from the forest. The local barracÃ£o would then transport the sm oked latex by river downstream to one of the larger Amazonian cities for eventual export ou t of the port of BelÃ©m at the mouth of the Amazon. The donos de seringais often prohibited th e seringueiros from engaging in any subsistence activities that competed with thei r rubber tapping activities. They preferred to recruit and hire single, poor men from the impoverished Northeastern region of Brazil to tap rubber, and women were scarce in the state of Acre during this first rubber boom. When these single male rubber tappers demande d that they be supplied with women, the large trading companies in Manaus and BelÃ©m sent women (many from the citiesÂ’ brothels) to the seringais in Acre (Campbell 1996). These single, male seringueiros lived in dispersed dwellings in th e tropical forests of the Brazi lian Amazon, which hindered the development of any collective pol itical strategy within the cl ass of seringueiros. Wages were unknown in the aviamento system, and a fe w rubber barons at the top of the rubber pyramid in Manaus and Belem profited mightily at the expense of the thousands of marginalized and frequently starving seringueiros in the Amazon. The present Brazilian state of Acre was annexed from Bolivia in 19032, an area that in 1899 supplied 60% of all the rubber produ ced in Amazonia (Hall 1989: 2). From 1900 to 1913, rubber accounted for one-third of BrazilÂ’s exports (Skidmore and Smith 1992: 2 See Church (1904) and Tambs (1966) for an expl anation of how Acre was acquired by Brazil from Bolivia.
34 153). The heyday of the Amazonian boom lasted from 1870 until 1912 when Asian production of cultivated rubber surpassed that of Brazilian production.3 Brazilian rubber production experienced a brief reprise during the Second World War as the Axis powers controlled rubber production in Asia, and the United States was forced to look to its neighbors to the south to s upply its rubber needs for the war effort. Following the termination of World War II, the Asian r ubber plantations in Malaysia and Ceylon resumed their place as the centers of worldwide rubber production. When the rubber economy in Brazil collapsed by the 1920s4, many donos de seringal simply abandoned their rubber estates as the economic value of these landholdings essentially ceased to exist. Th e population in the state of Acre fell by fourteen percent from 1920 to 1940 (Bakx 1988: 150). The seringueiros that remained behind, poor and without the means to migrate to other areas, were free to engage in a variety of subsistence activi ties including swidden agricu lture, hunting, fishing, and the small-scale extraction of forest products, incl uding rubber. These caboclos, a mixture of the culture of the migrants from the Northeast with the decultured descendents of Amazonian indigenous communities, were larg ely ignored by the State until after World War II. The collapse of the rubber economy in Amazonia by the 1920s transformed gender relations in the seringal and in the state of Acre as a whol e. As the donos de seringais 3 The collapse of the Amazonian r ubber economy has sometimes been credited to the Englishman Sir Henry Wickham who Â“smuggledÂ” rubber seeds out of Brazil in 1876 to be planted in Kew gardens for later establishment in the British colonies in Malaysia and Ceylon (Dean 1987). 4 The price per pound of Brazilian rubber was US$1.91 in 1910, falling to US$0.20 by 1922 (Bakx 1988: 149).
35 abandoned their rubber estates5, an autonomous rubber tapping strategy emerged, replacing the traditional debt-peonage av iamento system of rubber extraction. The single, solitary man tapping rubber became a th ing of the past as more women moved to Acre and seringueiro families became established. The family replaced the patrÃ£o as the organizational center of life in the seri ngal (Wolff 1999). Rubber extraction began to decline in relative importance for the seringue iro family as other subsistence productive activities gained in importance. These subsistence activitie s included hunting and fishing6, swidden agriculture, the collection and extraction of other fo rest products such as Brazil nut ( Bertholletia excelsa ), fruits (aÃ§aiÂ– Euterpe precatoria , cupuaÃ§uÂ– Theobroma grandiflorum ), palm hearts ( Euterpe edulis) , and small-scale animal husbandry. Consequently, the socio-economic life of the p easantry that had been deeply integrated with the market (though marginalized from its benefits) during the heyday of the production of rubber, transformed, lapsing to a subsistence mode of production, and becoming a Â“traditional populationÂ” (Wolff 1999 : 106) Â– a shift to livelihoods based on production for subsistence, rather than for ma rket. Additionally, for many rural peasant households in Acre, the dependence on the fore st as the critical site for the economic well-being of the family decreased as househol ds increasing turned away from extraction of latex to produce rubber, and increasingly towards swidden agriculture and the raising of domestic livestock. 5 Land was seldom bought and sold on the market in Amazonia during the rubber boom of the nineteenth and twentieth centuries as land ownership was de fact o, based on living on, and working, a piece of land. Many of the donos de serginais did not live within the bounds of the seringal, but financed the operation of production of rubber in the seringal. 6 Hunting and fishing by the seringueiros in the seringal were traditionally tolerated by the patrÃ£o as long as they did not interfere with the production of rubber.
36 Post World War II Amazonian Development With the fall of the rubber economy post-W orld War II, the cultural practices and livelihood strategies of indigenous peoples a nd the rural peasantry based on the extraction of forest products were looked upon as b ackward by the Brazilian government. The Amazon region was viewed as a vast, empty space full of natural resources, that was in dire need of modernization if the nation was to reach its full potential. Furthermore, this Â“empty landÂ” could be used to settle the la ndless peasantry, thus diffusing tension based on the gross inequality in la nd ownership throughout the c ountry. Indigenous peoples and the rural peasantry suffered due to the de velopment drive by the St ate and other elites in the tropical forests of the Amazon. Amazonian development from the 1950s onward witnessed the increasing concentration in la nd ownership, increasing violence directed at indigenous and caboclo communitie s as the elite gobbled up their parcels of land, and the increased use of a monetary economy with the introduction of c onsumer goods (Moran 1981; Moran 1983; Hall 1989; Schmink and Wood 1992; Little 2001). In 1953, the Superintendency for the Economic Valorisation of Amazonia (SPVEA) was created to stimulate the so cio-economic development of the Amazon region (Moran 1981; Hall 1989; Schmink and Wood 1992). The development drive was to be carried out through a seri es of five-year plans. The three top priorities of SPVEA included agricultural development to make th e region self-sufficient in the production of food and to increase produc tion of raw materials (wood, fish, game) for export and domestic consumption, to improve river trans port and port facilities, and lastly to pay attention to the regionÂ’s hea lth problems (Hall 1989). As part of the scheme to encourage the development of the interior of the count ry, the capital was move d from the coast in Rio de Janeiro to the interior of BrasÃlia. SPVEA also financed the construction of the
37 2000-kilometer highway that linked the city of BelÃ©m at the mouth of the Amazon to the new capital of BrasÃlia. Thousands of poor, landl ess settlers in search of a better life followed the road as it was constructed. Through the industrial in centives initiated by SPVEA, local elites were able to obtain large swaths of land that were used for cattle ranching or industrial agricu lture. Indian lands were frequently encroached upon (Goodland and Irwin 1975; Davis 1977; Br anford 1985; Treece 1990) and many other subsistence landholders lost their land as me rchants began to seize property in lieu of payment for outstanding debts (Schmink and Wood 1992). Frequently the land that was seized was rented back to the previous landowner, who continued his subsistence activities, yet without title to the land. The militarization of the Amazon began with the military coup in 1964 that overthrew the left leaning government of JoÃ£o Goulart. The military leaders instituted an aggressive policy to occupy and colonize th e Amazon frontier, as there was a shared vision within the military of the need to permanently settle the Amazon. SPVEA was replaced with the Superintendency for the Development of Amazonia (SUDAM) in charge of instituting Â‘Operation AmazoniaÂ’, th e governmentÂ’s development plan for the Amazon basin (Moran 1981; Lisansky 1985; Hall 1989; Almeida 1992; Schmink and Wood 1992). The governmentÂ’s goals of Amazo nian development we re threefold: to improve BrazilÂ’s foreign exchange with the in creased exploration and exploitation of the Amazon, to promote national integration, and to reduce social tensi ons in the Northeast by providing a settlement option for the poor and landless of this region (Hall 1989).7 7 Brazilian president EmÃlio MÃ©dici stated in 1970 that the primary motive of Amazonian development was, Â“to give men without land [northeas terners] a land without men [Amaz on]Â” (Moran 1981: 75). Thus, the government conveniently bypassed the issue of land reform.
38 Credit from the Bank of Amazonia (BASA) subsidized land acquisition schemes, as cattle ranching was aggressi vely promoted. The World Bank encouraged the Brazilian governmentÂ’s cattle mode of development w ith increasing loans for livestock production in the Amazon.8 By 1973, Brazil was estimated to have the third largest he rd of cattle in the world, surpassed only by the United Stat es and the Soviet Union (Davis 1977: 128). Land was increasingly in the hands of an elite minority, and the Gini coefficients for land ownership in the 1960s and 1970s increased in all Amazonian states except one (Rosenblatt 1992: 9). Small peasant farmers in creasingly lost their land (often violently) and were forced to move deeper into the Am azon to stake out a living. They worked as agricultural laborers on the large cattle ranches and agricultural estates, or they migrated to cities. There was a clear correlati on in Amazonia between the increase in deforestation, the increase in the concentration of landholdi ngs by rural elites, and rural violence (Schwartzman 1992). Furthermore, high national inflation led to Amazonian land holding speculative valu e, causing money to pour in to the region from the industrialized South, pushing the price of land beyond the means of the caboclo peasantry. Overall the militaryÂ’s devel opment efforts in the Amazon caused the peasantry to be increasingly removed from their lands while pushing them further westward into remote corners of the Amazon su ch as the state of Acre that borders Peru and Bolivia.9 The pace of environmental destruct ion escalated especially during the 1970s, the decade that Almeida (1992) labeled the Â“Decade of Colonization.Â” By 1997, it 8 The military regime in Brazil was able to finance its policies through the generous lending practices of the international financial sector in the developed world. By 1983, Brazil had acquired the largest foreign debt in the world between US$87 billion and US$100 billion (Skidmore and Smith 1992: 182). 9 Although the government initiated a series of directed settlement schemes for the peasantry in Amazonia, the overwhelming majority of peasant colonization was spontaneous, often as a result of landless migrants following roads as they were constructed throughout Amazonia (Smith 1982; Lisansky 1985; Hall 1989).
39 was estimated that approximately 13 percent of the Brazilian Amazon rainforest had been destroyed (Hall 2000: 100). Violence also became an increasing part of the social fabric in the Amazon as conflict intensified between the caboclo a nd large-scale development interests. Pistoleiros (hired gunmen) were ordered by the larg e landowners to drive small farmers off the land that they occupied. Diegue s (1992: 24) estimates that around 1,100 rural workers were killed in land conflicts in Brazil during the 1970s and 1980s. Peasant resistance movements from below increasingl y challenged the developmental paradigm of the government and elite interests (S chmink and Wood 1992). The return of democracy in 1985 allowed a space for these mo vements to articulate their competing claims for the resources of the nation. Th e Movement of the Landless (MST) demanding land reform acquired national attention by e ngaging in a series of well publicized land occupations. In 1985, the National Council of Rubber Tappers was created to protect seringueiros from the demands and desires of the ranchers who were interested in the conversion of forestland to pasture. Thei r leader Chico Mendes , a caboclo seringueiro from eastern Acre, gained international rec ognition and the support of environmental and human rights organizations worldwide as the seri ngueiros strove to protect the forest that they had traditionally inhabited. Unfortunate ly, cattle ranchers assassinated Mendes in 1988, but the movement of seringueiros continue d its mission. The concept of extractive reserves proposed by seringueiros was signed into law by the Brazilian government in 1990 and has been advocated as an appropriate developmental model that respects social and environmental sustainability in tr opical forests throughout the world.
40 Due to the rubber tapper movement engineered by Chico Mendes (and other leaders), and the national and international a ttention that it garnered, Acre became a site of interest in environmental po licy and a frame of reference in the fight to protect tropical forests from destruction and conversion to other land use types. International and national interest in the social movement of the rubber tapper s spurred increased developmental aid to the state as well as a flood of researchers in tent on investigating how sustainable forests and su stainable livelihoods could co -exist. Furthermore, Jorge Viana, trained as a forester, was elected govern or of the state of Ac re in 1998 and labeled his administration the Â“forest governmentÂ” si gnifying his commitment to protecting the stateÂ’s forests and the people who reside with in them. The goal of the Viana government has been to emphasize the importance of the forest environment in the social, cultural, and economic history and practice of the peopl es of Acre. Policies were directed at generating wealth thro ugh the maintenance of standing forests throughout the state. Additionally, Marina da Silva, who played a pivotal role in the social-environmental movement of rubber tappers in the state of Acre, and who represented the state as a senator, was hand picked in 2002 by the presiden t of Brazil, Lula da Silva, to head the national Ministry of th e Environment. Although isolated geographically from the rest of Brazil, over the past two decades the state of Acre regained the attention of the national and international community. Rubber in the Development of PDS SÃ£o Salvador Prior to being named the first sustainable development settlement within Brazil in 2000, Projeto de Desenvolvimento SustentÃ¡ vel (PDS) SÃ£o Salvador was commonly referred to as Seringal SÃ£o Salvador in defe rence to its use and designation as a rubber tapping estate. The word seringal is used in Brazil to desi gnate a current, or former
41 rubber estate. Even today, the residents of PDS SÃ£o Salvador commonl y refer to the area as Seringal SÃ£o Salvador rather than the name designated by the State in the year 2000. The rubber system developed slightly di fferently in PDS SÃ£o Salvador. Although the mechanics of the aviamento rubber syst em were much the same, rubber production began much later in the area, well after the initial peak of rubber activity throughout Acre at the turn of the past century, commencing only in the early 1930s. Possibly the process occurred later as a result of the difficulty of reaching the area, which lies on the extreme western border of the county with Peru. Even in 2005 there was not a highway that linked eastern Acre and the capital city of Rio Branco with western Acre. Few of the residents of PDS SÃ£o Salvador had ever ventured to the capital of their state, or much beyond the city of Cruzeiro do Sul which lay so me eighty kilometers from the seringal. Because of its relative isol ation, the political movement from below that defined the seringueiro movement of Chico Mendes in easte rn Acre, had not had as far reaching an impact in western Acre. According to documents from the fede ral land agency responsible for the development of settlement areas in Brazil, Instituto Nacional de ConizaÃ§Ã£o e Reforma AgrÃ¡ria (INCRA 1979), in 1933, Pedro de Morais (also spelled Pedro MorÃ£es) the owner of the seringal named SÃ£o Salvador that in cluded the colocaÃ§Ãµes of SÃ£o Francisco, Santa Luzia, Foz de Breguesso, Itacolomy do No rte and Aracoyaba, hired Luiz Gomes de Souza, or Seu Duda, to clear rubber tapping trails in the forest. There is no mention of how Mr. Morias came to acquire this tract of 37,532 hectares of land. It is possible that this land had been part of the homeland of the indigenous group, the Nukini, that currently occupies a reserve that borders PD S SÃ£o Salvador. In 2005, the Nukini and the
42 residents of PDS SÃ£o Salvador lived side-byside in relative peace, although there was little mixture between th e two communities. One of the frequent complaints that the N ukini had with outsiders regarded hunters entering their territory and encroaching on th eir lands. This was also a complaint of community members in PDS SÃ£o Salvador who stated that sometimes hunters came from the city centers of MÃ¢ncio Lima and Cru zeiro do Sul to illega lly hunt within their communities in order to sell wild game in th ese citiesa further violation of Brazilian law. There has been a cautious, recent movement of proposed meetings between the cacique and other leaders of the indigenous N ukini with the Conselho Gestor of PDS SÃ£o Salvador, the elected council with members of each of the ten communities in the settlement, to entertain discussions on how to address problems, opportunities, and management concerns that are common to both groups of peoples. The rubber system installed by Pedro de Mo rais was apparently not as harsh as it was elsewhere, as most of the older generati on that currently reside in PDS SÃ£o Salvador, who spent some of their adult years tapping r ubber, spoke fondly of him. The individuals that were recruited to tap rubber in the se ringal were mostly born in western Acre, although many of their parents were part of the wave of immigrants to the state that came from the northeastern Brazilian state of CearÃ¡. A few of the grandparents of the 2005 residents of PDS SÃ£o Salvador came from the state of Amazonas that lies just to the North of Acre. Another key difference with the earlier established rubber system was that the establishment of se ringueiro families within Seri ngal SÃ£o Salvador was not prohibited, as was customarily in other seringa is. Each seringueiro family worked two
43 rubber trails and was required to pay an a nnual fee of 50 kilograms of rubber to the Morais family.10 As Mr. Morais became too old to run the daily operations of rubber production in the seringal, he turned the ope ration over to his sons in the 1980s. Things deteriorated rapidly through careless management, and resi dents say that working conditions for the seringueiros became stricter, and eventual ly as the operation proved less and less profitable, the Morais family abandoned the seringal in the mid 1980s. The price of rubber in the nearby Alto JuruÃ¡ in we stern Acre fell from US$1.80/kg in 1982 to US$0.40/kg in 1991 (Pantoja 2004). By the la te 1980s, the residents of Seringal SÃ£o Salvador were left to their own accord, free to pursue whichever livelihood strategies best suited their households and families. An autonomous rubber system briefly develope d in the seringal for a few years that was quickly eclipsed when the subsidies fo r rubber were increas ingly relaxed by the Brazil government in the 1980s, and effectiv ely abandoned in the early 1990s. Rubber lost its central importance in the socio-cult ural and economic liveli hoods of the residents of the area, and by 2005 few, if any, resident s of PDS SÃ£o Salvador tapped rubber as a means of procuring a livelihood for their fam ilies. (None of the households interviewed in this study reported that they tapped rubbe r, although a few adult men wished to return to the rubber economy of the days of old.) Some households suffered in the transition from livelihoods based on the production of rubber to other forms of subsistence economy. One ex-rubber tapp er mentioned that the patrÃ£o (Pedro de Morais) was 10 Carneiro da Cunha and de Almeida (2000) estimate that in the JuruÃ¡ Valley in western Acre in the 1970s, each seringueiro household pr oduced, on average, 600 kilograms of rubber annually. Rubber tappers were charged 30 kilograms per rubber trail, per year, (60 kilograms rubber/year), as payment to their patrÃ£o. This represented 10% of a householdÂ’s yearly rubber production.
44 reasonable and not exceedingly harsh to him or his family; he knew what was expected of him, and he did not mind walking through the fo rest collecting latex to make rubber. The shift in household livelihoods from rubber to small-scale agricultural production left some of the older residents of PDS SÃ£o Salvador at a loss as to how to effectively engage with a new economic system that had broken from the patron-client relationship of the aviamento rubber system. The overwhelming majority of the residents of the settlement, though, enjoyed the autonom y they had acquired since the end of the rubber economy. Some residents stated that in the rubber syst em of old, they felt as if they were always working for someone else (their patrÃ£o), but now , they were working for themselves and their families, with household production based on the skills and needs of household members, not on the nece ssity of meeting production quotas set by outsiders. Some authors state that the aviamento sy stem persists today in the seringal, transfigured throughout the region (Raffles 2002). Though the rural producer was not as intametly tied to buyers of their product as oc curred in the rubber system in the past in Amazonia, buyers continue to purchase goods from the rural caboclo at below market value, while selling them manufactured items at above market price. An example of this is reflected in a conversation I had with an elder resident of PDS SÃ£o Salvador. He remarked that nowadays the economy of farinha in the seringal in many respects resembled the rubber economy of the past. Whereas regatÃµes (river traders) collected rubber from the seringueiros in the past, now they came from the cities of MÃ¢ncio Lima and Cruzeiro do Sul in their boats seeking farinha (manioc flour) produced by the rural caboclo. These river traders knew that the re sidents of PDS SÃ£o Salv ador frequently did
45 not have money to pay for gasoline for the journey into town to sell their agricultural product, yet had farinha ready for market, a nd thus, the traders we re able to purchase farinha at below market value while simulta neously selling manufactured goods sold in urban areas at inflated prices. Just as in th e past when the rural pr oducer was not able to come to the market, the market came to him/her. Post-rubber Political-econo my in PDS SÃ£o Salvador In the post-rubber rural economy in Seringa l SÃ£o Salvador in the early 1990s, the rural peasantry in the seringal went thr ough a quick succession of household livelihood strategies. First, households relied principally on the harvest of timber, then moved to the sale of wild game and animal skins11, and finally to reliance on small-scale, family based agriculture centered on the production of fa rinha for market (CÃ¢mara 2002). Farinha is the staple of Amazonian diets, and the main source of cash income for the majority of the residents of PDS SÃ£o Salvador. Farinha serv ed the dual purpose of a crop for eating, and a crop for sale. Whereas farinha was the principal cash crop in the summer months, it became the primary household staple in the lean winter months in PDS SÃ£o Salvador. In 1993, the Sociedade de Produtores AgrÃcolas do Rio Moa (Society of Agricultural Producers of the Moa River) was created by the hous ehold producers that lived along the Moa River as a means to insure that their interests were protected and accounted for, but principally as a means to po tentially acquire future credit. A few years later, the organization proposed to INCRA th at a settlement be created along the Moa River. With the establishment of a settlement comes (agricultural) credit for the settled 11 One resident stated that during the years of the trade in animal skins, some individuals would hunt wild game, remove their skins at the site of the hunt, then simply leave the meat to rot in the forest and return to the city centers with animal pelts to sell.
46 families. By 2005, though, membership in this rural producerÂ’s organization was largely symbolic, as the group did not engage in poli tical advocacy for its members. In 1994, the indigenous reserves of the Nukini and the Poianawa were established along the Moa River in close proximity to Seringal SÃ£o Salvador. The 1990s saw a critical shift in the li velihood strategies for the residents of Seringal SÃ£o Salvador, from rubber tapping to small-scale agricultur al production with some livestock production. Increasingly the st anding forest lost its central importance as the critical space for the survival of the household. Forested lands were cleared as agriculture became the economic center for most households in the seringal. Also, land was increasingly cleared for pasture to raise cat tle, as the rubber system collapsed in the area. Cattle often serves as a reserve of value for rural peas ant households in the event of a quick need for cash for the household. Potent ially the increase in this activity would lead to further deforestation, with the subseque nt loss in habitat for animals that dwell in the forest, thus, decreasing th e take of wild game from hunting and the consumption of meat by the residents of the seringal. Women within the household a nd community may have benef ited from this shift to agriculture from rubber tappi ng, as their resource procur ing activities gained in importance for the rural household. Women ar e key actors in the production of farinha, and many are critically involved in the s upply of food for the household through their participation in fishing. Fish may be a subs titute for wild game, and if women of the household participate in this activity, men may forgo hunting. As men in the household turned their attention away fr om the forest and rubber tappin g, the hunting of wild game decreased, and the importance of women in the provision of protein and food for the
47 household was heightened through their pract ice of fishing. Therefore, womenÂ’s productive activities had become more prominen t for the rural households in the seringal, whereas traditionally, rubber tapping domina ted by men defined the rural household in the area.12 In 1989, the Parque Nacional Serra do Divi sor (PNSD) was created on the western border of Seringal SÃ£o Salvador , abutting the frontier with Peru. The socio-diagnostic and ecological studies that resu lted in the mid to late 1990s concluded that there were many species of plants and wildlife that were po tentially at risk due to the inhabitants that currently resided within the boundaries of the newly created park (Zoneamento EcolÃ³gico-EconÃ´mico, ZEE 2000a). Brazilian law did not permit human residence in a national park. The federal government, along w ith the state government of Acre and nongovernmental organizations (NGOs), began the process of looking for other lands where the inhabitants of PNSD could be moved. Discussion, often heate d, still continued in 2005 with residents livin g within the bounds of the Nationa l Park, with residents vowing not to move and the government insisting that th ey must be resettled elsewhere. Seringal SÃ£o Salvador was acquired by th e Brazilian colonization agency, INCRA, in 1999 to be used as a resettlement project for peasant families living within the national park (PESACRE 1998). The abandoned sering al contained nearly 28,000 hectares of forest bordering the national par k, and consequently this area was an important link into the overall scheme of the buffer zone mana gement of the national park. INCRAÂ’s original plan was to resettle families living w ithin the border of the PNSD to Seringal SÃ£o 12 Conversely, a drawback for women in the transformation of household livelihoods from rubber to the production of farinha was that they may have had to spend more hours working as a result of their greater participation in fishing and farming in the household economy.
48 Salvador, into family owned small fa rms (PESACRE 1998). The year-long study conducted by a local Brazilian non-govern mental organization, PESACRE, with the collaboration of the University of Florida and other organizations, concluded that the resettlement project was not advisable, and could potentially have negative environmental and social impacts on the land an d the families that already lived there. Thus, at last estimate (ZEE 2000b: 252), PN SD was home to 5785 residents with a population density of 0.68 inhabitants/kmÂ², s lightly higher than the 0.67 inhabitants/kmÂ² that has been suggested to limit hunting pre ssure on wildlife (Bodme r et al. 1988; Peres 1990; Vickers 1991; Calouro 1995). The latest plan of INCRA was to resettle the residents of th e PNSD within the Projeto de Assentamento HawaÃ, in the munici pality of Rodrigues Alves in western Acre. Park residents had fiercely resisted this development, and negotiations had remained tense between residents of the park and government representatives from INCRA and IBAMA. An added development was in 2004 granting part of the southern portion of PNSD as a reserve of the NÃ¡ua indigenous pe ople. The caboclos living within the bounds of the newly declared indigenous reserve we re now required to resettle. Most moved north (or were planning to) to land occ upied by family members living in PNSD. The socio-diagnostic survey of the comm unities of Seringal SÃ£o Salvador in 1999 found seventy-six families, with a total of f our hundred twenty three inhabitants living within nine loosely grouped communities (CÃ¢mara 2002). Moving more families into this area from PNSD was not deemed ecologi cally or socially appropriate, yet now the State had to reckon with the residents that did live in th is area, as the management practices of these families, especially those of hunting and fishing, cr itically affected the
49 stocks of wild game and fish within the bounds of the national park itself. Wildlife roamed freely across the border of the park w ith the seringal, and residents of PDS SÃ£o Salvador living on the border of the two areas occasionally crossed into PNSD to hunt. PDS SÃ£o Salvador was still a criti cal site in the overall plan of buffer zone management for PNSD. Anxious residents of Seringal SÃ£o Salva dor, fearing that the government would move more families into the seringal, an area of land that many had called home for at least two or more generations , were brought into dialogue wi th INCRA with the help of PESACRE, to discuss resource management options within the 28,000 hectares of tropical forests. In the process of negotia tion that resulted betw een residents of the seringal and INCRA, Seringal SÃ£o Salvador became Projeto de Desenvolvimento SustentÃ¡vel (PDS, Sustainable Developmen t Project) SÃ£o Salvador on July 14, 2000. Thus, a new form of settlement was created in Brazil. Not only was it the first sustainable development settlement in Brazi l, but a settlement was created not with families that were transplanted to the area who had no history of working the land that they were settled upon, let alone a knowledge of the resource base, but with the residents that had already created their livelihoods and raised their families for at least the past several decades. Colonization projects in Br azil have historically le d to dire environmental outcomes for the resource base, as poor fam ilies typically were settled on land that was foreign to them (Fearnside 1984; Hall 1989: Chapter 1; Schmink and Wood 1992; PESACRE 1998). The immediate necessity of poor families raising crops to support their households, often without technical support, led to dest ructive agricultural practices
50 and the further conversion of forestlands. Additionally, agricultural credit programs often forced poor migrant peasants into pl anting annual food crops (rice, soy beans) whose long term production was ill-suited for the soils and climate of Amazonia (Smith 1982; Smith et al. 1995).13 The principal goal of the State in the overwhelming majority of these projects was to settle the landle ss peasantry, with secondary concern for the resulting effects on the ecosystems and land base of these sites (Fearnside 1984; Schmink and Wood 1992). Unfortunately, there has been a large degree of land ownership turnover in Brazilian coloniza tion projects, often leading to increased deforestation (Cronkleton 1998). In contrast, PDS SÃ£o Sa lvador was envisioned and created in dialogue with INCRA, NGOs with a sustained history of working with local peoples in the area, PESACRE and SOS AmazÃ´nia, and local residents who had a history of raising families in the proposed colonization site, as part of a scheme for regional management of resources in the area (buffer zone) intimat ely tied to the conser vation of the forest ecosystem of PNSD (CÃ¢mara 2002). Integrally tied within the conservation of the resource base in this larger unit of land management is the concern for the sust ainable management of fauna. Animals do not stop at the borders of the arbitrary land classifications of humans, and hunting pressure from poor residents outside the national park can severely limit the species and genetic range of animals, limiting the futu re stock of animal life in, and around, the protected national park. As mentioned above, animals are important disseminators of the seeds of tree species throughout the Amazon, and the loss of critical stocks of wildlife 13 Although manioc is well suited to the ecological conditions of Amazonia and has been cultivated by indigenous peoples for centuries throughout the Amazon basin, Â“a biased system of fiscal incentivesÂ” (Smith 1982: 73) insured that credit was not available for colonists to plant this crop.
51 can have dire consequences for the structur e and functioning of tropical forests, with potentially negative environmental and socio-cu ltural outcomes. Furthermore, areas in and around PNSD have been identified as ar eas of high human hunting intensity in the state of Acre (ZEEa 2000: 164). Therefore, the use of wildlife by the residents of PDS SÃ£o Salvador and the wise management of fa unal resources by these inhabitants in this area was of critical importance in the cons ervation and developmen t agenda and planning regionally, and in the creation of the new settlement (CÃ¢mara 2002). Within this new settlement, the federal government retained tit le to the land and residents were given leases to use the land th at they occupied within PDS SÃ£o Salvador. User rights were given to the households of PDS SÃ£o Salvador for twenty years, renewable if the rules and re gulations set forth by the State were observed by local residents. One key area of debate ea rly in the negotiation process between the government and the residents of the settlement centered on how the land was to be operated and managed by the residents of the se ttlement. In the initial stages of the process the government was proposing that land be co-managed collectively by communities, without individua l title to land, whereas the inhabitants of the newly created settlement were opting for individua l household lots. A dditionally, there was much heated debate centered on how much land each household was to be given to use to sustain their livelihoods. Eventually each hous ehold was given title to usage of a one hundred hectare lot of land, although by law onl y twenty percent could be cleared for agricultural use and/or livestoc k production. Provisions were made for future generations (future lot grantees) within th e settlement, as long as deforestation did not reach more
52 than twenty percent of the to tal area of the sett lement. It was estimated in 1999 that current deforestation within the settlement was less than five percent ( CÃ¢mara 2002). The communities of TimbaubÃ¡ to the north west of the settlement, as well as the northern part of the community SÃ£o Pedro, were incorporated by INCRA within the settlement after the initial talks began be tween residents of th e seringal and the government. In 2003, INCRA added an additi onal 25,000 hectares to PDS SÃ£o Salvador in the far northwestern portion of the sett lement bordering the community of TimbaubÃ¡ with the Nukini Indigenous Reserve. The Nukini had earlier petiti oned state and federal officials to incorporate this land within their te rritory, to be reserved as an area solely for hunting purposes. One hundred seventeen families in the newly named settlement registered with INCRA in 2000 to receive the credit offered by the federal government to settlement residents.14 Although the declaration of the new settlement on July 14, 2000 did not permit entry beyond this date, familie s previously living outside the bounds of PDS SÃ£o Salvador continued to move into the area. The map commissioned by INCRA dated January 2003, listed one hundred fifty agri cultural lots, ranging in size from a high of 35.8 hectares to a low of 17.3 hectares (s ee Figure 2.1). The overwhelming majority of lots were within twenty one to twenty hect ares in area. Recently residents had become much more vigilant in enforcing the rule that no new households were permitted to be established within th e bounds of the settlement. Comp laints were registered with IBAMA, the agency within the Brazilian gove rnment charged with the management of the environment. IBAMA shared a post with the military within the settlement, that 14 When word spread that INCRA was offering easy credit to households in the newly created settlement, some poor individuals from the closest town of MÃ¢ncio Lima scrambled to set up shelter in the seringal so as to acquire money and a piece of land to live and work.
53 Figure 2-1: Map of PDS SÃ£o Salvador Adapted from INCRA 2000.
54 Figure 2.1 Continued
55 Figure 2.1 Continued Commas under area (ha) heading signify decimal points.
56 monitored boat traffic along the Moa River, checking for suspicious cargo including drugs smuggled over the nearby border from Peru15, while also insuring the environmental regulations of the federal government were met by residents and other visitors to the settlement. Prior to declaring Seringal SÃ£o Salvador a sustainable development settlement, as a necessity to receive the credit offered by INCRA, households were required to group themselves within named communities. Hi storically, households were grouped within small family clusters, as was common for ma ny rural peasant households in the Brazilian Amazon (Harris 2000), from as few as two households to ten households, but in the resulting process, ten communities were established, most taking the old names of the colocaÃ§Ãµes of the rubber era. The process of shifting from a familial based socialpolitical structure to a community-based one was a difficult transition for most households in the settlement, with frequent community disputes (and resolution) defined on the basis of familial associ ation. Lingering disputes w ithin individual communities over hunting territory and practi ce often pitted one family gr oup against another. The bounty of the hunt, wild game, frequently served as the catalyst for th e intensification of animosity between competing family groups when meat was not exchanged across these boundaries. Each registered family with INCRA was pr omised a line of credit in the amount of $R3,900: $R2,500 for construction of a house (and well if need be), $R400 for food, and 15 Prior to the establishment of the military post within the settlement, residents of the seringal reported that it was not uncommon to see Peruvians hiking through the forests (transporting cocaine) attempting to evade the Brazilian authorities.
57 $R1000 for tools.16 Unwittingly the credit given to i nhabitants of the seringal would quicken the pace of deforestation, as severa l households purchased chain saws with the monies they received from INCRA. Interest payments were set at six percent, with twenty years allowed to pay in full. For th e overwhelming majority of the beneficiaries, this was the first time that they had been ex tended credit during th eir lifetimes. Few of the residents understand this requirement to repay the credit, as each household upon receipt rapidly spent most of this money, a nd it was doubtful that repayment will occur. PESACRE took the lead in developing mana gement goals with the residents of PDS SÃ£o Salvador in the years following the de claration of the settlement. The area of most contention with the ten communities pe rtained to hunting and the management of wildlife in the seringal. Fragoso et al. ( 2002) had found that the cu rrent level of hunting by the communities in PDS SÃ£o Salvador was not sustainable, and that some animal species, including threatened ones, might have become extinct in the area as a result of hunting pressure on wildlife. It has been noted by other researchers in Latin America that Â“the presence of healthy populations of wild game generally attests to healthy, su stainable landÂ” (Shaw 1991: 30). In 2005, residents in PDS SÃ£o Salvador remarked how it was rare to see monkeys and other larger animals close to th e communities, as they had just a decade before, when they often wandered in close pr oximity to the settlement areas, and hunters lamented the fact that they must walk further and further into the fore st to encounter wild game to hunt. This was due to the fact that critical habitat for wild animal species was 16 The exchange rate of the Brazilian Real with the US dollar was around US$1 to $R2.85 at the end of October 2003 when the majority of the households of PDS SÃ£o Salvador had finally received the credit promised them by INCRA.
58 increasingly lost in the conversion of fore st to agricultural fi elds close to human settlement areas, and because many large ga me species had been over-hunted in the forestlands that did remain standing close by the communities of the area. In a series of meetings lasting well over a year, PESACRE in conjunction with the residents of the ten communities of the settle ment created a series of regulations and management goals in regards to the use of natural resources within the settlement (Comunidade do Projeto de Desenvolviment o SustentÃ¡vel SÃ£o Salvador 2003). Community regulations set guidelines for resi dents in regards to hunting, fishing, timber exploration, animal husbandry, a nd the clearing of forestland. It was decided that if a resident of any of the ten communities w ithin PDS SÃ£o Salvador violated these regulations three or more times, the Consel ho Gestor would ask INCRA to remove this person from the settlement. As of 2005 this had not occurred, a lthough there had been denunciations to INCRA of the practices of some members of the settlement by other residents. Many residents though, were willing to overlook many of the violations of community regulations as long as they did not directly impact their livelihoods. In late 2003, an INCRA representative visiting the settle ment threatened to remove one resident from the settlement for his failure to plant his assigned lot, and due to the fact that he had recently taken work in the town of MÃ¢ncio Lima and had not lived on his assigned lot for several consecutive months. The area of the management of resources of the settlement that drew the most attention was that of hunting, with nineteen regulations documented to regulate the practice of hunting within the settlement (Comunidade do Projeto de Desenvolvimento SustentÃ¡vel SÃ£o Salvador 2003). (See Table 2-1.) Several of the regulations were
59 successful in modifying hunting practices with in the settlement, in cluding reducing the number of individuals that hunt ed with dogs, but other such regulations had not been as closely followed. A common tr igger point of animosity between households within communities in PDS SÃ£o Salvador remained the practice of hunting, and the use of wildlife within the settlement. Table 2-1: Community Hunting Re gulations in PDS SÃ£o Salvador Hunting porquinho [also commonly called cutia] ( Tayassu tajacu ) with dogs is prohibited. Hunting with dogs is prohibited. Dogs are permitted to accompany residents to keep wild animals from attacking residentsÂ’ domesticated animals. Hunting for food for a trip is permitted. Only 1/2kg per person per day up to a maximum of 20kg. Hunting the offspring of deer is prohibited. If you see a track of offspring of deer you are not permitted to shoot at the offspring. If you see tracks of the offspring of d eer or armadillo, do not kill the mother. Hunting to feed oneself is permitted. Hunting animals threatened with extinction is prohibited. Minors can only hunt with permi ssion from their parents. You must advise your neighbors when setting a trap [ armadilha ]. Traps are permitted only from 5pm until 7am the next morning. Traps must be accompanied with a warning si gn advising of their use in the vicinity. A fence must be put up around an armadilha. Traps are to be set up in th e buffer of oneÂ’s community. Hunters that do not live in the settlement are not permitted to hunt. Hunting to raise the offspring of wild animals is not permitted. Hunting to sell the meat of wild game is not permitted. Hunting turtles is permitted. * Translated by the author (from Portuguese) from Comunidade do Projeto de Desenvolvimento SustentÃ¡vel SÃ£o Salvador (2003). Another serious problem within the settle ment in regards to the hunting of wild game is the threat from outsiders who are not residents of PDS SÃ£o Salvador. Under Brazilian law, hunting for subsistence is permitted by residents of a settlement, but it is prohibited by individuals who do not reside w ithin the bounds of the settlement. Because there is little presence of the government bodi es that regulate hun ting practices in the
60 area, it is not difficult for outsiders to come to these areas and hunt, carrying their cargo of wild game with them to the city centers wh ere they live. Furthermore, residents of the seringal are afraid to confront illegal hunting in their lands because of the fear of bodily harm at the expense of these illegal hunters. Several residents of PDS SÃ£o Salvador informed me that they had been threatened with Â“a bullet in the b ack of their headÂ” if they denounced these illegal hunt ers with IBAMA. Finally, one additional problem that must be addressed in regards to hunting in PDS SÃ£o Salvador, was the selling of wild game in the nearby city centers. One resident of the settlement told me th at during the rubber era along th e Moa River, everyone hunted to sell the meat of the animals they hunt ed. Although illegal, at the end of 2003 wild game was selling in the city of MÃ¢ncio Li ma for around $R5 per kilogram. Many species of wild animals were considered deli cacies in the region including paca ( Agouti paca ), white-lipped peccary ( Tayassu pecari ), and red brocket deer ( Mazama americana ), and there were willing buyers of the meat from these animals in the cities. The selling of wild game occured mostly by hunters from outside th e settlement, but unfortunately also, (less frequently) by residents of PDS SÃ£o Salvador. Peer pressure was unlikely to solve these problems with the practice of hunting and the management of wildlife within the set tlement; instead IBAMA and INCRA must be vigilant with their patrolling of the resource pr actices that occur with in the settlement. A major limitation for these agencies was their lack of funding, and lack of staff, to adequately manage the thousands of hectares that fell under their jurisdiction. PESACRE had worked tirelessly over the past several years to educate the residents of PDS SÃ£o Salvador to the dangers of over-hunting the anim al resources of the se ttlement, while also
61 attempting to strengthen the Conselho Ge stor, the advisory council comprised of residents of the ten communities of the set tlement, that had some governing capacity. PESACRE had also attempted to act as an intermediary between the Conselho Gestor and INCRA and IBAMA. As the residents of PD S SÃ£o Salvador had little experience with Â“self governanceÂ”, undoubtedly the process woul d be slow with many setbacks, yet it was crucial for governmental and non-governmental agencies to continue to work closely with local residents whose reso urce practices had a major imp act on the sustainability of the resource base for themselves and futu re generations. The State of Acre The state of Acre is located in the fa r western portion of the Brazilian Amazon bordering the countries of Peru and Bolivia to the west and south, th e Brazilian state of Amazonas to the north, and the state of RondÃ´nia to the east. Acre is about the size of Georgia (153,149 square kilometers) and contains less than two percen t of the total area of Brazil. Nearly ninety percent of the stateÂ’s total area is classified as tropical forests. The total population of the state was around 550,000 (Souza 2002).17 The area that comprises the state of Acre was ceded by Boliv ia to Brazil with the Treaty of PetrÃ³polis in 1903. After over one-half of a century of classification as a territory in Brazil, Acre was elevated to the level of statehood on June 15, 1962. The state of AcreÂ’s relative isolation fr om outside markets and the rest of Brazil made the large-scale colonization projects that characterized Amazonian development elsewhere challenging to implement (Kainer et al. 2003). It was not until very recently, in 1992 when the final stretch of highwa y BR364 was paved, that Acre became 17 In 2002, AcreÂ’s population comprised less than Â½ of one percent of the total population of Brazil.
62 accessible year-round by land to other states in Brazil besides Amazonas and RondÃ´nia. The result of this was that deforestation was limited to less than 10% of the total area of the state, and Acre attracted researchers a nd planners intent on pr acticing novel ideas of sustainable development. The current govern ment in the state of Acre, popularly known as the governo da floresta (forest government), was an impor tant ally in the search for sustainable resource policies and had incorporated a development philosophy that sought ecosystem conservation, social equity, and cultural diversity (Kainer et al. 2003). The economy of the state of Acre, more so than any other region in Brazil, was based on the extraction of forest products. Se veral products were ex tracted and marketed from the stateÂ’s forests including fruits ( aÃ§ai , cupuaÃ§u ), Brazil nuts, honey, heart of palm, and to a lesser extent wood, but the ta pping of rubber defined the socio-economic relations of the state until recently. Th e Amazon rubber boom beginning in the late nineteenth century and lasting until just prio r to World War I, put Acre on the map. In 1995, the rubber economy produced around US$43.5 million for Brazil, of which 59 percent was generated in Ac re (Allegretti 1995: 167). Three distinct socio-cultural groups lived within the forest environment in Acre: agricultural colonists, seringue iros [caboclos], and indige nous peoples (Kainer et al. 2003). The colonists were relati vely new migrants to Acre co ming to the state within the past three decades, and did not have the same ecological sensibility as did the seringueiros and indigenous communities. Th e origin of these new migrants differed from the first wave of immigrants to the stat e at the turn of the twentieth century that came largely from the drought-plagued sert Ã£o region in the Northeast of Brazil18, in that 18 The current inhabitants of PDS SÃ£o Salvador were mostly descendents from migrants who came to Acre from the northeastern Brazilian state of CearÃ¡.
63 they were largely from the southern Brazili an states, enticed to the Amazon region not by the rubber trade, but by the opportunities to work in the cattle industry that sprang up in the region in the 1970s, as well as the promise of open land that they could work to raise their families. Whereas in 1971, three-fourth s of the land in Acre was classified as terra devoluta (unoccupied public land), by 1975, 80% of the land in Acre was owned by investors from southern Br azil (Branford 1985). The 1970s in the state of Acre was described as Â“paradise for large and medium size producers of cattleÂ” (Souza 2002: 99). The climate and vegetation in southern Brazi l bore little, if any, resemblance to the ecological conditions of the humid tropical forest environment found in Acre, and combined with the poverty of these new im migrants necessitated bringing land quickly into agricultural production to feed their families. Their survival strategy was predicated on slash-and-burn agriculture that necessitated the removal of forest cover in order to produce food crops in combination with small-scale livestock production. On the other hand, the seringueiros [ca boclos] and indigenous groups had a longterm, sustained, intimate relationship with th e forest that until recently was based on the extraction of forest products for their live lihoods. Weinstein (1985: 105) notes that the seringueiros had Â“Â…developed a common world view and mode of resistance that had greater significance than the obvious differences in ethnic or cultural backgrounds.Â” Their strategy for survival was based on the protection and preservation of the forest. Thus, the ecological experien ce of the seringueiros based on the knowledge of how to sustainably manage the tropical forests in Ac re, gained from decades of rubber tapping activities in the forest, led them to the political arena to stake their competing claims for the resources of the state. These claims were based on the survival and needs of the
64 family (PopkinÂ’s  rational actor model), th at were intertwined with the survival and needs of the forest and the community at la rge (ScottÂ’s  mo ral economy model). Neither peasant family, nor the community, ex isted in isolation, and the overlap between individual and community goals led Chico Me ndes and the seringueiros of eastern Acre to seek political action in the Â“fight for the forestÂ” (H eckt and Cockburn 1989). Even after the assassination of Chico Mendes in 1988, the movement of the seringueiros of Acre persisted, as among other rural peasan try in Latin America, Â“because of their capacity to translate ecological interdepe ndence into political powerÂ” (Anderson 1994: 15). However, resource use by the caboclo peas antry in PDS SÃ£o Salvador (and other areas throughout Acre) shifted beginning in the early 1990s from livelihoods based on the extraction of forest products (r ubber) to other land uses in cluding small-scale agriculture and the raising of livestock. This transf ormation necessitated de forestation in the conversion of forestland for crop and lives tock production. Some evidence suggested that the rural peasantry in PDS SÃ£o Salva dor and elsewhere in Acre was increasingly turning towards the domestication of cat tle as a livelihood strategy (Gomes 2001; Salisbury 2002). As tropical forests we re cut, habitat for wildlife was increasingly lost. Western Acre, where PDS SÃ£o Salvador is located, has always existed in relative isolation from the more populous eastern port ion of the state. Because as of 2005 a paved, year-round passable road still did not co nnect western with eastern Acre, western Acre had remained much less connected to ot her states in Brazil (except the state of Amazonas) than eastern Acre. Western Acre hi storically had closer ties to the state of Amazonas to the north, connected via the JuruÃ¡ River, than to eastern Acre. The JuruÃ¡
65 River was the route for the export of rubbe r during the rubber econom y, serving today as the method of transportation for farinha produ ced by the peasantry of western Acre. The lack of communication with the outside world limited the politically unifying force of the seringueiro movement in the western portion of the state. Furthermore, whereas eastern Acre had been heavily influenced by the libe ral politics of the Work ers Party (PT) over the past several decades, western Acre had not been as embracing of their liberal agenda, and tended towards electing conservative politicians to state office. The result of this isolation for peasant households in western Acre was that they tended to be poorer than member s of their class in the East, and they tended to have greater autonomy from the market than did the peasantry in eastern Acre. It was reported that the standard of living of the peasantry in PDS SÃ£o Salvador was lower than their rural counterparts in similar areas in the ot her major watersheds th roughout the state of Acre (Shaeff 2002). The peasantry [of PD S SÃ£o Salvador] were not (yet) solely commodity producers pertaining to a transito ry social stratum as described by linear models of the historic march of capitalism (Deere and de Janvry 1979). In agreement with BernsteinÂ’s (1982) proclamations for th e peasantry in genera l, because the rural residents of PDS SÃ£o Salvador had retained control over the orga nization of production Â– principally land and labor they had not become part of the proletarian class. As some sectors of the Brazilian peasantry were only pa rtially integrated into the market system they were able to continue to produce goods and services for use-value by household members, as opposed to production solely for exchange-value (goods bought and sold in the marketplace), in the subsistence house hold economy, relying on numerous forms of non-market transactions to reprodu ce social/economic relations.
66 Hunting by the peasantry in PDS SÃ£o Salvador was an integral part of a householdÂ’s livelihood strategy that was beyond th e scope of the market. First, animals in the settlement were not a private resource, but a comm on property resource.19 The bounty of wild animals in the forest was open to everyone in the community to exploit, free from a monetary cost. Second, because the product of the hunt (wild game) was freely exchanged in non-market transactio ns between rural peasant households, the peasantry was not totally encaptured by the sy stem of capitalist relations. Game meat was not bought and sold in the marketplace, but given, received, a nd reciprocated by peasant households. These reciprocal transac tions were not based on an equivalence of the amount of the product (meat) exchanged, but on an equivalence of the exchange of relations (kin) between households. Therefor e, it would be inaccurate to label the peasantry in its entirety as proletariats subj ect to the whims of the capitalist system that minimizes their contribution to the market (as important producers of cheap goods and labor), while pushing (firing)/pul ling (hiring) their labor power in and out of the system. Hunting, as currently practi ced by residents of PDS SÃ£o Salvador, was important in the reproductive regime of individual households and communities in the settlement. Subsistence production intimately tied to the reproductive needs of the household was further, a form of resistance to the encap turement of the peasantry by the capitalist system. Harriet Friedman (1980: 163) write s: Â“Â…if household reproduction is based on reciprocal tiesÂ…for renewal of means of production and s ubsistence, then reproduction resists commoditisation [sic].Â” Hunting performance in PDS SÃ£o Salvador was joined 19 Wild game was a common property resource within PDS SÃ£o Salvador, but its use was contentious which is reflected in the larger number of community rules and restrictions regarding its use. See Table 2.1 above.
67 with reciprocal relations be tween households, allowing for hous eholds to resist the end point of commoditization, simple commodity production, and thus, full integration into the market system. Hunting and the sharing of wild game acr oss households were key forms of peasant non-market exchange. This reciprocal exch ange likely would continue to occur in peasant communities in PDS SÃ£o Salvador (and elsewhere throughout Amazonia) as long as there remained an adequate stock of wild animals in th e tropical forests where these groups of peoples hunt and live. But, if hun ting practice were to continue on its present course, and the population continue to increase in the seringal, the future of the forest might reach the Â“empty forestÂ” that Redford ( 1992) spoke of, devoid of critical stocks of wild animals, and ultimately threatening the ecological sustainability of the tropical forest, as well as threatening th e livelihoods of the inhabitants of these forests. The Municipality of MÃ¢ncio Lima PDS SÃ£o Salvador is located in the municipa lity of MÃ¢ncio Lima in the far western portion of the state of Acre in the Brazilia n Amazon, the westernmost municipality within Brazil. A large percentage of the total la ndmass within the muni cipality is under the jurisdiction of the federal government. PNSD (267,000 hectares), the Nukini (27,000 hectares) and Poianawa (21,000 hectares) I ndigenous Reserves, and PDS SÃ£o Salvador (53,000 hectares) account for [well over] two-thir ds of landmass of the municipality of MÃ¢ncio Lima (470,000 hectares) (Toni 2003: 168). There are also two smaller settlement projects administered by INCRA in the muni cipality, the Projeto de Assentamento (PA) Rio Azul with 6,800 hectares, and PA SÃ£o Domingos with 1665 hectares (Cunha dos Santos 2003). In 2004, the NÃ¡ua indigenous gr oup was successful in their claim to the
68 legitimacy of land rights in the southern portion of PNSD and was granted 45,000kmÂ² of land (Campbell 2005). The municipality of MÃ¢ncio Lima cont ains 4,692kmÂ², a little more than three percent of the landmass of th e state of Acre (Government of Acre 2004). The population of the municipality was estimated to be 11,069 in 2002 (Souza 2002) with 46% living in rural areas and 54% in city centers (Gove rnment of Acre 2004). About 60% of the economy of the municipality was based on the public sectorthe federal, state, and municipal governments with the remaining 40% in family agriculture, small-scale animal production, and small commercial esta blishments (Cunha dos Santos 2003). The major commercial establishments in MÃ¢ncio Lima (food markets, distributors of gas) were controlled by the political leaders of the region. This monopolis tic control over the consumer sector combined with the isolation of the municipality of MÃ¢ncio Lima resulted in higher prices for consumers living in the municipality compared to other residents throughout the state of Acre. Subsistence agriculture ce ntered on the production of farinha dominated the productive sector for the majority of small farmers in the area. The majority of the production of farinha, sold in 50km sacs, was br ought to the cities of MÃ¢ncio Lima and Cruzeiro do Sul for eventual export to the cap ital of the state of Acre, Rio Branco, or to the city of Manaus in the state of Amazonas to the north of Acre. A large percentage of the consumer goods found in MÃ¢ncio Lima we re imported from other municipalities via the city of Cruzeiro do Sul. The period n ear the end of the dry season in October and early November was a time of a dearth of many consumer goods in the region, as the highway BR-364 that seasonally connected the cities of MÃ¢nc io Lima, Cruzeiro do Sul,
69 and Rodrigues Alves with eastern Acre was of tentimes too wet for large trucks to pass, and the water level of the JuruÃ¡ River (a crit ical mode of transportation for goods, people, and information from Western Acre to the out side world) was too low for large boats to transport products. The Research Site PDS SÃ£o Salvador PDS SÃ£o Salvador is located between the coordinates of latitude South 07Â°22'14" and 07Â°18'14", and 73Â°08'41" and 73Â°22'14" l ongitude West (INCRA 1998). When PDS SÃ£o Salvador was declared a settlement in 2000, it contained 282kmÂ² slightly more than 28,000 hectares. Around 18% of this area (52kmÂ²) was considered varzea (seasonally flooded land) and the rest terra firme (land that is above the highe st rise of the rivers and streams) (CÃ¢mara 2002). The varzea cont ains the most productive agricultural land deposited with topsoil, and other nutrients that erode from the terra firme during the winter rains. The alluvial deposits of the varzea were planted with corn and beans during the dry summer months prior to their submer ging by the rising waters brought on by the rains of winter. The majority of the homes of residents of the settlement were perched on the restingas , the ridges that overl ooked the varzea. In 2003, an additional 25,000 hectares were added to the settlement. As of 2005, this area was uninhabited, designated as a rese rve for hunting by the residents of PDS SÃ£o Salvador, and there were no plans to provide lo ts for additional families to settle in this area. This land on the northwestern end of the settlement was land that the Nukini had petitioned INCRA and IBAMA to incorporate into their reserve. This no-manÂ’s land with unclear boundaries had been a source of conflict with hunters from PDS SÃ£o Salvador, and hunters from outside the set tlement encroaching onto land held by the Nukini.
70 The settlement was located 40km from the muni cipal seat of the municipality of MÃ¢ncio Lima, which also went by the name MÃ¢ncio Lima, and 80km from the city of Cruzeiro do Sul, the second most populous ci ty in the state of Acre. The settlement was bordered by PNSD and the Nukini Indigenous Reserve to the west, Fazenda Monte Bello to the east, the Brazilian state of Amazonas to the north, and Seringal Valparaiso to the south. Both Fazenda Monte Bello and Seringa l Valparaiso had large herds of cattle, which was the principal income generating source for the owne rs of these two properties. The owner of Fazenda Monte Belo occasionally hired resi dents from PDS SÃ£o Salvador (men only) when the seasonal needs of the farm could not be met by the labor of family members alone. The vegetation of PDS SÃ£o Salvador was characterized as belonging to the dense tropical forest type that is a combination of hardwood tree species, mixed with assorted palms and bamboos (CÃ¢mara 2002). Commerci al timber species were widespread and not found in dense concentrations anywhere w ithin the settlement. Several palm species were used locally to produce bevera ges for the household including aÃ§ai ( Euterpe precatoria ), bacaba ( Oenocarpus bacaba ), buriti ( Mauritia flexuosa ), and patauÃ¡ ( Oenocarpus bataua ). Brazil nut ( Bertholletia excelsa ) which was found in significant quantities in eastern Acre and was an im portant non-timber forest product (NTFP) for many households in that portion of the state, was not found in PDS SÃ£o Salvador. In 2005, residents of the settlement did not market NTFPs. The climate of the area was characterize d by two seasons, each lasting about six months: the rainy season that ran from early November through April, and the dry season
71 Figure 2-2: Map of Acre Adapted from Salisbury 2002
72 that occurred from May until the end of Oct ober. In a conversation with one long-time resident of the seringal, he explained to me how the climate had changed in the area: Paolo: Before there were two seasons: when it was wet, and when it was dry. Now there are more times when it is dry than when it rains. Eric: Why is this so? Paolo: I think it is becaus e the forest is being cut. He also went on to explain that it was important to cut down only as much forestland as you planned to pl ant with crops in that year. If you cut more than you would use, he said, you wasted the land. Then after a few years when you needed to let the earth rest, you had to go further and further in to the forest to cut trees to have land to plant your crops. The result was that you ended up cutting more of the forest, and faster, because you wasted land. As more land was deforested for agriculture, habitat was increasingly lost for the animals that lived in the settlement. An imals retreated deeper into the forest and hunters were thus forced to walk further and fu rther from their houses to encounter wild game w ithin the settlement. The average annual precipitation fluctuated between 1750mm to 2300mm, with the humidity from 80% to 90% monthly thr oughout the year (INCRA 1998). Temperatures in the settlement were relatively stable, with an average high temperature of 26ÂºC and an average low temperature of 24ÂºC (INCRA 1998). Dry season friagens (cold fronts), usually two or three days of below normal cold temperatures occasionally swept across the state of Acre along the mountains of the Andes from the south, and temperatures in the seringal fell to around 10ÂºC. As might be expected, most activity within the settlement came to a standstill as residents huddled together within their houses waiting for the friagem to pass.
73 Development within the settlement was mini mal at best, as residents lived without running water and electricity. Infrastructu re was almost non-existent; there were no paved roads within, or to, th e settlement, and there were no stores where supplies could be purchased. Several of the ten communities that comprised the settlement had grade schools20 (none offered secondary education), but the schools lacked adequate supplies and qualified teachers to give instruction to the students. Many of the schools were taught by a resident of the community that was literate. As might be expected, the quality of education the children of the settlement were receiving was suspect. Having schools at all was a relatively new phenomenon within th e settlement, as during the epoch of the rubber economy within the sering al, schools did not exist. As a result, the a dult illiteracy rate was higharound 53% (CÃ¢mara 2002). Health care was also severely lacking w ithin the settlement. There was a health post within the community of Sede, but it was operated by a local resident of the community. As one resident said, Â“The hea lth post gives shots and pills. If you need anything else you have to go to MÃ¢ncio Lima .Â” Frequently the health post of the settlement lacked even thes e basic necessities. Malaria was a persistent problem for residents of the settlement. Other co mmon ailments included hepatitis, stomach problems resulting from contaminated water, snake bites, flu and the common cold, and infections from cuts and bruise s that did not heal readily in the humid tropical forest environment. Once per year the federal government transported nurses via military helicopter to a central location within th e settlement (the community of Sede) to distribute vaccinations free of ch arge to the residents of the settlement and nearby areas. 20 Those children in communities lacking schools had to wa lk up to one hour, or row in canoes, to achieve an education in the nearby communities that had a school.
74 Common shots that were administered include d vaccinations for yellow fever, measles, mumps, and rubella, tetanus, and hepatitis. The Moa River cut through th e heart of PDS SÃ£o Salva dor and was the principal mode of transport into, and out of, the settlement. Most of the households in the area were situated on the banks of the Moa or the streams that run off the Moa. Only one community, Vai Quem Quer, was not located along a waterway, but rather a thirtyminute walk (during the dry season) inland from the Moa River. Because of this lack of access to the principal waterway of the settle ment, this community had great difficulty is transporting its agricultural produc ts to market for sale in the cities of MÃ¢ncio Lima and Cruzeiro do Sul. There were ten named communities with in PDS SÃ£o Salvador. As stated previously, residents were required to group themselves within communities to obtain the credit that the government offered during the es tablishment of the settlement. This was a difficult process for many households that were accustomed to political alignments within the seringal that hist orically were based along familial ties. Tension within communities frequently revolved around differe nces between family groups living in the same community. This was clearly seen in disputes between households regarding the use of wildlife and with the practice of hunting.
75 Figure 2-3: The Moa River Cut thro ugh the Heart of PDS SÃ£o Salvador. Seasonal Calendar of Resource Use Peasant reproductive and productive stra tegies in Amazonia are critically dependent on the seasonal nature of the e nvironments that they inhabit (Chibnik 1994; Harris 1998, 2000). The yearly calendar of activ ities and responsibilit ies of the members of the household in PDS SÃ£o Salvador reflec ted the change in seasons (see Figure 2.4 below). The rainy season ( inverno ), started in November and lasted into April. Generally the wettest months were January th rough March. This was also the period of the year that residents called mau de rancho Â–bad for food During these several months, agricultural production was minimal, the catch of fish decreased due to the swift currents of wint er runoff, and the rivers had flooded the forests, giving fish plenty of area to hide and making food resources difficult to acquire. Since most residents of the settlement did not have sufficient income to purchase food,
76 Figure 2-4: Seasonal Ca lendar of Activities
77 they Â‘tightened their beltsÂ’ (Chayanov 1986) and subse quently decreased their food intake during these months of difficulty. The dry season ( verÃ£o ) began in May, continuing through October. The warmest m onths were those of July, August, and September. The school year began in late February or early Marc h and ran into early December. Although most respondents indicated th at hunting was an on-going activity throughout the year, they also stated that the peak season for hunting occurred during the rainy season, especially in the months of Fe bruary and March. Beckerman (1994) says in most cases in the Amazonian environment, hunting is driven by fishing, as people hunt more when fishing resources are harder to obtain (in the rainy season when the forests are flooded). Descola (1996) has also shown that hunting by the Achuar in Ecuador drops dramatically in response to in creased availability of fish resources. This fishing-hunting strategy could be predicted based on rainfall. Hunting was particularly important for the families of the seringal during inverno when agricultural production was minimal and food resources in the seringal were scarce. Th e lack of available s ources of food in the seringal during the wet winter months encour aged hunters in groups of normally two to five men, to make overnight extended hunting tr ips into the areas of the settlement that were devoid of human settlements, where pr esumably there was a greater concentration of wild animals. These extended group hunts we re seasonal in nature, as they came to a halt with the dry season and when the demands of agriculture for the household were at its peak. Fishing was another activity that occurred throughout the year. The peak time of the year when fish were most abundant was near the end of inverno (April) into the
78 warmer summer months. The Moa River was al so at its fullest normally in the month of April, facilitating faster transport along the waterways of the seringa l. The Moa reached its low point in the year as the dry summer m onths came to a close in October. The low water level in the settlement in late summer made it difficult to reach some of the communities that lay along the streams that f eed from the Moa. The low water levels also prohibited these communities from trans porting their product to market. The collection of NTFPs in the forest such as aÃ§ai, buriti, bacaba, and patauÃ¡ was at its peak during the wetter months of the ye ar from January through March. Households made a beverage from the fruit of these pa lms to which sugar and farinha were added. These NTFPs were consumed within the househ old and were not marketed in the city centers. Agricultural production began in earnest as the rain subsided and hot days of summer prevailed. In the months of June a nd July, men and adolescent boys were in the planting areas ( roÃ§ado and quadra )1 preparing these areas for the planting of the summer crops. This was a time of intense physical labor as brush and w eeds were cut by hand with machetes in the summer heat. Some members of the community, generally single mothers or retired individuals that had a re gular source of income coming from outside the community, hired other individuals in th e community to work for them, preparing land to plant at the daily rate of R$10.00. In this way, outside income was circulated amongst the community members, passing from th e hands of those that had, to those that had not, a sort of economic leveling mechan ism in the community (Scott 1976). This money, though, passes from the hands of the olde r married men or older single women in 1 The roÃ§ado refers to the area planted with manioc. The quadra is the area where other seasonal crops (rice, beans, corn) are planted. A quadra is further broken down into tarefas Â¼ of a quadra.
79 the community to the younger adult men and adolescent boys in the community. Although women were reported to assist in agricultural work and production in the community, they were not hired as daily wa ge laborers. Many members within each community trade days of labor ( troca dias ) with other members in the community, first working together in one of the quadras or roÃ§ados belonging to one of the two community members, and then when this work was completed, both working together in the fields of the other. Often these same individuals that traded labor days would hunt together as well. After the quadra or roÃ§ado wa s weeded, the cut brush was burned in late August or early September. Planting begins several weeks after the brush was burned, in late September into early October. Rice, corn and beans took three months to mature and were sold beginning in January. Generall y, though, these crops were consumed by the household, with corn being an important sour ce of food for domestic livestock. Manioc took nine months to mature. In each co mmunity there existed at least one small casa de farinha , where manioc was processed into farinha. Farinha was the principal income generating source for the majority of the families throughout the PDS SÃ£o Salvador. Farinha was processed and sold throughout th e year, although produc tion slowed during the wet winter months. Because of this , income decreased for households in the settlement during the winter months. Farinha was also an important food source in the seringal, and no meal was complete unless it is accompanied by farinha. A quadra or roÃ§ado was planted for two to three years in su ccession and then left idle to recuperate, normally for five to seven years. Capoeira nova referred to land that had been left idle for less than seven years. Capoeira velha was designated for land that had been left
80 unplanted for more than seven years. Genera lly in any year, a household would have two or three quadras or roÃ§ados in production. Travel in PDS SÃ£o Salvador In her work with the indigenous Mera tus Dayaks in Indonesia, Anna Tsing explained how travel was an essential ingred ient in the socio-poli tical construction of individuals and households. Furt hermore, travel was Â“part of the ordinary way of making a livingÂ…Â” (Tsing 1993: 154). Travel was also highly gender stratified, with men enjoying the benefits that the ability to travel offered in terms of status and political clout, and women restricted via their lack of movement. Women were restricted in their travel due to their domestic responsibilities, which limited their participation in the market and their ability to forge political connections. Travel was similarly important in the cons truction of self and in the well being of households in the seringal throughout Brazil. In Amazonia, trad itional subsistence lifestyles were lifestyles of traveling. Traveling within the seringal was necessarily integral to the livelihood system s of the rural peasantry, and oneÂ’s availability to procure resources for the household was heightened thro ugh travel. The peasantry traveled from their households to thei r agricultural fields, th e rivers and streams where they fished and collected water for household use, and the fo rests where they hunted and collected nontimber forest products. They also traveled from the seringal to th e city to sell their product and to purchase goods that they did not produce themselves. Individuals and households that were tempered in their ability to travel faced threat to their livelihoods. Resource users in the seringal were forced to travel to obtain the dispersed resources that are integral for human surv ival. Tropical forests are very heterogonous ecologically, and the botanical and faunal reso urces that are needed for human survival
81 are highly dispersed, necessitating that human s wander within the forest to obtain these resources (Peters 1994). The slash-and-burn agricultural syst em typical of the indigenous and rural peasantry in Amazonia, an agricultu ral system adapted to the climate and soils of the tropics (Posey and BalÃ©e, eds. 1989; Anderson ed. 1990; Redford and Padoch, eds. 1992), is an agricultural system of travel, as fi elds are cleared, planted, worked and then fallowed, necessitating a movement to the succ eeding agricultural site where the process is repeated every few years. These mobile fields of agriculture entail a cyclical movement through the system as the fallowed fields ( capoeira ) are returned to production in a renewal of the process of agricultural production. Travel is necessary in the construction and maintenance of livelihood systems of households, and further, is critical in the creation of social-politi cal bonds and alliances across households in PDS SÃ£o Salvador. MenÂ’s travel was often more flexible and varied in time and space than was womanÂ’s travel, re flecting the greater status accrued by male members in the household and larger community. The nature of travel in the seringal afforded men greater opportunity than wo men to forge alliances and accrue sociopolitical power. WomenÂ’s travel was centered on the house and its maintenance, especially that of its members, whereas me nÂ’s travel tended to be out in the world, engaged in productive and socio-political purs uits. Furthermore, the travel of men was also frequently free of surveillance by hous ehold and community members, as opposed to the travel of women that was accomplished under the watchful eye of the men (and communities) in their lives. One way in which men engaged in crea ting alliances across households by traveling within the seringal is throug h the hunt. Group hunting practiced among men
82 from different households within the community stre ngthened alliances across households. Often these group hunts entailed extended-stay, over night hunting outings that required men to be absent from their hous eholds for upwards to a week at a time. Women of the settlement did not participate in this type of hunting activity. The meat from the hunting of wild animals was an important source of food for the rural inhabitants in the seringal. Animal meat has also been shown to hold special significance in the food preferen ces of rural peoples in the Amazon region and is said to incur elevated status to the providers of meat for the household and community (Siskind 1973; Kensinger 1983; Stearman 1989, 1991; Hill and Kaplan 1993; Theil 1994; Wolff 1999: 177). Additionally, the reciprocal distribu tion of wild game fr om a successful hunt between households further aligns households in the community. Hawkes (1993) argued that the principal benefit in the sharing of resources, such as wild game, between households, comes not from the food that is received, but in the development of social bonds between different households within the community. Women were restricted in th eir ability to participate in the hunting of wild game through a set of cultural and lingu istic practices that reserved this activity for men. The hunting of wild game normally (though not excl usively) occurred within the forested areas of the seringal, a space that traditionally had been reserved for use by men. Both men and women spoke of the dangers of th e forest environment, that was deemed unsuitable for the presence of women. Men we re also the owners of the shotguns that were used in the hunt, and often women in th e household did not know how to fire a gun. These beliefs and practices acted as barriers to the participation of women in the hunt of
83 wild game, and thus, limited their ability to create alliances acr oss households in the seringal. Although women rarely participated in the k illing of wild game in the forest, they were often involved in the partitioning of game meat to other households in the community once the hunted animal had been br ought into the household. In this manner, women were afforded a space to build bonds with other members and households in the community, although the division of game meat was viewed as a subsidiary activity to the actual killing of the wild animal by the male hunter. In chapter 5, I explore in depth womenÂ’s participation in meat exchange be tween households. Other researchers have also noted that a gender defined task a ssigned to women in peasant households in Amazonia entails the division and sharing of game meat between households in the community (Harris 2000). WomenÂ’s spaces were critically involved with the domestic and reproductive functioning of the household. A critical space reserved for women was the house. WomenÂ’s travel in the seringal could be said to radiate from the centrality of the house. Here women were responsible for the clean ing of the home, cooking of, and cleaning after, meals, and maintaining the area close-by the house where women cared for small animals that were domesticated and consumed by the household2. The school, where mothers took the lead in the education of their children and where women were the teachers of the children of the community, a nd places involved with the health of the 2 Generally men cared for the larger animals of th e household (pigs, cattle) and women cared for the smaller animals of the household (chickens, ducks). These larger animals were frequently viewed as potential sources of income for the household, whereas the smaller animals raised by the women of the household were consumed by the members of the household and as such, were infrequently sold to neighboring households or in the markets in town.
84 members of the household such as the health outpost and hospital, were also considered feminine sites. Shared spaces between the genders tend to be centered around water, such as the rivers and streams where both men and wome n fished, bathed, collected water for the household, and traveled into, and out of, the se ringal. Riverbanks were important areas where women across households could gather to converse and share their lives with each other. Waterways were also open areas wher e women could easily be observed by their male family members and the larger community. Travel in the seringal was normally accomp lished on land by foot, or by boat in the rivers and streams that ran throughout the seri ngal. Paved roads were unheard of in the seringal, and the majority of the forest tra ils that ran throughout th e seringal were the remnants of the dirt paths that seringueiros utilized during the epoch of the rubber economy. These forest trails connected hous eholds within the community as well as joining neighboring communities. Residents ha d to tread knee-deep in mud to traverse these trails during the winter rainy season. Households in the settlement did not own horses or mules. Hunters traveled by land and via water in their pursuit of game. Most often, though, hunters traveled by foot from their hom es into the forest when hunting wild game. Hunters travel by themselves or in small groups of men (frequently not more than two people per group) when hunting in the fo rests near settlement areas. Sometimes hunters would travel by canoe one-half hour or more from their households on the waterways that flowed near their communities, to seek new hunting grounds. This
85 generally occurred if the hunter had not succes sfully captured wild game in his last several consecutive hunti ng trips. The necessity of traveling in the performance of the hunt afforded men and women of the household some free time from one anot her. While men engaged in the masculine defined activity of hunting and socialized wi th other men in group hunts, women pursued their gender defined tasks (chi ldcare, cleaning, fishing) whil e also having the opportunity to mingle with other women in the community. An aspect of the enjoyment of the hunt for men was the ability to escape the women in their lives through th e practice of hunting. Correspondingly, for women, they were free from the demands of the men in their lives while their husbands and sons were in th e forest pursuing wild game. A householdÂ’s desire to continue hunting (the ir emotional commitment to hunti ng) was partly tied to the necessity of traveling to hunt the wild animals that lived in the seringal, which allowed both genders some freedom from the other. Most long distance travel w ithin the seringal, and to communities outside of the seringal, was accomplished by motorized canoe . Canoes transported people, products, animals, and information within the seringal. Frequently a familial cluster of households grouped within a community would share in th e cost and maintenance of a canoe and its small motor. Households without a canoe, or the resources to pur chase gasoline for the canoe, had to rely on their canoe-owning ne ighbors for transporta tion on long-distance journeys. Passengers in the canoe owned by ot hers normally paid a small fee for travel, dependent on the distance to be traveled. These households lacking ownership of a canoe were among the poorest households in the comm unity and their accessibility to markets outside of the seringal to sell their product, or to inform ation that is often passed by word
86 of mouth, was hampered. Furthermore, the co st of passenger travel for individuals in poor households lacking a canoe of their own frequently prohibited all members of the household from traveling to the world outside the seringal. Often women in the poorest households had to sacrifice their travel for th e travel of men, as the familyÂ’s budget could not support the travel of all members in th e household. This helped to reinforce the notion of the house being a womanÂ’s place and the outside world as the place of men.
87 CHAPTER 3 METHODS This chapter examines the methods used throughout this stud y. There were two principal sets of objectives of this research pr oject. The first set of objectives was 1) to understand the influence of household orga nization and decision-making on peasant hunting strategy and behavior, and 2) to understand the effects of inter-household relationships on hunting practices. Additionall y, this project considered the historical trajectory of resource use by residents in PD S SÃ£o Salvador (explained in the previous chapter), and the implications of hunting for other forest based activities and the sustainability of the forest resource base within the settlement . The quantitative and qualitative methodology used in this project, as well as the research design of the study, were designed to not only test hypotheses of inter-household, and intra-household, relationships that influence hunting behavior, but further to understand how material, social, and symbolic realities and behaviors construct the practice of hunting with peasant households and communities in PDS SÃ£o Salva dor. Participant observation of productive and reproductive activities by residents of house holds within the comm unity enriched this ethnography of hunting by the peasantry of PDS SÃ£o Salvador. The unit of analysis in unde rstanding the practice of hunting in PDS SÃ£o Salvador was the household. Throughout much of Latin Am erica it is at the level of the household that the practices of liveli hood are carried out (Gudeman and Rivera 1990). It is argued here that the household is the appropriate uni t of study in reference to the practice of hunting, as the decision to hunt or not to hunt , with whom, where, and how to hunt, is
88 rarely left solely with the hunter himself/he rself. There always exist a range of intrahousehold, as well as inter-household, culturall y appropriate obligations that influence the practice of hunting. The household resource procurement strategies and practice of all household members critically affect the practice of huntin g. The hunter as a producer of wild game is not separate from, a nd cannot function without, his household and community, and the hunting household is not separate from, a nd does not function without, the hunters that resi de within the household. Households are different from businesse s in that the household is primarily interested in supporting and maintaining its elf, whereas the businessesÂ’ outlook is focused towards expansion of the unit (Gudema n and Rivera 1990). Furthermore, rural households in the seringal produce the m eans to reproduce themselves whereas corporations purchase the inputs necessary to produce their output. Because of this, the household Â“is never fully engaged with the market or dependent on itÂ” (Gudeman and Rivera 1990: 10). In PDS SÃ£o Salvador, this was especially true in the wet winter months when household production for market was lo w, or non-existent for some households, and the energy of members of the household is directed at maintenance of the household unit rather than production for market. Often the means to insure the reproduction of the household was through Â“belt-tighteningÂ” (Cha yanov 1986), necessitating a drop in consumption by household members. The concept of the household is a useful an alytical tool because it allows us to investigate economic, social, cultural, and poli tical life at a microcosmic level one step above that of the individual. The househol d is a rather small grouping of individuals where one can view how processes are c ontested, negotiated, a nd resolved between
89 individuals with often competi ng interests. As human bei ngs are social creatures, the existent relationships amongst household me mbers can serve to inform local-micro processes that build models to explain more complex human relationships that exist in communities, counties, or countries. In addition, the household is frequently a microcosm of societal symbol s, practices, and behaviors. The study of the household therefore allows for an understa nding of relationship s at the local level and can act as a link to how policies that are de signed and governed at a larger level of human complexity influence, and are acted upon, by the household and its members. Households, though, are always embedded in wider social and economic networks. Kin and other social obligations between hous eholds affect liveli hood strategies and how resources are procured, distributed and used. In the seringal, the re ciprocal obligations between households in the community were necessary for the survival of individual households and the collective survival of the community as a whole. Labor units comprised of members of two or more hous eholds sharing in agricultural work or hunting, and the reciprocal distribution of food resources between different households including principally wild game and fish resour ces, are examples in which the practice of constructing and maintaining oneÂ’s house hold was embedded within inter-household relationships. The Household The productive and reproductive activit ies of men and women within the household, and across households, influence hunting practice. Gender plays an often defining role in the rights, rules, and res ponsibilities of househol d members and in the assignment within the household of thes e productive and repr oductive activities by
90 household members. Gudeman and Rivera (1 990) state that the householdÂ’s connection to the material world is marked by gender. Production is defined as activit ies that increase, or have the potential to increase, the assets, capital, and/or income of the household. MenÂ’s production has been seen as primary to that of womenÂ’s production, which is viewed as supplemen tal in the realm of the householdÂ’s productive activities. Femini sts have commented on the cultural myth of the male breadwinner (Safa 1996) that elev ates the productive role of men in the household while simultaneously devaluing wo menÂ’s roles in insuring the productive strength and survival of the household. Reproductive activities are those that do not directly increase the income or wealth of the household, but include the range of func tions that insures the health, survival, and well being of all household members. Household reproduction occurs at three simultaneous levels: biological reproduc tion, generational reproduction, and daily reproduction (Ellis 1993). Biological reproduc tion entails the birt h and nurturing of infants. The care, upbringing, socializati on, and education of ch ildren in the household falls under the rubric of generational repr oduction. Children are socialized in the framework that is practiced with in the household and that is c onsistent with the social and cultural norms where they are raised. Finally, daily reproduc tion includes all of the other domestic tasks such as cooking, cleaning, care for the house, and caring for household gardens and/or domesticated animals that are necessary for household members to create and sustain livelihoods. Historically and cross-culturally, women in the household have been charged with ensuring that the overwhelming majority of the householdÂ’s reproductive needs are
91 satisfied, and mothers have been defined in accordance with their roles as the reproductive providers for the household. Fu rthermore, women are often encumbered with both reproductive and productive househol d responsibilities, while men typically have been free to concentrate in productiv e pursuits. Men, on the other hand, have generally been free of the reproductive care of the household and its members and have concentrated on ensuring that the productive need s of the household are met. As the productive and reproductive activit ies of all household members shape the resource strategies of the household, it is be st not to investigat e these practices in isolation from one another, but a more fruitf ul approach evaluates how the intersection of the activities of both genders within the househol d critically affects th e practice(s) of the other, thus, critically affecti ng the practice of the household as a whole. Furthermore, in many instances these activities are not self -contained by the memb ers of an individual household, but are routinely operated a nd managed across many households. Group hunting by residents of more than one househol d within the community, and the resulting sharing of wild game from a successful hunt , are two obvious examples of coordinated resource procurement strategies by more than on e household within the community. House and Household in Movement The house and household (residents of the house) were critical spaces of movement and travel within the seringal. Neither the actual structure of the house, nor the members who reside within the four walls of the house, were fixed ent ities within the settlement. This problematizes the definition of the house as a fixed structure, and the household as a set of regular, defined members (Grinker 1994; AugÃ© 1995; Rapport and Dawson 1998). A house has been described as a dynamic space that entails movement within and through this place, and is Â“both a concrete physical place and a personal space of
92 identificationÂ” (Olwig 1998: 225). In the pr ocess of movement th at is a search for identity (Rapport and Dawson 1998), houses, homes, and households expressed the interrelationship between Â“a conceptual spac e of identificationÂ” and Â“a nodal point in social relationsÂ” (Olwig 1998: 236). The physical placing of the house, and the members of the house that move into, within, through, a nd out of, the structur e of the house, were both highly fluid in the se ttlement. Although families in the Amazon region desire to live in loosely-grouped settlements of relatives (Harris 2000), households are constantly on the move in search of a better locale for the home (da Cunha and de Almeida 2002: 230). Houses in PDS SÃ£o Salvador were constructed with wood from trees that was available locally in the forests nearby the house, and the roof was commonly made of palm leaves, or increasingly, from sheets of aluminum purchased in town. Clim atic factors of high humidity and lots of rain, combined with insect and fungal pests, led to the rapid natural deterioration of the structure of the house within five to ten year s, necessitating locating another site for the construction of a new dwelling. The new house might be built next door to the old, abandoned house, or built far from the ol d house on another piece of land within the domain of the household. The older house, if located close to the new house, was frequently left standing, and wa s used as storage, or sometimes as a temporary house for a newly married son or daughter and his or her spouse, as the new couple built up a stock of resources to eventually cons truct their own house. During the epoch of the rubber system, houses in the seringal were widely dispersed, each located within the householdÂ’s colocaÃ§Ã£o the area of forestland that contained the rubber trees that each household tapped to produce rubber. Because of the
93 heterogeneous nature of natural stands of Hevea brasiliensis , the common rubber producing tree in the Amazon, that are not spac ed closely to one another, the colocaÃ§Ã£o was a relatively large expanse of forestland a nd houses were often located thirty minutes walk or further from one another. When th e rubber system collapsed in the late 1980s, and rubber lost its central importance in th e socio-cultural and economic livelihoods of the residents of the settlement, the residents of the seringal began to move out from the forest and towards the riverbanks and stream s that ran through the seringal, and houses began to congregate within small family sett lements of often only tw o or three houses, to as many as ten to twenty houses. Househol ds increasingly divers ified their livelihood strategies as hunting, fishing, and the gathering of forest products gained in importance for the survival of the household and its me mbers. Recently though, as agriculture gained increasing economic significance for the ca boclo household, some houses were in a process of retreat from locations along the ma in river-ways, and were moving back to the forestlands that were then cleared to become fields for cash-crop agricultural production. As was the case during the rubber epoch, on ce again increasing e ngagement with the market had led to the isolation of houses and households within the seringal. One male resident in the community of Rio Azul stated that he moved his house from the riverbank into the forest to be cl oser to better hunting gr ounds. Prior to this movement, his home was located in a grouping of six other houses that lacked edible fauna right next to the settlement grouping. As his household was one of the poorer of this household group (and fell towards the poor er end of the spectrum for all households within PDS SÃ£o Salvador), wild game was a critical source of food for his family. Moving his house into the forest meant easier access to wild game, less competition with
94 other households for the game resources that exist within proximity of the community, and less travel time for himself during hunting outi ngs to secure food for the household. Locating the house within fields of agricu lture benefited men of the household. It was men who took the lead in clearing land, planting, and maintenance of the householdÂ’s cash crops, and in the transport and sale of these crops, and thus the location of the house within these fields of production prioritized the producti ve activities of men in the household. Women who were responsib le for the majority of the householdÂ’s reproductive duties, including gathering wate r for household use a nd the cleaning of clothes and cooking supplies, were forced to walk longer distances to accomplish these tasks when the house was moved from the riverb ank to the roÃ§ado (agricultural field). Household membership was also in fl ux. Generational change in households occurred with the addition of new household members through births , and the loss of its members through deaths and marriage. In th e seringal, adolescent boys and girls became available for marriage in their middle to late teens, and new houses and households were established as these adolescents married a nd moved out of their parentÂ’s house. Generally people in the seringal married others that resided within their own community, or close by their communities people that they had known since their earl y childhood years. They established new households w ithin the community of the parents of the husband or bride, depending on the residentia l rules of the cultura l group, and if there was sufficient land and opportunity for th e new couple to cr eate a livelihood. Family members frequently traveled from households in the city to the seringal or vice-versa, residing in different households for periods of time from days to weeks, to months to years. Some of these mobile indi viduals shifted their lo cation of residence in
95 cyclical movements throughout the year, de pending on work, familial, and personal obligations and desires. Often a household in the seringal would re quest labor help from family members living in the ci ty during critical times in th e yearly agricultural calendar, when the work regime was greater than what could be accomplished by members residing in the household. Individuals lacki ng steady and secure employment in the city had the freedom to accommodate the labor demands of their families that lived in the seringal, and the household in the seringal recei ved a laborer that was trusted, could work with little guidance or supe rvision, and was usually knowle dgeable of the terrain and work assignments that were expected of him or her (Netting 1989). Separated and divorced mothers would travel to, and from, th e seringal, to reside with family members who could give them emotional and materi al comfort and support during their time of need. Jilted lovers would also Â“escapeÂ” to, or from, the se ringal, sometimes for several months at a time to lessen the pain of lost love and to attempt to forget their past lovers. Although there was frequent movement in te rms of the structure of the Â‘houseÂ’, and the members of this structur e that collectively form the Â‘householdÂ’, in this study, the household is conceptualized as members who shared a common roof, generally ate from the same pot, and who collectively shared the productive resources of the household. Only permanent (year round), or semi-perma nent members who lived within the house, and shared in the productive and reproductive wo rk regimes within the household for half of the calendar year or more , were counted as members of the household. The critical distinction for membership within a hous ehold was the sharing and/or pooling of productive assets. In several instances in the settlement, teachers from the cities of MÃ¢ncio Lima or Cruzeiro do Sul lived with a family in the community where they taught,
96 the family provided them with shelter a nd food, but income earned from teaching was kept separate from the wealth accrued by the peasant household. Accordingly, the peasant household kept their productive assets separate from the teachers that they provided with a roof over their heads, and food. Fieldwork Field research began in late July 2003, lasting until early June 2004. Data collection within the communities of PDS SÃ£o Sa lvador lasted for ten consecutive months from August 2003 to May 2004. Data was coll ected by myself and two local research assistants who lived in two of the ten communities of PDS SÃ£o Salvador. Several different methods were used in this st udy, including household surveys, a list of productive and reproductive activities of house hold members, time allo cation analysis of daily activities of household members, and surveys of hunting outings by selected households in two communities. Partic ipant observation within households and communities was also used throughout the fieldwork portion of this study. Each of these methods is described in more de tail below. Figure 3-1: Research Assistant Interviewi ng Resident of the Community Boa Vista.
97 Sampling within each community was conduc ted in consecutive periods of from one to three week visits to a community. A ll necessary supplies for each data collection visit to a community were tr ansported in motorized canoe up the Moa River to reach the community where research was conducted. Du ring my stay in a community, I resided with a family in one of the households in the community where data we re to be collected. Research assistants living in the communities of Boa Vista and Vai Quem Quer were responsible for collecting spot sampling surveys of daily activities of household members within their communities. The research assi stant living in the community of Boa Vista also collected hunting outing data in her co mmunity. All other data were collected by myself. Frequently, data from several differe nt methods were collected in each visit to a particular community. Table 3.1 shows the fiel d research schedule of data collection. Table 3-1: Field Research Schedule 2003 2004 Method July AS ONDJan F M A M June Household Surveys XXXXX X X Productive/ Reproductive Activities XXXXX X X Time Allocation XXXXXX X X X Hunting Outings X X X X Participant Observation XXXXXX X X X X Methodological Considerations The methods used throughout this study were employed in the investigation of two sets of objectives. The firs t set of objectives was to te st hypotheses related to the influence of household (inter-household) or ganization and decision-making on peasant
98 hunting strategy and behavior. Second, was to understand the effects of inter-household relationships on hunting practices. Hypothe ses were focused on the understanding of material, gender and symbolic relationships within each household that influenced the practice of hunting in PDS SÃ£o Salvador, and social obligations acro ss households in the community that affected hunti ng practice of households. Methods used to test intra-household (H1 Â– H7) and inter-household hypotheses (H8, H9) on hunting output were derived primarily from household surveys. Productive and reproductive household surveys, and time allocation surveys provided important supplementary data to conclusions reached through hypothesis testing. These methods were critical in assessing the relationship of gender in the behavioral output of adult members of the household in terms of live lihood strategy and labor use by the household, as well as the seasonal nature of hunting in the settlement. Ethnographic material resulting from hunting outing surveys, partic ipant observation, and taped interviews with hunters supplemented methods used to test inter-household hypotheses (H8, H9). These methods contributed to understa nding the social nature of th e hunt as a means of joining households in the community. Each of these met hods is described in further detail below. Household Surveys Household surveys were collected in six of the ten communities of the settlement: Boa Vista, SÃ£o Francisco, Vai Quem Quer , Girassol, ConceiÃ§Ã£o, and Rio Azul. A sampling frame of all of the households that exist within PDS SÃ£o Salvador was not available, and the possibility of constructing one was beyond the scope of this project. Several of the communities we re virtually inaccessible at th e end of the dry season (SÃ£o Pedro, TimbaubÃ¡) when the small streams that dr ain into the Moa River flow to a trickle, preventing even the smallest of motorized canoes to pass. The entrance to another
99 community located within the forest in PDS SÃ£o Salvador (Vai Quem Quer) floods at the height of the rainy season, with the water level too high to walk through with supplies on oneÂ’s back, yet too low for boats to pass. Th erefore, it was decided that a random sample of all households within the settlement was not feasible in this study. It was decided to select several communitie s to sample, based on several criteria. First, community leaders and other members of the community were asked if they wished to participate in the stud y. Although I had originally intended to conduct household surveys in four different communities, it quick ly became apparent that the small sample size (n) would have severely lim ited the running of statistical analysis on the basis of the samples collected solely in these four co mmunities. Two additional communities were therefore sampled with household surveys. Six of the seven communities that were approached by me agreed to par ticipate in this study. Another criterion used in the selection of which communities to sample was the geographical layout of the community. One crit ical factor was if the houses within the community were grouped together in a central ar ea (in clusters), or if they were spread along the banks of the rivers that run thr oughout PDS SÃ£o Salvador. This was important in testing the effect of inter-household relationships on hunting, based on whether households had regular, daily contact with other households in the community, and if households had a clear line of sight to other households in the community. Additionally, it was desired to select communities based on the distance, and time, one must travel by motorized canoe to reach the city of MÃ¢ncio Limathe principal market where residents of the settlement sell their agricultural pr oducts and purchase supplies. The communities of Boa Vista and SÃ£o Francisco were the two closest communities to the city of MÃ¢ncio
100 Lima, about a five hour boat ride downstr eam, whereas the community of Rio Azul, on the far western end of the settlement borde ring the Serra do Diviso r National Park, lay from eight to twelve hours by boat from MÃ¢nc io Lima depending on the time of year one was traveling, and the size of the boat in wh ich one was traveling. It was thought that communities further removed from the market centers might have a different relationship with the natural resource base than those communities that had easier access to market their household products. Households that we re located on the peri phery of the market system might exhibit market avoidance (l ivelihood strategies relying heavily on nonmarket transactions including meat shari ng and trading labor days with neighboring households) and therefore production might be directed at sustaining household consumption needs rather than increased mark et integration. Finally, another important criterion in the selection of communities was accessibility. One community was not favorable to participating in this research project and was therefore, inaccessible. Two othe r communities were difficult to reach for several months of the year, and furthermor e, the households in these communities were widely spread along the banks of small stre ams that run to the Moa River, making the logistics of reaching these households very difficult. Lack of time, money, and manpower prevented me from working within these communities. Unfortunately, the resource use by households within these co mmunities that lay on the margins of the settlement, and are difficult to access (SÃ£o Pe dro, TimbaubÃ¡), might yield interesting results that are different from the patter ns of resource use by households in other communities in the settlement.
101 Due to the small number of households th at exist within the communities of PDS SÃ£o Salvador, an attempt was made to sample all of the households w ithin each of the six communities sampled in the study. Unfortunately, this was not always possible. The households that were not interviewed within the six communities sampled in the study were each visited on more than three occasions, but the members of these households were traveling outside of the community when I visited their houses. Repeated visits to these households during different months of the fieldwork calendar yielded similar results. Household surveys were conducted with all adult members that resided within the household present at the time when the surv ey was given. An adult member of the household was defined as someone fifteen years of age or older who lived and shared in the productive and reproductive assets of the household for si x months or longer of the year. At age fifteen, adolescent boys within the community begin to be hired as daily laborers at a rate of R$10 per day1 by other households in the community and many begin to hunt without supervision by elder male members of thei r family. Adolescent girls become marriage eligible at around fifteen years of age. Often, one or more adult members of the household were not present when I was in the community. Repeated efforts to conduct surveys with these households were unsuccessful because adult household members were traveling for various reasons. Therefore these households were not sampled in the study. Table 3.2 below shows the communities wher e sampling of households occurred and the number of households sampled in each community. It should be noted that the 1 This was the equivalent in late 2003 of about US$3.30 for an eight-hour day of agricultural labor.
102 number of households in each community may have changed since fieldwork was completed in June 2004, as new residen ces evolve through the creation of new households via marriage within the communit y, or by the movement of family members from the city centers of MÃ¢ncio Lima, Cruzei ro do Sul, and other ne arby urban areas to PDS SÃ£o Salvador. Table 3-2: Household Sampling Community Total No. of Households Number of Households Sampled Percentage of Households Sampled Boa Vista 13 13 100 SÃ£o Francisco 6 6 100 Vai Quem Quer 7 7 100 Girrassol 10 7 70 ConceiÃ§Ã£o 10 8 80 Rio Azul 18 18 100 TOTAL 64 59 92 Household sampling occurred in six of th e ten communities that exist in PDS SÃ£o Salvador.2 A total of fifty-nine households were sampled in these six communities. This represented ninety-two percen t of the total number of house holds (sixty-four) in the six communities as of June 2004. All of the househ olds were sampled in four communities, with seventy percent of the households samp led in the community of Girrassol, and eighty percent of the households sample d in the community of ConceiÃ§Ã£o. Household surveys collected basic socio-ec onomic data of all household members. In addition, the questionnaire asked specifi c questions that pertained to the hunting behavior and beliefs of adu lt household members. An attempt was made to collect information from both men and women of th e household, although in many instances it 2 The most recent map of the settlement (completed by INCRA in January 2003 ), listed a total of 150 numbered lots corresponding to 150 different households. Therefore, the 59 households sampled in this study represented 39% of the total number of known households in the settlement.
103 was the senior male of the household who gave the majority of responses. The household survey was used to collect information as to how different independe nt variables affected the dependent variable, frequency of hunting (measured in terms of number of hunting outings by household members per month). Additionally, open-ende d questions in the household survey were designed to uncover attitudes and beliefs important in the construction of the practice of hunting by house holds in the settlement. See Appendix A for an example of the househol d survey used throughout the study. Productive and Reproductive Household Survey The productive and reproductive household survey measured typical daily work activities that were th e responsibility of adult members of the household. An attempt was made to ascertain the household work regi mes of men and women within the household. These data were used in support of the anal ysis of hunting behavior s and patterns of the household as a whole. This survey was an important tool in measuring the impact of gender on the practice of hunting with households in the settlement. Clearly delineating the productive and reproductive activities that men and women within the household were engaged in, gives clues as to how hun ting practices and beliefs were constructed within the household. See Appendix B for an example of this survey. The productive and reproductive household survey was constructed with the same households where the general household survey was completed. As with the household survey explained above, an attempt was made to capture the responses of all adult members of the household. This was the case in all but a few instances . In total, fiftynine of these surveys were recorded.
104 Time Allocation Surveys Time allocation methods have been used by researches in Amazonia to measure hunting practice and activity. Hames (1979) us ed time allocation surv eys to estimate the amount of time Yanomamo men spent hunting. Yost and Kelley (1983) collected data for nearly one year on hunting by the Waorani of eastern Ecuador. These data were used to suggest the hunting superiority of the shotgun over native weaponry (blowgun, spear). Time allocation surveys are used by social scientists to measure how individuals spend their time throughout the typical workda y, season, or year. By directly recording activities as they occur in the field, the rese archer is able to empirically measure Â“the behavioral Â‘outputÂ’ of decisions, prefer ences, and attitudesÂ” (Gross 1984: 519) of individuals as they perform th eir normal daily routines. Seasonal variations in productive and reproductive tasks can be qua ntified through the use of time allocation surveys. This is important in Amazonia as seasonal variat ions have a tremendous effect on patterns of resource use by indigenous and peasant groups (Chibnik 1994; Beckerman 1994; Descola 1996; Harris 1998; Harris 2000). Beckerman (1994) writes that the cy clical seasonal change in rainfall is the critical variable in the Amazonian environment that determines hunting (and fishing) strategy. People hunt mo re when fishing resources are harder to obtain in the wet winter mont hs as the tropical forests of Amazonia flood, allowing fish more areas to hide within the tangled forest understory. Time allocation surveys can measure the amount of time individuals devo te to hunting, fishing or other productive tasks, and the seasonal effects on the use of time by individuals. The effects of gender on the work regime s within the household can further be analyzed using time allocation surveys. Cr itically, the amount of time men and women engage in reproductive and productive activit ies can be quantitatively estimated through
105 the use of time allocation methods. Femini sts have long argued that the reproductive functions that women perf orm within the household such as cooking, cleaning, and childcare, allow men to engage in productive pursuits that bring greater prestige to individuals involved in thes e productive pursuits (Bener Ãa 1979; Bouquet 1984; Dwyer and Bruce 1988; Moore 1988; Kabeer 1994; Leach 1994; Safa 1995; Rocheleau et al. 1996; Sachs 1996; Deere and LeÃ³n 2001). Of im portance in this study was to measure how much time men and women were devoting to productive and reproductive activities (and leisure), and to attemp t to quantify the amount of time that the household participated in hunting. Gender was a key va riable in the performance of the hunt in peasant households and womenÂ’s engagement in productive and reproductive household activities critically a ffected the hunting output of men in the household. Therefore, an empirical measurement of the behavior out put of menÂ’s and womenÂ’s daily productive and reproductive activities allows the resear cher to see how hous ehold decisions and preferences on resource use and la bor assignment are carried out in practice. In this study, spot sampling of households was used as a proxy to estimate the amount of time adult men and women engaged in productive and reproductive activities, as well as leisure. Spot sampling entail s unannounced randomized visits to households recording observations of behavior as they na turally occur. The researcher attempts to record behavior as it happens before the obser ved individual is aware that he or she is being watched. If the house is empty when the researcher arrives to record behavior, then he/she can search for household members in the community to discover what activities they are engaged in, or if this is not po ssible, one can ask neig hbors in the community what the individuals of the household in que stion are doing at that point in time.
106 Spot sampling of household membersÂ’ behavior in PDS SÃ£o Salvador was conducted for a period of nine consecutive months from August 2003 through May 2004 in the community of Boa Vista, and for eight consecutive months (September 2003 through May 2004) in the communities of Rio Azul and Vai Quem Quer.3 Samples were recorded by myself in the co mmunity of Rio Azul, one rese arch assistant living in the community of Boa Vista, and another livi ng in the community of Vai Quem Quer. Research assistants were trained over a peri od of two days on how to observe and record behavior, prior to conducting these samples w ithout direct supervision. Behavior was recorded only for adults of the household4 and was taken for all adult individuals in the house. If no adults were present at the hous e when samples were scheduled to be taken, children of the household that were present were asked what their parents and older siblings were doing and this was recorded as behavior of the adults of the household at that time. Often, though, if a sample was sche duled to be taken later in the day and adult household members were known ahead of time to be away from the community center for most, if not all of the day, engaged in act ivities such as hunting, fishing, or processing of farinha for example, these behaviors were recorded at the time of the sample. If no members of the household were present and I, or the two research assistants, did not know what activities the adult members of the household we re doing at the time of the scheduled spot sample, it was noted that no one was present at the time of the sampling. 3 The first three months of the dry season in 2003 (May through July) were missed in the spot sampling exercise. 4 Although the work of younger children (less than fifteen years of age) is often essential to the productive, and reproductive, maintenance and output of the household, it proved difficult to track their activities as often children accompanied their parents and older sibli ngs in their daily work activities (for example, the making of farinha), but the work activities of these you nger children in these vent ures were sporadic and inconsistent.
107 Appendix C shows an example of the behavi ors that were observed and recorded in the three communities throughout this exercise . These behaviors mirrored the activities that were explored in the pr oductive and reproductive household survey described above. In addition to these activities, leisure activities, travel, eating, sleepi ng, and resting were recorded of adult participan ts. Although individuals were often engaged in several activities simultaneously, it was left to the researcherÂ’s discretion to record only the principal behavior as it was observed. It is not uncommon especia lly for women of the household to be involved in several daily activ ities at once (often involving childcare) as a time saving stratagem, but the decision was made to prioritize, and note, one behavior as primary. This allowed for one to place a percentage on the amount of time individuals were engaged in specific behaviors, enta iling that the sum of all of the observed behaviors added to one hundred percent, al though it might underestim ate activities that are frequently combined. All spot sampling household visits were randomized, visits to randomly selected households (using a random number table). Spot samples were conducted four times per day, between 6am and 6pm. It was reasoned th at we could comfortably visit households four times per day given the other daily obl igations of the stu dy and with minimal interference in the daily live lihoods of the two research assistants, both mothers, whose typical day required them to participate in the reproductive and productive maintenance of their households. Furthermore, samples we re recorded only duri ng the daylight hours for two principal reasons: 1) as a means to avoid unnecessary intrusion in the lives of individuals and families in the communities where the sampling occurred, to respect the householdÂ’s privacy, and 2) sampling at ni ght was potentially dangerous, walking by
108 flashlight often through forests where anim als and other hazards awaited unsuspecting travelers. Spot samples also noted if it was raining or not. Spot samples were recorded on consecu tive days, Monday through Saturday only. Sunday was generally a day of rest in the se ringal and therefore, it was decided that sampling on this day would not adequately capture the typical daily reproductive and productive work regimes of adult househol d members. The number of days of continuous sampling within one community, th e day of the week, which week in the month to begin sampling, and the time of day of sampling, were chosen using a random number table. Samples were taken for five to seven consecutive days in the communities of Rio Azul and Vai Quem Quer, and for seven to fifteen consecutive days in the community of Boa Vista. All households were in the poten tial pool of sampled households in the communities of Boa Vista and Vai Quem Quer, and a subset of seven houses in the community of Rio Azul were sampled. With the exception of one household in the community of Boa Vista, a ll other households were within a thirtyminute walking distance from the researcher . The attempt was to avoid approaching households by motorized canoe, as it is customary for indivi duals in the seringal to halt what they are doing to observe approaching boa ts as they travel along the waterways in the seringal. Spot sampling necessitates capt uring behavior unexpectedly as is occurs naturally, and because households in the co mmunity of Rio Azul are widespread along the banks of the Blue River (Rio Azul), re quiring one to travel by motorized canoe to visit all households in the community, this pr ecluded spot sampling of all households in Rio Azul.
109 Hunting Outings The hunting outing survey was designed to investigate the ac tual practice of hunting by hunters in the settlement and the so cial interaction of hunters from different households in the community. The output of the hunting party was captured in terms of the number and kind of prey that was captured, as well as the makeup of the hunting party. In addition, the locati on of the hunt, period during th e day (or days) that the hunt occurred, and the number of shots fired duri ng the hunting outing were all recorded. It was also noted if it was raining or not the da y of the hunting outing survey. This methodology followed hunting practi ces of hunting households in two communities over a period of four months (January 2004 through April 2004). Four hunting households were sampled in the commun ity of Boa Vista and three households in Rio Azul. The households that were selected to participate in this exercise were based on the results obtained from the household su rvey explained above. Hunting households were followed for seven consecutive days pe r month for four consecutive months. The day of the week and the week of the m onth that sampling occurred were randomly selected using a random number table. At th e end of each of the seven consecutive days hunters were interviewed, after they had retu rned to their households, about their hunting outing. Hunters that made extended-stay, ove rnight hunting trips were interviewed when they returned to their households and communities, following the conclusion of their hunting activity. A drawback to this methodology was the relatively short dur ation of time over which the samples in this exercise were cap tured. Hunting outing data was measured for only four months of the year and should not be extrapolated ove r twelve months to estimate yearly hunting activity of hunting house holds in the settlement. Furthermore,
110 hunting outings were only measured in one s eason. To adequately measure if hunting activity does indeed fluctuate seasonally in Amazonia, sampling should occur both during the wet and dry seasons. Unfortunately, it wa s not possible to collect hunting outing data during the dry season in 2003, as this tim e was devoted to visiting and meeting communities, presenting the proposed research to residents of PDS SÃ£o Salvador, capturing household survey data of the households selected to participate in this project, along with time allocation survey s used throughout the duration of fieldwork. In spite of this, the four months of h unting outing data gave some clues about hunting activity by hunting households during one season and th e interaction amongst hunting households in the community. Participant Observation The ten months of continuous fieldwork in western Acre allowed me to participate in the regular daily activities of the peasant households in PDS SÃ£o Salvador. I ate, slept, bathed, conversed with, and lived side-by-side with the members of the households that I stayed with throughout my fieldwork in the settlement. I accompanied residents of the settlement during their daily work regimes in the roÃ§ado and other agri cultural fields, as they hunted in the forests and fished in the rivers that run throughout the seringal, and as they prepared farinha for market. I watche d and noted social interaction of household members and between residents of the differe nt households in the community. I also visited schools, traveled with local residents attending meetings and workshops organized by NGOs and state government officials in the seringal and in the ci ties of MÃ¢ncio Lima and Cruzeiro do Sul, and visited with residents of the settlement as they traveled to visit their families living in nearby urban areas. I accompanied residents to parties in nearby
111 communities, played soccer with my hosts, and traded life stories with them, sharing our pasts and dreams for the future. Participant observation afforded me the opportunity to capture residentsÂ’ livelihoods in motion, thus enriching th is ethnography of hunting beyond purely the results captured from quantitative met hodology. Conversations over dinner, while traveling by land or water in the seringal , and in the urban environment provided different environments in the dialogue of this ethnographer and the subject(s). Qualitative methodology throughout this proj ect complemented quantitative data collection, improving the accuracy and quality of the information collected. The research design of this project necessitated the marriage of quantitative and qualitative methodology throughout the fieldwork, analysis of data , and writing of this dissertation.
112 CHAPTER 4 HUNTING WITHIN THE HOUSE HOLD (INTRA-HOUSEHOLD) This chapter traces hunting practices fr om the perspective of the individual household. Later chapters will discuss inte r-household influences on hunting and their effects on hunting within the settlement. A de scription of the practice of hunting as well as how household relationships affect hunting practice and behavior, are included in this chapter. We will also look at how gender role s within the household play a critical role in the assignment of the householdÂ’s produc tive and reproductive activities. The differences in frequency of hunting across households, as well as differences across communities are analyzed in this chapter. The seasonal nature of hunting in PDS SÃ£o Salvador will also be explored. One key research question of this st udy was to understand the influence of household organization and deci sion-making on peasant hunti ng strategy and behavior. We tested hypotheses pertaining to the mate rial realities of th e household, the gender relationships of household members, and the emotional commitment to hunting of household members that influence hunting pr actice by household members. Hypotheses testing included T-tests of mean values for material variables (H1-H3) and gender variables (H4-H6). The vari able Â“emotional commitment to huntingÂ” (H7) was analyzed with PearsonÂ’s Correlation (r). In addition, a linear regression equation was used to model the frequency of hunting dependent on th e material and gender variables analyzed in hypotheses testing.
113 The Hunt in PDS SÃ£o Salvador The overwhelming majority of households surveyed in this study were hunting households. The average number of times a ll households surveyed in the study hunted per month was 2.08 (sd = 1.85) with a median of two hunting outings per month. Fortynine of the fifty-nine households interv iewed (83.1%), had at least one household member that hunted wild game in PDS SÃ£o Sa lvador. Households varied in their frequency of hunting activity, though. Almost one-half of the households surveyed in this study either did not hunt, or hunted one or fewer times per month (n = 28; 47.5%). Only eight households reported that they hunted more than three times per month (13.6%). Figure 4.1 shows hunting frequency (i n terms of hunting outings per month) by household. 11 13 16 11 2 3 2 1 0 2 4 6 8 10 12 14 16 18012345610 Number of Times Hunt/MonthNumber of Households Figure 4-1: Freque ncy of Hunting
114 Note that the frequency of hunting is skewed to the left and does not approximate a normal distribution with the skewness of the distribution 1.68 and Kurtosis 4.79. Skewness shows the symmetry of the distribut ion while Kurtosis indicates how peaked, or flat, the distribution is (Pallant 200 1). A normal populati on distribution has a skewness and Kurtosis of less than 1.0. Th e decision was made here to transform the dependent variable, frequency of hunting, usi ng the logarithm of th e responses of the dependent variable measured in the survey, so that it conformed to a normal population distribution.1 This transformation of the variable gave a skewness of -.225 and a Kurtosis of -.398. This allowed for the use of parametr ic statistics (eg. P earson correlation) that are a more powerful form of statistical analysis than non-parametric statistics (SpearmanÂ’s rho) that do not assume nor mal population distributions. All predictive statistics used in subsequent analysis in this dissertation includ ing students t-tests, PearsonÂ’s correlations, and multiple regression utilized the logarithm transformation of the dependent variable, frequenc y of hunting. Many hunters stated that they hunted more in the past (early to middle 1990s) when their livelihoods depended on th e sale of wild game, and wh en additionally, the seringal was seldom patrolled by the governmental bodie s that are responsible for enforcing the prohibition on the sale of wild game throughout Brazil. One hunter who sold meat from wild game in the past said that he stopped hunting when the area was declared a settlement and he was no longer able to sell th e meat from the animals that he had hunted in the forest. As agriculture became the dominant form of livelihood for households in 1 There are several techniques that can be used to tr ansform variables (mathematical ly modify) so that they conform to a normally-distributed curve, including using the square root, logarithm, or inverse of the variable to be transformed. Pallant 2003 suggests using the logarithm to the base 10 transformation for population distributions that resemble the one in Figure 5.1.
115 the area in the past decade, hunting frequenc y waned, and many residents turned to the substitution of domestic animals for meat for their households. Over half of the households interviewed stat ed that they did not enjoy hunting (n = 31; 52.5%). These households hunted primarily because they were obligated to provide food for their families2 (see Table 4.1). Others cited th at hunting interfered with their agricultural work, part icularly the production of farinha that was the main source of income for households in the settlement. These hunters stated that their agricultural responsibilities ( serviÃ§o ) did not afford them adequate time to hunt. Many hunters complained that because of past hunting, animals were no longer close to their communities and they had to walk several hours to matar um bicho (kill an animal). Older men in the settlement sa id that the physical demands necessary to be a successful hunter were beyond their capabilities, and th at hunting was restri cted to younger men who had the stamina to pursue wild game throughout the forest. Table 4-1: Households that Enjoy Hunting Yes No No Response Households that enjoy hunting 24* (40.7%) 31 (52.5%) 4 (6.8%) *Numbers refer to number of households (n = 59). Some hunters stated that they enjoyed hunting and would not cease this activity even if they had another reliable source of food for their families (e.g. domesticated animals, fish, or purchasing meat). Men also sa id that they walked the forest in pursuit of game, sometimes, to escape the barulho (noise) of the household.3 Another hunter 2 A study of peasant hunters in the Peruvian Amazon similarly found that hunting was primarily for subsistence meat, with economic returns from meat sa les in urban markets a secondary concern (Bodmer 2004). 3 Women also commented that they prefe rred to Â“escapeÂ” to their work in the roÃ§ado (field of manioc) to forget the daily pressures they had in their lives. The roÃ§ado though, was usua lly located close to the
116 replied that Â“the forest helps peopleÂ” becau se it is Â“calm in the forestÂ”. Still another hunter said that hunting is part of Â“the culture of the regionÂ”, and th at he hunts because Â“it is the jeito4 I have.Â” Adolescent boys frequently said they liked to hunt because they liked to fire their guns and kill wild animals. Prior to beginni ng to be educated in the ways and means of hunting with firearms by their fathers, older brothers, or other adult relatives in the community, it was common to find boys, individua lly or in small groups, with slingshots taking aim at birds that were close to thei r houses. This early hunting experience was invaluable for these future adult hunters and heads of households, to learn the habits of the wild animals that lived within their midst, and the terrain that these animals inhabited. Adolescent boys began to accompany their adul t male relatives on hunting outings into the forest in their early teen years, and began to hunt alone usually by the time they reached their late teenage years. There was a marked similarity in the age at which the household head and his son(s) began to hunt in the se ringal. Table 4.2 shows that on average, both fathers and sons began to hunt at age 14. There is more variability though, in the age at which the household head began to hunt (sd = 4.61) with a range of from 8 to 30 years of age as opposed to the age their sons began to hunt (sd = 2.65), ranging from 8 to 18 years of age. This partially reflects that all of the s ons who at that time engaged in the practice of hunting were born in the seringal, whereas se veral of the hunters who were heads of house, and was an open field where oneÂ’s activities could be observed, whereas visibility was limited within the closed canopy forest where most hunting activity occurred. 4 Jeito has a variety of English meanings that may include some of the following: Â‘wayÂ’, Â‘mannerÂ’, Â‘knackÂ’, Â‘skillÂ’, Â‘appearanceÂ’, Â‘personalityÂ’, and/or Â‘style.Â’
117 households were born in urban ar eas, moving as adults to the se ringal in search of land to work and an area to raise a family, therefore l earning to hunt as adults after they migrated to the seringal. An additional difference between hunting fathers and their hunting sons was that all of the sons in terviewed who hunted reported th at their fathers taught them how to hunt, whereas their fathers stated that they were trained to hunt (in addition to their fathers) by their uncles, older brot hers, neighbors, or in a few cases, taught themselves how to hunt. Table 4-2: Age that Hunters Began to Hunt Number Mean Minimum Maximum SD Household head 49 14.08 8 30 4.61 Sons 9 14 8 18 2.65 There was some minor variability in the hunting output of households as compared across communities in the settlement. One might expect to find differences in hunting frequency based on the relative access to the market in each of the ten communities. Studies have shown that the market is a signi ficant factor in infl uencing resource use and livelihood strategy of peasant and indigenous households (Godoy et al. 1997; Henrich 1997; Godoy 2001; Wallace 2004). One way that market access can be measured is by the travel time required to go from comm unity to market centers where household products are sold. It is possi ble that communities closer to market centers may prioritize income generating pursuits (in this case agri culture, primarily the production of farinha) because they have easier access to these mark ets, over other activities that do not bring income to the household (such as hunting). Here we investigat ed hunting frequency within the community as a f unction of the travel time by motorized canoe required to reach the city of MÃ¢ncio Lima, the closes t to, and principal urban center from, the settlement where household products were marketed.
118 Table 4.3 shows average hunting frequenc y, per household, per month in the six communities where household interviews we re conducted. Also shown are average travel times by motorized canoe from the communities in PDS SÃ£o Salvador to the city of MÃ¢ncio Lima.5 Table 4-3: Hunting Frequency across Communities Community Travel Time (hours)* Households Interviewed (N) Hunting Households Mean Stan. Dev. Min. Max. Boa Vista 7.5 13 10 1.62 1.61 0 6 SÃ£o Francisco 7.75 6 6 2.13 0.75 1 3 Girrassol 8.5 7 6 2.71 1.80 0 6 ConceiÃ§Ã£o 9 8 7 2.13 1.96 0 5 Vai Quem Quer 9 (+ 20min walk) 7 6 1.86 1.68 0 5 Rio Azul 10 18 14 2.22 2.37 0 10 TOTAL 59 49 2.10 1.85 0 10 *Travel time was from the community (downstream) to the city of MÃ¢ncio Lima during the high-water summer months. Travel times upstream might add several hours to the journey. From Guerra (2002). Hunting frequency in the communities of PDS SÃ£o Salvador was somewhat responsive (though slightly) to the distance fr om market. The community of Boa Vista which lies the closest to the city of MÃ¢ncio Lima had a mean frequency of hunting, per 5 Although all communities in the settlement were far from urban markets, travel times could vary greatly depending on the rise and fall of the waterways in th e settlement. Near the end of the dry season in the months of September and October, large boats were not able to pass on the Blue River (Rio Azul) and even small boats frequently got stuck on the bottom of th e shallow river requiring boat operator and passengers to climb out of the boat, pushing and pulling the boat free from obstruction. This could add several hours to oneÂ’s journey to, and from, the city. Near the end of the dry season in the month of October, the low water levels of Rio Azul did not permit the passage of large boats. Therefore, residents of the community of Rio Azul were restricted in the amount of their product (farinha sold in 50-kg sacks) they were able to transport to market. Often residents of the farthest communities from the city of MÃ¢ncio Lima (Rio Azul, TimbaubÃ¡) would spend two days in travel time returning to their community, spending a night in one of the closer communities (usually Boa Vista or Sede) on their return trips back to their community. The community of Vai Quem Quer had the opposite problem in that because the community was situated a 20 to 30-minute walk inland from the Rio Moa, the rising waters in late winter prevented individuals from walking sacks of farinha on their backs to their waiting boats on the Moa for transport to urban markets. Some individuals chose not to travel out of their community during the months when river transport became increasingly difficult.
119 month, per household of 1.62 outings/month (s d = 1.61) whereas the furthest community from MÃ¢ncio Lima where households were interviewed in this study, Rio Azul, had a mean of 2.22 hunting outings/month (sd = 1.85). The mean for all six communities combined was 2.10 hunting outings/month (sd = 1.85). The other four communities that lay between the communities of Boa Vista a nd Rio Azul had varying degrees of mean hunting frequency, ranging from a low of 1.86 hunting outings/month (sd = 1.68) for households in Vai Quem Quer, to a high of 2.71 hunting outings/month (sd = 1.80) within the community of Girrassol. Distance to market appears not to be an influence on whether or not a household engages in hunting activity. There was a negligible difference in the percentages of hunting households in Boa Vista (76.9%; 10 of 13 households), the closest community to market within the settlement, with Rio Azul (77.7%; 14 of 18 households), the furthest community to market measured in this study. Randomized, spot sample visits were ma de to households in three communities, Boa Vista, Rio Azul, and Vai Quem Quer, to measure the produc tive and reproductive activities of all adult household members. Samples taken between 6 am and 6pm, four times per day, monthly, over a nine-month, c onsecutive period were used as a proxy to account for the percentage of time hous ehold members engaged in productive and reproductive pursuits, personal care and hygiene, leisure activities, a nd travel outside of the community. A total of 1585 observations were recorded fr om August 2003 through April 2004: 893 (56.3% of all observations) in Boa Vista, 393 (24.8%) in Vai Quem Quer, and 299 (18.9%) in Rio Azul. Table 4.4 below shows spot sampling data for households, by community.
120 Table 4-4: Spot Sampling of Adult Activities by Community Activity Boa Vista Rio Azul Vai Quem Quer Total all 3 comm. Cat. total Productive Farinha 71* (8.0) ^ 31 (10.4) 26 (6.6) 128 (8.1) Fish 77 (8.6) 20 (6.7) 13 (3.3) 110 (6.9) Clean/ clear land 66 (7.4) 4 (1.3) 17 (4.3) 87 (5.5) Hunt 29 (3.2) 13 (4.3) 22 (5.6) 64 (4.0) Plant 24 (2.7) 2 (0.7) 17 (4.3) 43 (2.7) Teach 23 (2.6) 0 0 23 (1.5) Sub-total 290 (32.5) 70 (23.4) 95 (24.2) 455 (28.7) Reproductive Child care 48 (5.4) 13 (4.3) 46 (11.7) 107 (6.8) Cook 83 (9.3) 14 (4.7) 6 (1.5) 103 (6.5) Wash dishes/clth 61 (6.8) 4 (1.3) 21 (5.3) 86 (5.4) Collect firewood 23 (2.6) 1 (0.3) 7 (1.8) 31 (2.0) Care animals 8 (0.9) 6 (2.0) 7 (1.8) 21 (1.3) Collect NTFP 6 (0.7) 0 14 (3.6) 20 (1.3) Saw timber 4 (0.4) 5 (1.7) 7 (1.8) 16 (1.0) Collect water 8 (0.9) 0 3 (0.8) 11 (0.7) Sub-total 241 (27.0) 43 (14.4) 111 (28.2) 394 (25.0) Personal care Rest 107 (12.0) 27 (9.0) 14 (3.6) 148 (9.3) Eat 57 (6.4) 9 (3.0) 14 (3.6) 80 (5.0) Sleep 7 (0.8) 13 (4.3) 6 (1.5) 26 (1.6) Sub-total 171 (19.1) 49 (16.4) 34 (8.7) 254 (16.0) Leisure Meet friends 8 (0.9) 26 (8.7) 11 (2.8) 45 (2.8) Play sports 8 (0.9) 4 (1.3) 5 (1.3) 17 (1.1) Sub-total 16 (1.8) 30 (10.0) 16 (4.1) 62 (3.9) Travel outside community 78 (8.7) 79 (26.4) 78 (19.8) 237 (15.0) 237 (15.0) *First number is the total number of observations. ^Second number in parenthesis is the percentage of observations/community.
121 Table 4 Â– 4: Continued Activity Boa Vista Rio Azul Vai Quem Quer Total all 3 comm. Cat. total Other activity 46 (5.2) 19 (6.4) 36 (9.2) 101 (6.4) 101 (6.4) Whereabouts unknown 58 (6.5) 2 (0.7) 21 (5.3) 81 (5.1) 81 (5.1) 100% 100% 100% 100% 100 % Total observations 893 (56.3) 299 (18.9) 393 (24.8) 1585 The comparison of spot sampling data in the communities of Boa Vista, Rio Azul and Vai Quem Quer, showed some differences between the three communities. Similar to what was shown with the mean hunting fr equency for households in Table 5.3 above, the communities that were further from urban markets (Rio Azul, Vai Quem Quer) hunted slightly more often than in the community of Boa Vista that is closest to the city of MÃ¢ncio Lima. Overall, hunting occupied 4% of adult time in the three communities together, a little less than the average in Boa Vista (3.2%), and slightly more in Rio Azul (4.3%) and Vai Quem Quer (5.6%). In contrast , fishing occurred more frequently in Boa Vista than in the other two communities. A possible explanation for this is that whereas households in both Rio Azul and Vai Quem Qu er line the riverbanks , the community of Boa Vista has an ox-box lake that forms duri ng the dry season right at the feet of the community, capturing a stockpile of fi sh that are easily accessible to fishermen/fisherwomen in the community. In total, adults spent much more time occupied with productive pursuits in Boa Vista (32.5%) than they did in either Vai Quem Quer (24.2%) or Rio Azul (23.4%). Adults also spent a greater percen tage of their time
122 traveling in the two communities that lay fu rthest from urban centers, Rio Azul (26.4%) and Vai Quem Quer (19.8%), than did adults in Bo a Vista (8.7%). Hunting Techniques There were four primary techniques that hunters in PDS SÃ£o Salvador used in the pursuit of wild game. Daily, often solita ry, hunting excursions into the forest; pastorar ; hunting with the use of an armadilha ; and extended-stay group hunts comprised the four methods of hunting by residents of the set tlement. The first three mentioned hunting techniques are described here. Group hunts, which normally involved hunters from two or more households in the community, will be explained in the following chapter on hunting between households (interhousehold) in the community. The most common form of hunting was when hunters walked into the forest and returned to their households in the same day. Often an individual from the household would leave for a hunting outing into the forest if the househ old lacked fish or meat. These daily hunting excursions were usually accomplished by a solitary hunter who left in the early morning hours, around 6:30am to 7:30am, and returned to his household in the early or late afternoon. Hunters would often head b ack to their communities soon after killing a large anim al such as a veado ( Mazama americana ) or queixada ( Tayassu pecari ), but would frequently stay in the forest in the attempt to capture additional wild game if they had only killed an embiara6 (small animal) or bird. 6 Animals that are classified as embiara were those th at were hunted for food that were small bodied, and were eaten at one sitting by members of the household. If an embiara animal was seen early in the hunting outing, some hunters would not fire th eir guns at them, preferring to sa ve their few shells of ammunition that they carried with them (normally from five to te n shells) for larger game that they might encounter later in the day. Embiara animals that were hunted in PDS SÃ£o Salvador included small monkeys such as parauacu ( Pithecia monachus ) and zogue-zogue ( Callicebus cupreus ), cutia ( Dasyprocta fuliginosa ), quatipuru ( Sciuridae sp. ), paca ( Agouti paca ), and smaller species of tatu (Dsaypus sp. ). These animals were generally consumed solely by members of the household of the hunter who killed the animal, as there was not sufficient meat from these animals to shar e with other households in the community.
123 Sometimes two, or rarely three, other i ndividuals would enter the forest to hunt together. Almost always these small groups of men and adolescent boys were relatives by blood or marriage. If adol escent boys were part of th ese groups, they would normally not venture far from the sight of the lead hunter. If skilled adult hunters entered the forest together, they usually broke off from each other at some landmark in the forest (a forest trail or a large tree) and then regrouped at some agr eed upon time later before they returned to their communities. Generally hunters had a few select areas, within a few hours walk from their houses that they searched when they walked the fo rest in the pursuit of game. Hunters rotated their visits to these hunting locations, spread ing their kill over a wider geographical area so as to not overhunt any one area, or simply to change their luck. Experienced hunters used the sunlight shining thr ough the forest canopy as their guide during hunting outings. Sometimes an unexpected rainstorm would diso rient a hunter, who usually resorted to seeking shelter from the rain until the downpour passed, until he decided to resume pursuing wild game, or return to his comm unity. Less experienced hunters hunted within the sight of well worn trails in the forest, fe aring that they would become lost if they ventured far from these forest paths. Hunt ers in one community utilized a compass as a guide when hunting within the forest. Arbor eal game (monkeys, birds) was located by sound, whereas ground game was located by sight , including animal tracks and plants broken by animals in the forest. Another common practice in the repert oire of hunting techniques was to pastorar . Pastorar generally meant to lie in wait for a wild animal to pass by where the hunter was stationed. Hunters differe ntiated between a hunt ( caÃ§ar ), which (usually) implied
124 Figure 4-2: Hunter from Rio Azul with Tucano ( Ramphastos vitellinus) and Nambu Galinha ( Tinamus guttatus ). walking through the forest to pursue wild game, and pastorar , which meant to stalk game, normally from a perch in a tree or a hammock slung between trees in the forest. Hunters would sometimes pastorar in, or near, a tree th at was in fruit in the forest, that attracted wild game. Pastorar was also the term used when the residents of the settlement carried their guns with them to the roÃ§ado (land where manioc was plante d). Residents did so either as an opportunistic way to hunt while chiefl y engaged with agricultural work, but more often, to rid their fields of animal pests that were eating the househol dÂ’s crops. They said that they would pastorar in the roÃ§ado if the paca ( Agouti paca ) or cutia ( Dasyprocta fuliginosa ), offenda o roÃ§ado (bothered the roÃ§ado). Hun ting within oneÂ’s agricultural fields has been termed Â“garden huntingÂ” (Linares 1976) and is also a common hunting technique used by indigenous peoples in Amazonia.
125 Usually, though, pastorar occurred close to the house, done in th e late afternoon to early evening after other work obligations had been met. Animals that were frequently pursued in this manner are birds whose habi tat was the forest floor including jacu ( Penelope jacquaou ) and jacamim ( Psophia leucoptera ), or monkeys such as parauacu ( Pithecia monachus ) and zogue-zogue ( Callicebus cupreus ) that climbed in the trees near the houses of the community. Adolescent boys would frequently grab the gun on their return to the house after comple ting their daily agricultural ta sks, and join with another friend or two to pastorar in the forest s near their houses. These young hunters went through an informal apprentices hip, having already passed from the slingshot, to pastorar and accompanying older experienced hunters on hunting excursions, to eventually hunting alone in the forest. A third way that hunters enga ged their prey was with the armadilha . An armadilha was a trap made of the sawed-off barrel of a rifle, with the trigge r being tripped by small animals that commonly burrowed into downed, ro tted logs, or directly into a hole in the forest floor. The two animal s most commonly hunted with the armadilha were the paca ( Agouti paca ) and different species of tatu ( Dasypus sp. ). Both of these animals were nocturnal, and it was not uncommon to hear th e blast of a rifle from oneÂ’s bed in the middle of the night, signifying that one of th ese animals had been shot by an armadilha set in the forest. Armadilhas were commonly placed a few me ters off walking paths in the forest where the tracks of a paca or ta tu had been seen, late in the afternoon, and then checked early the next morning to see if an animal had been shot. Sometimes armadilhas were placed in agricultural fields to kill small anim als that were eating the householdÂ’s crops.
126 Armadilhas were commonly set the day following a rain, when the tracks of the paca and tatu were easily visible in the mud of th e forest floor, announcing their presence close by in the forest. Thus, armadilhas were more frequently used in the rainy season. One hunter in the community of Rio Azul said he only sets an armadilh a after it had rained, where animal tracks are visible, because it was a waste of time to do so otherwise. It was not uncommon for a hunter to set out an arma dilha after he had returned from a day of hunting walking through the fore st. Commonly this was the task of adolescent boys in the household. Sometimes more than one armadilha was set at sunset along a path where the tracks of a paca or tatu had been spotted. Material Realities of the Household and Hunting Wild game was an important source of food, and protein, for the residents of PDS SÃ£o Salvador. This was especially true fo r households in the wet winter months from February into April, known locally as times of mau de rancho (bad for food) when the catch of fish is drastically reduced and agricultural food sources are diminished. Although the majority of hunters in terviewed in this sample stat ed that they did not enjoy hunting, they hunted to provide meat for thei r families. Wild game did hold elevated status in the food preferences of residents of the settlement. For example, one adult resident from the community of ConceiÃ§Ã£o replied that Â“ carne Ã© o melhor rancho Â” (meat is the best food). Frequently when I enquire d what a particular household was having for dinner, a common response if they did not have any meat (or fish) to eat was Â“nothing, just farinha.Â” Farinha was the staple of th e diet of households in PDS SÃ£o Salvador, and no meal was complete without it, but wild game holds special si gnificance within the dietary preferences of the household. Animal meat was served during festas (parties) for the inaugurations of schools, churches, or during other cong regations of households and
127 communities in the area. No one would consid er serving just fish and farinha to their guests. Hunters in the hosting household(s) would often hunt in the days preceding the festa in order to have game meat to serve to their guests. Hunters would also hunt the day preceding a planned trip by household memb ers to urban areas, to provide food for family members to eat during the long jour ney by boat into town. In Chapter 1, a set of hypotheses was stat ed related to the effect of material conditions of the household on the practice of hunting. Frequency of hunting was the dependent variable hypothesized to be de pendent upon the set of material conditions, including number of children living within the household, do mesticated animals raised by the household, and outside sources of income, which were each independent variables. Three different hypotheses were said to accoun t for the material reality of each household sampled in the study as it related to hunti ng by each household. Hypothesis 1 stated that households with lots of child ren hunted more than households with fewer children. This reflects the necessity to provide food for me mbers of the household and one would expect that the more mouths to feed in the h ousehold, the greater the frequency of hunting (measured as hunting outings per month) by household members. In order to test hypothesis 1, it was first necessary to determine how many children resided within each household. Throughout much of Brazil, there is the interesting child rearing phenomena known as criar , meaning Â‘to raiseÂ’, wherein a child lives with, and is cared for, someone else other than his or her biological parents (Scheper-Hughes 1992). These children are frequently, but not alwa ys, biologically relate d to household members where the additional child will be raised a nd live. This frequently occurs when households with older, established parents ra ise the children of their offspring (their
128 grandchildren). The grandparentsÂ’ household us ually has more resources at their disposal than does the parentsÂ’ house hold, and thus, the movement of children from parent to grandparent places less burden on the parent sÂ’ household while also directing more resources to the grandchildren. In addition, these children help their grandparents with daily household tasks that become cumbersome to elder couples in the seringal. Nearly every community in the seringal has househol ds where grandchildren are being raised with their grandparents (Panto ja 2001). Frequently these filhos de criaÃ§Ã£o (children by rearing) (Scheper-Hughes 1992: 104) are registered with the State by their adoptive parents as their own children. This may occu r in locales where grandparent and parent reside within the same community or in co mmunities where they do not live together. In this study, one of the areas we we re concerned with was resource use by households (hunting frequency) as dependent on the material conditions of the household as a whole. It was important to accurately capture how many children resided within each household. Therefore, here, the children of the household included only those children that were permanent residents of th e household (six months or more per year), including both the biological offspring of a man and woman of the household, and the adopted children (criar) as described above . These adopted child ren living within the household of their adoptive parents were not counted as members of their biological parentsÂ’ household. Also of importance to address hypothesi s 1, was to determine what was meant by Â“households with lots of children.Â” Fi gure 4.3 shows the number of children residing within a household by the number of households measured in the study. It shows that the number of children living with in a household ranged from a low of zero to a high of
129 seven. The mean number of children per househ old measured in this study (n = 59) was 2.54 (sd = 1.92) 7. Adding one standard deviation (1.92) to the mean nu mber of children per household (2.54) equals 4.46 children/househol d. Since it is not possible to have less than whole numbers of child ren residing within a househol d, for our purposes here, a household with five or more children was cons idered a household with a lot of children. 7 1313 11 7 3 5 0 2 4 6 8 10 12 14 1234567 Number of ChildrenNumber of Households Figure 4-3: Number of Child ren Living in a Household A students-t test was used to test for statistical significance of hypothesis 1 that households with lots of child ren hunted more than households with fewer children. There was a statistically significant difference, at the 90% level of probabi lity, in the frequency of hunting per month of househol ds with lots of children ( 5 children) as compared with hunting frequency by households with fewer children (n = 15, p = .077; df = 57). LeveneÂ’s test for equality of variances of the two sample populations (households with lots of children, households with fewer chil dren) showed that equa l variances could not be assumed (F = 5.821; p = 0.19; df = 57). 7 The socio-diagnostic report of the settlement completed in 1999 found that 59% of the residents of 79 families surveyed in ten communities (total population = 423) were eighteen years of age of less (CÃ¢mara 2002:19). This gives a mean of 3.28 children per family.
130 Hypothesis 2 stated that households lack ing domesticated animals would hunt more than households that had domesticated animal s. This was based on the substitution effect that domesticated meat (chicken, pigs, cattle ) would have as a replacement for the meat of wild game hunted in the settlement. One young hunter in the community of Rio Azul specifically stated that he was required to hunt to provide for his family because he did not raise animals. Households that did not domesticate animals (n = 7) hunted 3.71 times/month in comparison to the hunting out put of 1.87 times/month of households that did have domesticated animals (n = 52).8 This suggests that the meat from domesticated animals may be a substitute for the meat of wild game. Hunting frequency of households with cattle almost reached st atistical significance in compar ison with households that did not have cattle (n = 30, p = .101; df = 57). Equal variances were assumed between the two sampled populations. The majority of households surveyed ha d domesticated animals (52 = 88%), but there was variability in the number and kind of animals that households owned. Table 4.5 demonstrates that less than half of the households had cattle (29 = 49%) or pigs (19 = 32%), whereas the overwhelming majority of households had domesticated chickens and/or ducks (52 = 88%). In addition, househol ds tended to have small herds of cattle, but large numbers of chickens and/or ducks . The distribution of cattle and pigs per household was highly skewed, reflecting that a few households tended to have a large number of cattle and pigs and the rest had fe wer than five (see Table 4.6.). This was not 8 Inferential statistics (t-test) were not used in the co mparison of mean values of households that did and did not domesticate animals because of the very small sample size of households without domesticated animals (n = 7).
131 the case for chickens and ducks , though, with the di stribution of thes e two animals per household showing a normal population distribution. Table 4-5: Domesticated Animals by Household #Animals/household Households w/ Cattle Households w/ Pigs Households w/ Chickens and/or Ducks 0 30 40 6 1 9 3 0 <5 17 5 6 >5 12 14 46 *29 households had cattle, 19 households had pigs, and 52 households had chickens and/or ducks. Table 4-6: Mean, Standard Deviation of Domesticated Animals/Household Animal Mean SD Skewness Kurtosis Cattle 3.83 9.48 4.41 22.65 Pigs 3.27 6.50 2.31 5.27 Chickens/Ducks 19.63 15.03 .622 -.369 This may reflect the fact that chickens and ducks were consumed within the household, rarely being sold (but frequently bartered amongst nei ghbors in the seringal), yet cattle and pigs were held as a reserve of value, sold in the marketplace, and as such were infrequently consumed by the househol d. The population of the settlement was quite young, without having had many years to acquire and store income, and very poor, and these elements combined with the highe r cost of purchasing cattle and pigs as opposed to chickens or ducks, had limited many householdsÂ’ ability to raise larger domesticated animals. In this study, the twelve households that had larger herds of cattle (> 5 cows) have all lived in settlement for over fifteen consecutive years, and twelve of the fourteen households that had larger herds of pigs (> 5 pigs), had lived in the seringal for ten or more consecutive years. These hous eholds with larger he rds of cattle and pigs also tended to be the wealthiest households in the settlement, largely as a result of their investment in domesticating animals.
132 The third hypothesis related to the material condition of the household stated that households that lacked an outside source of income (wages, retirement checks) would hunt more than households that had an outside source of income. Retirement checks paid individuals one or two minimu m salaries if the retiree wa s over 60 years of age, or physically incapable of working.9 These checks were an important source of income for some elder couples living in the settlement. The assumption was that households with an outside source of income might have money available to purchase meat for the household instead of being required to hunt to obtain meat for the household. Of the 59 households surveyed, 14 had an outside source of income (mainly monthly retirement or disability checks). The lack of an outside source of cash income did not show a statistically significant increase in hunting output in comparison to households that did have receive cash income (n = 45, p = .924, df = 57). Wild game continued to be an important source of food for the residents of the settlement. The data confirmed that a househol dÂ’s material realities significantly affected the hunting output of individual households. Households with five or more children hunted more frequently to feed the member s of the household than did households that had fewer than five children. The productive pursuits of the household increased as the number of consumers in the household in creases (Chayanov 1986 ). Additionally, those households lacking domesticated animals, that did not have a substitute for wild game, hunted with greater frequency than househ olds that did raise domesticated animals. Households that did have domesticated anim als generally had larger numbers of ducks and chickens that were raised for home consum ption than they did of pigs and cattle, used 9 The minimum wage in Brazil at the end of 2003 was $R180/month (about $US62.94).
133 as investments for future needs. House holds within the settlement invested in domesticated animals (specifically cattle) to enhance their future livelihood (Ellis 2000). Households that had the larges t herds of pigs and cattle tended to be households that had lived along the Moa River for one or more decades. Two studies in Acre, including one in th e community of SÃ£o Pedro in PDS SÃ£o Salvador (Salisbury 2002) concluded that sm all land holders were in creasingly turning to cattle production as a form of livelihood (Gom es 2001). This does not bode well for the conservation of Amazonia environments, as in creasing cattle producti on has been linked with increases in deforestation (Hecht a nd Cockburn 1990; Loker 1993; Fearnside 1997). As Loker (1993) wrote, there is an inherent rationality for poor peasant households to turn to cattle production, due to the reserve of value of cattl e (insurance value), the low labor input required in maintenance of ca ttle, and production of simultaneous use value (milk) and market value (beef) of cattle. For poor rural households fr equently restricted in their ability to pay for hired labor, cattl e raising with its lo w labor input and high monetary returns is an attract ive livelihood strategy (Homma et al. 1992 sited in Smith et al. 1995). Furthermore, Wallace (2004) found in his study of rubber tappers in the Chico Mendes Extractive Reserve in eastern Acre that households invested a greater portion of their wealth in cattle as they became wealthier. (It is beyond the scope of this project to determine if indeed cattle production is on the rise within the settlem ent or if increasing wealth translates into an increa se in cattle production.) Whether or not a household member had an outside source of income as a means to purchase meat did not affect a householdÂ’s hun ting output. This may reflect first that there were few opportunities to purchase meat in the settl ement. Small domesticated
134 animals were infrequently sold, usually consumed by the household that raised them. Larger animals were infrequently butchered, selling meat by the kilo, but rather when sold, were usually purchased as a live animal , at a cost prohibitive to the overwhelming majority of households in the area. Wild game was rarely sold in the settlement, and as we shall see in the next chapter, the excha nge of game meat between households was an important social mechanism that joins diffe rent households. Second, the income that residents of PDS SÃ£o Salvador received in the form of retirement checks or as teachers in the settlement schools was minimal, and af ter the purchase of manufactured goods in urban areas that were not produced by rural hou seholds there was little money left over to purchase meat for the household. Generally hous eholds used the prec ious little cash that they acquired to purchase items that they c ould not produce themselves or could not find substitutes for. Household Gender Relationships and Hunting The forest has traditionally been associat ed as a masculine space since the days of the rubber economy, and although th e collection of latex to pr oduce rubber had ceased to define the livelihoods of the residents of PDS SÃ£o Salvador , both men and women continued to think of the forest as a manÂ’s sp ace. Men, who used the forest to hunt wild game, said that the forest holds unexpected dangers including snakes, wild animals, and mythological figures that protected the forest.10 Women remarked that the forest was 10 Some of these mythological figures include the Mapinguari , a Big-foot like creature that is said to protect the animals of the forest (Wagley1976; Smith 1996; Melo 2000). Others include the MÃ£e da Mata (Mother of the Forest), and Pai da Mata (Father of the Forest), who similarly protect the animals and other resources of the forest environment. These creatures are all said to be fero cious, foul smelling, and mysterious in their habitats and ways of being. Many of the older generation (greater than forty years of age) that spent the majority of th eir working years within the rubber system of the days of old in the seringal, believed that these creatures were real an d still existed within the forests, while the younger generation that had greater freedom of movement within, and without, the seringal, and for whom the forest was not the critical space of making a living, yet just one more place in the range of spaces that comprised
135 dark and scary, often difficult to traverse, and that they lacked expe rience with the ways and means necessary to negotiate the forest space. One woman from the community of Boa Vista told me, Â“Ã‰ fea a mulher anda na ma ta.Â” (It is ugly for women to walk in the forest.) Another man in a different community stated, Â“O homem tem serviÃ§o na mata. A mulher na casa.Â” (Men work in the forest . Women in the house.) Both men and women stated that pursing wild game within the fo rest was a tiring chore that women were not equipped to handle. Table 4.7 shows a partial listing of the pr oductive and reproduc tive activities of adult household members in the fifty-nine house holds that were interviewed in this study. Productive household activities are defined as activities that could potentially generate income, or could be a step in the process of generating income, for the household in the seringal. The first three hous ehold activities listed in Ta ble 4.7 Â– hunting, fishing, and producing farinha were consider ed productive activities, as were the activities of care for cattle, care for pigs, the sale of the householdÂ’s marketable products, and daily work as a paid laborer. Although Brazilian law pr ohibits the sale of w ild game for profit, allowing for the use of wild game resources as a source of food for the household, hunted game (and fish) was normally exchanged be tween households within the community, occasionally bartered amongst households, and very infrequently sold in the local markets in the towns of MÃ¢ncio Lima or Cruzeiro do Sul. Reproductive household activities were defined as those activities th at did not have the pot ential to generate income for the household in the seringal, yet were essential in the day-to-day sustenance of household members. The last four item s in Table 4.7Â– cooking, childcare, washing the livelihood systems of the peasantry of the seringal, tended to distrust the belief in such creatures of the forest.
136 clothes and dishes, and care of the householdÂ’s chickens were reproductive household activities. The care of a householdÂ’s domesticated animals served both productive and reproductive functions. Cattle, for example, provided milk for household members and cash income when these animals were slaughtered. Cattle were also a reserve of value for the household in times of need such as a medical emergency. Chickens were usually consumed by the household, but were also infr equently sold or bart ered with neighbors for such things as gasoline for passage into town, manufactured foods purchased in urban areas, or cigarettes and alcohol. Table 4-7: Productive and Reproducti ve Household Activities by Gender Activity Men Women Both Productive Hunting 48* 0 1 Fishing 27 4 27 Farinha 7 0 47 Care cattle 22 1 2 Care pigs 19 0 2 Daily paid labor 34 0 1 Sell products 46 0 7 Reproductive Cooking 5 49 4 Childcare 2 46 4 Wash clothes/dishes 4 52 2 Care chickens 10 30 11 *Numbers indicate the gender of the household member(s) engaged in this activity. Many work activities were gender specifi c, yet would sometimes be accomplished by the Â‘otherÂ’ gender if the n eed arose. This often occu red in households that were female-headed (lacking a husband) where the adult woman took a greater involvement in productive household activities , and in male-headed households lacking a wife where the adult man was forced to invest more time and energy in insuring the householdÂ’s reproductive needs were met. In male-headed households lacking a spouse or other adult
137 female, often some of the reproductive needs of the single man (and the other members of his household if he has small children) such as clothes washing or cooking were met by his married daughters living in ot her households in the community. Historically cross-culturall y, women of the household have been responsible for taking care of the reproductive household activities, while men and women in the household share some of the productive hous ehold activities (Boserup 1970; Dwyer and Bruce 1988; Leach 1994; Slocum et al. 1995; Rocheleau et al. 1996; Schmink 1999; Deere and LeÃ³n 2001). In this regard, PDS SÃ£o Salvador was not exceptional, as men and women shared in the activities of fishing and in the production of farinha (the main source of cash income for the majority of residents in the seringal), while women were overwhelmingly responsible for insuring th at the householdÂ’s reproductive tasks were met, including caring for children, washing clothes/dishes, and c ooking. Predominately male productive activities includ ed hunting and the care of large domesticated animals, including cattle and pigs, while a predominately female productive activity was the care of the householdÂ’s chickens. Men also took the lead in the sale of the householdÂ’s agricultural products, an d predominately men in the settlement were hired as daily laborers ( diÃ¡ristas ).11 Therefore, generally a househol dÂ’s cash income passed through the hands of the men (usually the senior male) of the household before it reaches the rest of the household members. In her study of r ubber tappers in the Ch ico Mendes Extractive Reserve in eastern Acre, Campbell (1996) found similar results, with men in control of 11 A diÃ¡rista was usually hired by his neighbor in the community on a daily basis to clear an area of primary forest or capoeira (fallowed agricultural land) for plan ting. These informal contracts generally ran for less than two consecutive weeks. Although women were not hired for such work in the settlement, the wife of the husband of the contracting household was usually required to provide a cooked meal as part of the daily payment to the diÃ¡rista.
138 the householdÂ’s finances and the distributi on of the householdÂ’s cas h income. Within PDS SÃ£o Salvador, frequently wives complained that their husbands journeyed to town to sell farinha, and ended up dri nking some of the householdÂ’s pr ofits that could be better spent on needed household items, or spent money on their girlfriends (imagined or real) in town. The spot sampling data analyzed by gender (Table 4.8) generated many of the same conclusions raised above, that men concen trated in productive pursuits, while women were responsible for making sure that the hou seholdÂ’s reproductive n eeds were met. In this sample, men spent 40% of their time engaged with productive activities, while women were occupied with productive pursu its only 14.5% of their time. Hunting was overwhelmingly a masculine activity with men involved with this activity 7.1% of their time, and women negligibly participated in hunting outings (0.1%). A Pearson correlation indicated that the association between hunting and gender is highly significant (r = .174, p .01). A hunter stated, Â“ O homem Ã© chefe da famÃlia. Tem que dar alimentaÃ§Ã£o a famÃlia .Â” (A man is the head of the family. He has to provide food for the family.) One productive activity in whic h women were important actors was the production of farinha, spending 5.8% of their time dedicated to this activity. Women also fished about one-third of the time (3.9%) that men (9.3%) did. The only productive activity in which women invested more of their time than did men was teaching, with menÂ’s participation in this activity negligible (0.1% of their time). The amount of time adult men and women in the household focused in reproductive work varied between the genders even more than that of productive work. Women spent fully 40% of their time with reproductive work and men only 13.5% of their time. Time
139 devoted to personal care and leisure was about the same for men (16.1%, 4.3%) and women (15.9%, 3.5%). Interestingly, the data showed that women traveled a slightly Table 4-8: Spot Sampling of Household Activities by Gender Activity Men Women Item total Category total Productive Farinha 88 (9.9) 40 (5.8) 128 (8.1) Fish 83 (9.3) 27 (3.9) 110 (6.9) Clean/ clear land 84 (9.4) 3 (0.4) 87 (5.5) Hunt 63 (7.1) 1 (0.1) 64 (4.0) Plant 36 (4.0) 7 (1.0) 43 (2.7) Teach 1 (0.1) 22 (3.2) 23 (1.5) Sub-total 355 (40.0) 100 (14.5) 455 (28.7) Reproductive Child care 16 (1.8) 91 (13.2) 107 (6.8) Cook 16 (1.8) 87 (12.6) 103 (6.5) Wash (dishes/clothes) 4 (0.4) 82 (11.8) 86 (5.4) Collect Firewood 30 (3.4) 1 (0.1) 31 (2.0) Care Animals 17 (1.9) 4 (0.6) 21 (1.3) Collect NTFP 15 (1.7) 5 (0.7) 20 (1.3) Saw timber 15 (1.7) 1 (0.1) 16 (1.0) Collect water 8 (0.9) 3 (0.4) 11 (0.7) Sub-total 121 (13.5) 274 (40.0) 392 (25.0) Personal care Rest 83 (9.3) 65 (9.4) 148 (9.3) Eat 47 (5.3) 33 (4.8) 80 (5.0) Sleep 14 (1.6) 12 (1.7) 26 (1.6) Sub-total 144 (16.1) 110 (15.9) 254 (16.0) Leisure Meet friends 22 (2.5) 23 (3.3) 45 (2.8) Play sports 16 (1.8) 1 (0.1) 17 (1.1) Sub-total 38 (4.3) 24 (3.5) 62 (3.9) Travel 126 (14.1) 111 (16.0) 237 (15.0) 237 (15.0) Other activity 61 (6.8) 40 (5.8) 101 (6.4) 101 (6.4) Whereabouts unknown 48 (5.4) 33 (4.8) 81 (5.1) 81 (5.1) 100% 100% 100% 100% Total observations 893 (56.3) 692 (43.7) 1585 (100) greater percentage of their time (16.0%) than did men (14.1%). This was unexpected as men had more opportunities to travel than di d women, including the pursuit of wild game
140 in the forest, working as day laborers on othe r land owners propertie s, and the selling of the householdÂ’s agricultural products in urban areas all of which women seldom participated in (see Table 4.7).12 Although hunting has been traditionally vi ewed as a male resource procuring strategy (Siskind 1973; Hurtado et al. 1985; Kaplan and Hi ll 1985; Jochim 1988; Stearman 1989, 1991; Robinson and Bennett 2000; Stedman and Heberlein 2001), hunting practice and behavior is best viewed as a nego tiation of the productive and reproductive rights and responsibilities of household members. The hunting of wild game is one of the varied ac tivities that rural households engage in to sustain their families, and is dependent on the collectiv e livelihood strategies of all household members. The men of the household would not have time to hunt if their reproductive needs were not met by the women of the house hold, and the decision to hunt or not to hunt is frequently based on the food resource s accrued by all members of the familyboth men and women. As one woman in the community of Rio Azul told me, each person in the household does their part to help maintain the family: men hunt and clear the forest to plant, and women fish. One male hunter stat ed, Â“When there are no fish [fishing being a shared activity between men and women in th e seringal], one must hunt.Â” As Table 4.7 shows, women were important contributors in insuring the household had a steady supply 12 The fact that spot sampling data collection covered all six months of the wet season (November into April), but only three months of the dry season (August into October), may have accounted for the lower travel percentage of men in comparison to women in the study. The dry season offered many more opportunities for men to travel in the pursuit of productive household activities than were available for women. Therefore, the data analyzed here probably underestimated the percentage of time men traveled, and overestimated the percentage of time women traveled. In addition, women in the settlement frequently traveled with their younger children to urban areas du ring the rainy season (when schools in the seringal were not in session), and stayed with relatives for seve ral weeks at a time. In part, this was a mechanism to lessen the number of mouths to feed during the wet, winter months when food resources were more difficult to capture in the seringal.
141 of food based on their work in the production of farinha, the catching of fish, and the care of chickens which were generally raised for home consumption. Though women of the household did not partake directly in the physical act of hunting, their productive and reproductive work activities with in the household livelihood syst em were essential to the overall hunting strategy of the household. Gender Hypotheses and Hunting Practice Hypothesis 1 stated that male-headed households would hunt more often than female headed households. Hunting was defined as a masculine activity in the seringal, one in which women infrequently partook (see Table 4.7). As such, hunting was a culturally coded activity for me n in the household and wider community. Therefore, in male-headed households we would expect to find a greater frequenc y of hunting than in female-headed households that 1) might lack an adult male, and 2) where hunting was not an fundamental part of the gender defi ned activities of household members. A total of 54 households sampled were male-headed, with 5 households having a female head. There was a difference in the mean frequency of hunting outings per month between the two groups (2.22 for ma le-headed .60 for female-headed).13 This is suggestive of differences in hunting output re sulting from which gender is the head of household. As hunting was a male activity in the settlement, we would expect that households with a male head were inclined to hunt with greater frequency than in households where the tradition of hunting was not part of the repe rtoire of livelihood activities of the household head (i.e. female headed households). 13 Inferential statistics (t-test) were not used in the comparison of mean values of male-headed and femaleheaded households because of the very small sample size of female-headed households (n = 5).
142 The second hypothesis related to the gende r dynamics within the household stated that households with adolescent boys and/or adult men would hunt more often than in households lacking adolescent boys and/or adult men. Part of the soci alization process of growing into manhood in the peas ant households in PDS SÃ£o Salv ador required that boys learn how to hunt. The elder males in the household (fathers and older brothers) took on the role of tutelage of the younger males, teaching them the ways and means of the animals of the seringal and how to stalk a nd hunt them in the forest. One man who did not hunt stated that when his young sons grew older, he would resume hunting to teach them this important skill. Women also desi red, as partners, men th at were successful hunters. Lau, a prolific hunter in the co mmunity of Rio Azul, said women knew a good hunter would always be able to sustain (feed) them. A comparison of the mean values of house holds with adolescent boys (n = 12) and those lacking adolescent boys (n = 47) proved to be slightly significa nt at the 90% level (p .099, df = 57). LeveneÂ’s test showed th at equal variances could not be assumed between the two groups (p .020). Finally, hypothesis 6 stated that households where wome n did not fish would hunt more than in households wher e women do fish. This partially reflected the substitute effect of fish replacing wild game as a s ource of food and protein for the household. In addition, the gender responsibilities of hous ehold members came into play as the productive activities of wome n living within the household affected the practice of hunting by household members. Women fished in thirty-four of the fift y-nine households surveyed. There was, though, no statistical difference in the number of times per month household members
143 hunted, with respect to whether or not wome n of the household engaged in fishing (n = 34, p = .674, df = 57). Equal variances were assumed between households where women did fish and those where wo men did not fish (F = .269, p .606, df = 57). The gender dynamics within a household played an important role in the householdÂ’s hunting output. Hunting was an activity that is hi ghly gender stratified, reserved for men, although women did play a ro le in the practice of the hunt which will be explored in greater depth in the following chapter. Hunting was part of the productive household activities that men sp ecialized in. Women were res ponsible for maintaining the reproductive needs of household members while also participating in the productive output of the household. MenÂ’s reproductiv e needs were satisfied by women of the household allowing them the opportu nity to hunt wild game. Beyond purely the work regimes of househol d members, the performance of the hunt was wrapped up in the process of becoming a man within the household and community. Men wished to teach their sons the practice of hunting as their fathers had earlier taught them how to hunt. Men were seen as the providers for household members (even though as we see, from the data above, women were important parts of the productive process of the hous ehold) and hunting prowess wa s an important method of demonstrating this. The forest, where most hunting activity o ccurred in the seringal, was also a highly gendered space. This was the environment of men as opposed to the house that was the domain of women. Linguistic devices were used by residents of the settlement that mark the forest as a space of hidden dangers, includ ing snakes and mythological creatures that
144 only men were capable of confronting. Both men and women said that hunting within the forest environment was manÂ’s work. Male-headed households did hunt more often than did female-headed households, although the findings of difference in hunting fr equency were at best suggestive due to the small sample size of the female-headed household group. This is not surprising, as hunting activity was a culturally coded behavi or for men in the community and thus, we would expect that households lacking a senior male would not attach similar importance to hunting as households that were male-head ed. Further, households that did have adolescent, hunting-age boys hunted at a significantly greater frequency than households that did not have adolescent boys living in the household. Hunting was a key process of bringing adolescent boys into manhood, and a hous eholdÂ’s hunting output increased with the presence of adolescent boys in the household. Wh ether or not women of the household participated in fish ing did not prove to influe nce a householdÂ’s hunting output. Emotional Commitment to Hunting A householdÂ’s emotional commitment to hunting is a measure of the householdÂ’s satisfaction with hunting as an integral part of the livelihood strategy for the household. Whereas some hunters begrudgingly partook in the practice of hunting due to necessity or because of cultural or social obligations, for other households, hunting was more than merely a part of the overall subsistence strategy of the household. The emotional commitment to hunting measures the householdÂ’s enjoyment with hunting and its willingness to continue with this practice ev en if the household has available alternative sources of meat and protein for household members. For some individuals, hunting was a form of relaxation or a means to escape to the solitude of the forest, leaving behind oneÂ’s daily pressures and problems back in the
145 community. As mentioned above, one hunter stated that hunting allowed him the opportunity to escape the Â“noi seÂ” of the community. Â“ A mata ajuda a gente Â” (The forest helps people) says Manuel, and Â“sozinho na mata, eu passo o dia todo sem que sentir riava.Â” (I spend the entire day in the forest without getting angry.) One hunter in the community of Vai Quem Quer would routinely grab his gun and wander into the forest in the late afternoon hours just prio r to sunset after spending th e day preparing farinha. He said that hunting was his break ( descanso ) after working the entire day. Adolescent boys would also frequently pastorar in the forests close to their houses with their friends, after finishing an exhausting day of agricultural work. As the overwhelming majority of househol ds did not have the funds to pay for labor, many daily activities in the settlemen t required that both men and women work together throughout the day. The preparation of farinha, which almost every household in the settlement partook in, was one such activity that husbands and wives engaged in together throughout the work day.14 The practice of hunting allowed both husband and wife some freedom from each other. Hunt ers throughout the world express the sentiment that hunting offers an opportunity for men to escape their wives (Fine 2000). While their husbands were off in the woods hunting, of ten it was an opportunity for wives across households to socialize. When hunters returned to their households late in the day, their wives resumed their work activities, includi ng preparing the wild ga me that was brought back to the household by their husbands, and fixing a quick snack (coffee with farinha) for their tired hunterhusbands. 14 In this study, 54 of the 59 households interviewed (92%) produced farinha. Of these 54 households, in 47 of them (87%), men and women of the household worked together. See Table 4.7.
146 Other hunters said that they liked to hunt because they enjoyed walking within the forest. For them, hunting was (additionally) a form of recreation. These were generally hunters that were born and raised in the seringal ( criado na mata ) who viewed agricultural activity and the raising of livestock as Â‘workÂ’, and hunting as Â‘pleasureÂ’. One hunter mentioned the sensaÃ§Ã£o (sensation) that he got when he was hunting in the forest. Many of the newer male migrants to the se ringal did not hunt, while some replied that they hunted purely out of necessity to provide food for th eir families. The emotional commitment to hunting was m easured with a scale of a series of eight statements scored with a 3-point Like rt-scale Â– 1) never/no, 2) sometimes, and 3) always/yes. All adult members of the househol d together were read each of the eight statements listed below. Where there wa s disagreement between household members on a particular question, the answer was scored with the middle value in the Likert-scale. The eight statements rated by all adult memb ers of the households sampled in the study were: 1. We like to hunt. 2. Hunting is easy. 3. Hunting is not a waste of time even if we return home without killing any wild animals. 4. We like to hunt more than do any other work activity. 5. We will continue to hunt even if we do not need the meat from the hunt to feed our household. 6. We prefer to hunt rath er than purchase meat. 7. Hunting does not interfere with our other work responsibilities. 8. We hunt to socialize with our sons, friends, and neighbors. A reliability analysis was used to dete rmine if the scale used to measure the emotional commitment to hunting was intern ally consistent, in fact measuring the construct emotional commitment to hunting. CronbachÂ’s alpha coefficient is commonly used to measure the internal c onsistency of a scale. The clos er the alpha coefficient is to
147 1.0, the more reliable the scale. The Cronbach Â’s alpha coefficient for the eight questions listed above in the scale, emotiona l commitment to hunting, was .5416. During questioning I noticed th at the majority of respons es to statement 6 listed above, appeared to base their answers on the desire to have money (to purchase meat), rather than on the intent of the question to measure if hunting activ ity would continue if they did in fact have the means to purchase meat for the household. It seemed that this question was not measuring an emotional commitment to hunting, but the desire for money. CronbachÂ’s alpha coefficient showed that by removing question 6 from the scale, the overall alpha coefficient for the scale in creased to .6087. Therefore, question 6 was removed in the calculation of a householdÂ’s emotional commitment to hunting. A PearsonÂ’s correlation (r) was used to de termine the strength of the relationship between the independent variable, emotiona l commitment of hunting, and the dependent variable, frequency of hunting. These are not ca usal relationships, but merely indicate an association between variables. Pearson correlations take on va lues ranging from -1 to +1 with the size of the absolute value of the re lationship (ignoring the sign before the value) indicating how strong the relationship is between the two variables in question. A perfect correlation gives a value of -1 or +1, and a value of 0 indicates no relationship between the variables. PearsonÂ’s r for the two variab les measured here, emotional commitment to hunting with the frequency of hunting, wa s statistically si gnificant (r = .435, p .001) showing a positive relationship between the variab les. This indicates a strong correlation between these two variables. Seasonality and Hunting Although hunting was an activity that occurs throughout th e year, residents of PDS SÃ£o Salvador said it peaked during the wet m onths of January through April when fishing
148 resources were harder to obtain due to the rising rivers that allowed fish to hide in the flooded forests, and the swift moving currents th at hindered the successful catch of fish. The rivers in the seringal in winter were al so cloudy, filled with soil that had eroded from the riverbanks and therefore, fish were difficult to see and catch. Hunters stated that the wet forest floor during the rainy season allowed them to stalk their prey without drawi ng undue attention to themselves and thus, scaring away the animals they were hunting. They said that the leaves on the dry forest floor of summer crackled underneath their feet, alerting animals to their presence in the forest. Animal tracks were also visible in the mud of the forest floor during the rainy season. Several hunters mentioned that the trees and palms th at fruited during wint er months provided a food source for animals, a meeting place for an imals in the forest, and thus, serve das a focal point for hunting activity. One member of the community of Rio Az ul stated that it was necessary to conserve these fruiting trees and palms so as to have a food source for the animals that inhabited the forest. Furthe rmore, agricultural activity slowed drastically during the rainy season which nor mally ran from the end of November to the beginning of May, affording hunters more freedom to pur sue prey in the forested areas of the seringal. Residents also said that th e months of February into April were mau de rancho (bad for food), as fishing was poor, and agri cultural food production was low during this period, meaning that hunters were often rec onciled to pick up their gun and walk the forest ( dar uma volta ) in the hopes of killing wild game to provide food for their families. As Maria of Rio Azul said, Â“Nosso mercado Ã© a mata.Â” (Our market is the forest.) Table 4.9 shows spot sampling data on hous ehold activities as a function of the weather Â– if it was ra ining or not when the observation was recorded. A total of 114
149 (7.2%) samples were recorded when it was raining, and 1471 (92.8%) on clear days when it was not raining. Several pa tterns emerged from the data . First, productive activities were drastically reduced when it was raini ng in comparison to when there were clear skies. The percentage of time devoted to productive activities when it was raining (11.4%) was less than one-half of the time sp ent in productive activities when it was not raining (30.2%). Fishing and agricultural work (planting and land preparation) were all substantially lower when it rained (0%, 1.8%) than when it was dry (7.5%, 6.9%). Hunting activity was also lower when it was raining (2.6% compared to 4.1% when dry), but not by as great a percentage drop as o ccurred with other produc tive activities. The production of farinha, though, only dropped slightly when it rained (7.0% compared to 8.2% when dry). Farinha was the anchor of income for most househol ds in the settlement during the summer months, and the most im portant food source during the wet months when supplies of food were difficult to come by in the seringal. Second, the amount of time devoted to re productive activities remained fairly constant regardless of the weather, except for childcare which increased substantially from 6.4% to 11.4% of adult daily activity wh en it was raining. This was largely a function of the school year, which had vacation during two to three months of the rainy season (dependent on the communityÂ’s school ca lendar year); thus, parents invested more time in childcare when their ch ildren were not in school. Third, as may be expected, adults spent much more time resting when it was raining (17.5% = tw ice as much time), than when it was not raining (8.7%). Finally, travel outside of the community was slightly increased when it was dry (15.1%) as opposed to when it was wet (13.2% ). One woman in the community of Vai
150 Quem Quer mentioned that she did not vent ure outside the community during the rainy season because travel into, and out of, the area was difficult and not worth the effort. Many people in the seringal escaped to urban areas during the wet wi nter months when productive activities slowed dramatically, spending weeks up to months visiting with relatives that lived in MÃ¢ncio Lima and Cruzeiro do Sul. Summer travel, though, was based on the householdÂ’s produc tive activities: the house holdÂ’s labor requirements (primarily agricultural), and the sale of household products (farinha) in urban areas. Table 4-9: Household Activities as a Function of Rain or Sun Activity Rain Dry Item total Category total Productive Farinha 8 (7.0) 120 (8.2) 126 (8.1) Fish 0 110 (7.5) 110 (6.9) Clean/ clear land 2 (1.8) 85 (5.8) 87 (5.5) Hunt 3 (2.6) 61 (4.1) 62 (4.0) Plant 0 43 (2.9) 43 (2.7) Teach 1 (0.9) 22 (1.5) 23 (1.5) Sub-total 13 (11.4) 444 (30.2) 455 (28.7) Reproductive Child care 13 (11.4) 94 (6.4) 107 (6.8) Cook 7 (6.1) 96 (6.5) 103 (6.5) Wash (dishes/clothes) 5 (4.4) 81 (5.5) 86 (5.4) Collect Firewood 2 (1.8) 29 (2.0) 31 (2.0) Care Animals 3 (2.6) 18 (1.2) 21 (1.3) Collect NTFP 2 (1.8) 18 (1.2) 20 (1.3) Saw timber 0 16 (1.1) 16 (1.0) Collect water 0 11 (0.7) 11 (0.7) Sub-total 32 (28.1) 363 (24.7) 392 (25.0) Personal Care Rest 20 (17.5) 128 (8.7) 148 (9.3) Eat 6 (5.3) 74 (5.0) 80 (5.0) Sleep 11 (9.6) 15 (1.0) 26 (1.6) Sub-total 37 (32.5) 217 (14.8) 254 (16.0) Leisure Meet friends 3 (2.6) 42 (2.9) 45 (2.8) Play sports 1 (0.9) 16 (1.1) 17 (1.1) 4 (3.5) 58 (3.9) 62 (3.9) Travel 15 (13.2) 222 (15.1) 237 (15.0) 237 (15.0)
151 Table 4 Â– 9 Continued Activity Rain Dry Item total Category total Whereabouts unknown 2 (1.8) 79 (5.4) 81 (5.1) 81 (5.1) 100% 100% 100% 100% Total observations 114 (7.2) 1471 (92.8) 1585 (100) Modeling Intra-household Eff ects of Hunting Frequency Intra-household effects on a householdÂ’s hunting frequency were theorized in Chapter 1 to be the result of three sets of independent variables: 1) material (lots of children, domesticated animals, outside sour ce of cash income), 2) gender (male or female-headed household, hunting age boys/adu lt men in the household, women fish) and, 3) the householdÂ’s emotional commi tment to hunting. After normalizing the dependent variable, frequency of hunting, usi ng the log base10 transf ormation, T-tests of the statistical significance for the mean valu es of each of the independent material and gender variables were computed. These resu lts are summarized in Table 4.10 below. Table 4-10: T-test for Significance of In tra-household Material & Gender Variables Independent Variable p-value Material Lots of children .077* Domesticated animals ---^ Outside income .924 Gender Male-headed ---^ Hunting age boys .099* Women fish .674 *Significant at 0.10. ^T-test not computed. Statistical significance at p 0.10 was found for the independent variables Â“lots of childrenÂ”, Â“domesticated animalsÂ”, and Â“hun ting age boysÂ” in the household. A Pearson correlation found a high degree of statistical significance of an association of the householdÂ’s emotional commitment to hunting with the frequency of hunting (r = .435, p .001). The data indicate that the most salient variable influencing a householdÂ’s
152 frequency of hunting was a householdÂ’s Â“emoti onal commitment to huntingÂ” Â– the desire to continue with hunting as an important part of the householdÂ’ s livelihood strategy. Second, the importance of game meat to the hou seholdÂ’s food security influenced hunting output by the household as those households with lots of children (increased consumers within the household) and those lacking domesticated animals (minus substitutes for wild game), hunted more often than households with fewer children or those that domesticated livestock. Third, gender appeared not to be as important a variable in the determination of a householdÂ’s hunting output. Here we use multivariate linear regression to model the combined effects the independent variables categorized as mate rial, gender, and emotional commitment to hunting have on the dependent variable, fr equency of hunting. First, a Pearson correlation was run to uncover if any of the independent vari ables were highly associated with each other. Multicollinearity occurs when two independent variables are highly associated with one another and therefore, leaving both variables in the regression model usually does not add much explanatory power to the model. Only the two independent variables, Â“lots of childrenÂ” and Â“hunting age boysÂ” indicated a high (positive) association with one another (r = .661, p .001). T-tests run on mean values of these two variables showed statistical significance at p .10; therefore, re gression models were tested with these variables in separate models. Multiple regression was used to model the re lationship of the independent variables with the dependent variable, frequency of hun ting. The material variables were entered in the first model. The gender variables and the variable, Â“emotional commitment to huntingÂ”, were run in Model 2. Finally, Model 3 includes only those independent
153 variables that proved to be st atistically significant in either Model 1 or Model 2. Table 4.11 presents the results of these regression models. Table 4-11: Multivariate Regression M odeling of Intra-household Influence on Frequency of Hunting Model 1 Model 2 Model 3 Variable -coefficients -coefficients -coefficients Lots of children .178 Domesticated -.249* -.122 animals Outside income -.005 Male-headed .237* .219* Hunting age .300** .313** boys Women fish .077 Emotional .462*** .429*** commitment hunting Observations 59 59 59 Adjusted rÂ² .031 .241*** .249*** *Significant at p 0.10; **significant at p 0.05; ***significant at p 0.01. Model 1 incorporated the material hypot heses H1 Â– H3. The model indicated a very weak fit in explaining the dependent va riable (adjusted rÂ² = .031, F = 1.611). Of the three independent material variables, Â“domes ticated animalsÂ” was weakly significant (p 0.10). The coefficient of -.249 for the Â“domesticat ed animalsÂ” variable indicated that as a householdÂ’s domesticated animals decreased by 24.9%, hunting frequency increased by a factor of 1. Model 2 tested the gender variables (H 4 Â– H6) combined with the variable Â“emotional commitment to huntingÂ” (H7). Th e model proved to be highly statistically significant with the independent variables of the model explai ning a little of 24% of the variance in the dependent variable, Â“freque ncy of hunting.Â” The exploratory Model 3 was only a slightly better f it than Model 2. Of the thre e regression models, Model 3, incorporating of the independent variables that achieved statistical significance in Model
154 1 (Â“domesticated animalsÂ”) and Model 2 (Â“ male-headed householdÂ”, Â“hunting age boysÂ”, and Â“emotional commitment to huntingÂ”) indica ted the most explanatory power of the dependent variable frequency of hunting (rÂ² = .249). The multiple regression models indicated that the most important variables influencing a householdÂ’s hunting output were those of the Â“emotional commitment to huntingÂ” (cultural) and Â“hunting age boysÂ” (g ender) in the household. The variable Â“male-headed householdsÂ” (gender) also was sl ightly significant, but due to the small sample size in this study of households that were not male-headed (i.e. female headed households, n = 5), this result can only be s uggestive of its influence on the dependent variable Â“frequency of hunting.Â” The material variable Â“lot s of childrenÂ” that reached statistical significance in a co mparison of mean values with a t-test, did not achieve significance when modeling this variable comb ined with the other material variables (Model 1). In addition, the variable Â“domestic ated animalsÂ” that was also statistically significant with a t-test and in Model 1, was not significant in the regression Model 3. The regression coefficients of the rest of the intra-household variab les with the exception of the variable Â“women fish,Â” moved in the direction that was expected.15 Apparently, fish was not a substitute for game meat in PDS SÃ£o Salvador. Taken as a whole, the material variables had much less explanatory power than did cultural and gender variables in influenc ing a householdÂ’s hunting frequency. This suggests that hunting in PDS SÃ£o Salvador is primarily motivated by reasons other than securing food for the household. Hunting output, then, reflects a livelihood choice of the 15 The variables Â“lots of childrenÂ” and Â“male-headedÂ” household showed a positive influence of hunting output, and the variables Â“domesticat ed animalsÂ” and Â“outside incom eÂ” showed a negative influence on hunting output.
155 household, yet is also dependent on the importan ce of this activity as a marker of male identity in the settlement. Conclusion In this chapter we examined the influen ces of intra-household relationships on the output of hunting. A variety of methods incl uding in depth house hold questionnaires, productive/reproductive household surveys of adult activity, spot sampling of adult behavior, hypothesis testing, and multiple regression were used to explore these relationships and how they affected a house holdÂ’s frequency of hunting. Intra-household independent variables were group ed into three sets: material realities of the household, gender dynamics and relationships within the household, and a householdÂ’s emotional commitment to hunting. We also briefly inve stigated the seasonal nature of hunting and other productive and reproductiv e activities of adults w ithin the settlement. The first part of this chapter focused on understanding the material needs of households in PDS SÃ£o Salvador. The majority of the households sampled in this study were hunting households, although it is significant that ove r one-half of these hunting households stated that they did not enjoy hunting. Hunting output has waned for most households as livelihood strategi es have rapidly shifted in th e last ten to fifteen years from rubber tapping to small-scale agriculture centered on the producti on of farinha. The extraction of rubber necessitated adults (mostly men) walking within the forest to tap rubber, placing them in close contact with the animals of the seringal, whereas agricultural work entails clearing forestland for crop production, and thus, the clearing of animal habitat in the seringal. Therefore, one of the results of deforestation for agricultural production is less daily contact with the anim als of the seringal, which
156 lessens hunting opportunity. Furt hermore, many hunters were unwilling to sacrifice their agricultural output for th e prospect of hunting. As much of adult activity in the settle ment is based on fulfilling the consumptive needs of household members, the material realit ies of the household played a critical role in a householdÂ’s hunting output. The more mout hs to feed in the household necessitated an increase in hunting output by the household. Households that had five or more children hunted with greater frequency than hous eholds that had fewer than five children. Most hunters indeed stated that they hunted pr imarily to supply food for their families. The domestication of livestock was shown to be a potential meat substitute for wild game for households in the settlement. T hose households that did not raise animals hunted with greater frequency than househol ds that had domesticated animals. One hunter even specifically said that he was forced to hunt because his household did not raise animals. Smaller animals including chickens and ducks were raised for home consumption whereas larger animals such as cattle and pigs were domesticated as a reserve of value for house holds. Cattle did, though, also provide use value for households, seen in the production of milk for home consumption. As stated above, increasingly small landholde rs are turning to ca ttle production in Amazonia. Salisbury (2002) found that this is also occurring within one of the ten communities of the settlement PDS SÃ£o Salvad or. Generally, cattle implies deforestation in Amazonian environments that are ill su ited for long-term cattle production. The prospects of increasing cattle production may not fit within the management framework of a sustainable development settlement as was envisioned by INCRA and NGOs in the creation of PDS SÃ£o Salvador. In spite of this, households in this study who raised cattle
157 were less likely to hunt than households that did not have cattle. The paradox is that while plant diversity de creases with conversion of fores tland to pasture, wild animal resources in the area may be protected ( not hunted as frequently) as households domesticate cattle. It is un likely though, that increasing cattl e production for residents in the area is a strategy th at resource managers are likely to (or should) pursue. Next, gender relationships within the hous ehold were investigated. Hunting was found to be overwhelmingly a culturally-coded male activity within the seringal. The forest environment where most hunting ac tivity occurred in the settlement had traditionally (since the rubber era) been reserv ed as a space for men. Men were said to work in the forest; women in the house. Both men and women spoke of the dangers that the forest held, including snakes and mythol ogical creatures that women were supposedly incapable of confronting. For their part, wo men seemed content to let their men wander around the forest in the pursuit of wild ga me, an often physically draining all-day activity. An integral aspect of the gender-defin ed roles of men in the household and community was wrapped into hunting activity. Part of the socialization process of becoming a man meant that adolescent boys were taught the ways and means of hunting by their elder kin in the h ousehold and community. Hypot hesis testing showed that households that had adolescent boys hunted with greater frequency th an those households that did not have adolescent re sidents. Boys progressed fr om firing slingshots at birds near their houses, to pastorar close to their houses with thei r elder brothers or adolescent boys from other households in the community, to accompanying their fathers, uncles or older brothers on hunting outings, to finally hunting alone in the fo rest by their late
158 teenage years. This progre ssion coincided with the progres s of a boy becoming a man in the eyes of the community, as adolescent boys became marriage elig ible around the time that they began to hunt on their own. Men specialized in produc tive pursuits and women t ook care of the householdÂ’s reproductive needs while also participating in productive household activities. Men took the lead in hunting, the care of the householdÂ’s large animal s (cattle, pigs), the sale of household products, and were hired as laborer s in the seringal. Women cooked, washed clothes/dishes, cared for child ren, cared for chickens, and also engaged in the productive activities of fishing, and making farinha that was the main income source for the majority of households in the settlement. The spot sampling data showed that men spent 40% of their time in productive activ ities and 14.5% in reproductive activities, whereas for women the numbers were almost the exact opposite: 40% in reproductive work and 13.5% in productive work. The practice of the hunt elevated the stat us of men in the household, as the product of hunting, wild game, held a special place in the food preferences of residents in the settlement. Game meat was the most a ppreciated food in the seringal (Wolff 1999). Residents of PDS SÃ£o Salvador often stated to me if they lacked meat to eat that they had Â“nothingÂ” to eat. Animal meat was required at all festas in the settlement, with men of the hosting households hunting one or two days prior to celebr ations in order to provide meat for their guests. WomenÂ’s work did not afford similar opport unities to acquire the special status that men achieved through the performance of the hunt. Some households hunted because they were more emotionally committed to hunting as an important part of the overal l livelihood strategy for the household. A
159 Pearson correlation showed a high degree of statistical association between the independent variable Â“emotional commitment to huntingÂ” and the dependent variable Â“frequency of huntingÂ”. Many hunters derived satisfaction from hunting in the forested areas of the settlement that allowed them to escape the daily pressures of their households and community. Hunting for these men was mu ch more than a means to provide food for the dinner table or a social obligation related to their ma nhood, but a form of recreation and relaxation. Some hunters stated they would not cease th e practice of hunting even if they had substitute sources of meat available for their family. It is unclear if these households would lessen their hunt ing output if they raised do mestic livestock. Finally, we investigated the seasonal nature of hunting in the settlement. Most hunters replied that they hunted more often in the wet, winter season (late November to early May) than in the dry season. Animal tr acks were visible in the mud of the forest floor in winter, and agricultural activity dram atically slowed in winter affording men more opportunity to hunt. Along the Moa Rive r, several months in winter (February through April) are known as mau de rancho (bad for food), with both fish and agricultural food sour ces declining, requiring men to pursue wild animals to feed household members. In addition, many trees fr uit in winter attracting animals, thus serving as a focal point for hunting activity. The environment played a key role in the yearly cycle of live lihood strategies of households in the settlement. In general, households were concerned with production (farinha) in the summer months, and consump tion in winter. In the study spot sample data, adults spent almost three times as much time devoted to produc tive activities in dry
160 weather as they did when it is raining, yet more time in reproductive activities when it was wet outside (28.1%) than they did wh en there were clear skies (24.7%). Hunting output by households in PDS SÃ£ o Salvador reflected multiple intrahousehold influences. The importance of wild game as a food source for rural households cannot be overstate d, yet as we see above, ther e were other factors that critically affected hun ting behavior by peasant househol ds. Hunting and gender roles and relationships were interwoven within the hous ehold and as such, reflect a negotiation by members of the household. Yes, men in th e aggregate were the hunters, but womenÂ’s work in insuring that the reproductive needs of household members were met was critical in the opportunity men have to hunt. Wome n were also import ant producers of food resources for the household. Men and women of the settlement sa id that men hunted while women fished. One male resident from the community of Rio Azul stated specifically that he did not have to hunt because his wife fished. He, himself, concentrated in the production of farinha and other agricultural pursuits because his wife participated in supplying other f ood resources for the households. Some men hunt because it was an enjoyabl e activity for them. These are the men that had hunting in their blood16 ( jeito ), satisfied with their hunti ng lifestyle. It would be interesting to question these men in the futu re to see if their hunting output did change, and if it did, the reasons fo r this change. Maybe increasing agricultural work, or an increase in animal domestication, leads to a decrease in hunting frequency, or conversely, more children leads to increased hunting activity? 16 Hell (1996) reports that European hunters also refer to hunting as being in the blood. Some of these hunters use the metaphor of Â‘black bloodÂ’ related to Â‘hunting feverÂ’.
161 The regression models indicated that cu ltural and gender variables had a much greater effect on a householdÂ’s hunting output than did material variable s. This indicated that the performance of the hunt satisfied ot her desires and preferences of the household than simply capturing the product of the hunt (game meat) to feed the household. Hunting is a cultural practice in the sett lement that brought satisfaction to those individuals that engaged in this activity while also conf erring the status of manhood on these practitioners of hunting. Furthermore, the product of the hunt, game meat, not only tastes good, but has elevated importance in the food preferences of residents of the settlement. Plant foods and fish do not infer th e same dietary status as does animal meat. Therefore, households hunt primarily as a ma tter of choice because th ey desire to do so, because it is a masculine defined activity, a nd because Â“game meat is the best food,Â” rather than out of necessity to feed the members of the family.
162 CHAPTER 5 HUNTING ACROSS HOUSEHOLDS (INTER-HOUSEHOLD) The previous chapter explored hunti ng as a function of intra-household relationships. In this chapter, we will in vestigate how relationships between different households within the community (inter-household) affect the practice of hunting. This will include an examination of group hunti ng by individuals from several different households in the community. Hunting outing data from two communities, Rio Azul and Boa Vista, will be explored to uncover the nature of group hunts by members of different households. We will also look at meat exch ange amongst households and the symbolic and social importance that it hol ds for peasant households. Here we will also test hy potheses raised in Chapter 1 that were posed to explain inter-household hunting behavior. Hypotheses testing will incl ude T-tests of mean values for inter-household variables. Finally, lin ear regression will be used to model the dependent variable, frequency of hunting, ex plained by the inter-household variables analyzed in this chapter. Hunting Across Households In the previous chapter the three hunting techniques that individual hunters use in the settlement were discussed. Here we l ook at the fourth common technique of hunting that hunters employed in their pursuit of wild game in PD S SÃ£o Salvador, that of group hunts which included hunters from severa l households in the community. Hunters in PDS SÃ£o Salvador infrequently used extended, overnight hunting outings lasting several days to about a week. Hunters spoke of walk ing six to ten hours
163 to reach o centro (the center)1, an uninhabited area in the southe rn part of the seringal that had been deemed a hunting reserve by the resi dents of PDS SÃ£o Salvador. Hunters that lived in the northern part of the settleme nt (communities of Ti mbaubÃ¡, ConceiÃ§Ã£o, and SÃ£o Pedro) made these trips to the forested area of the settlement that bordered the Nukini Indigenous Reserve. This area was rife with tension as the boundary between indigenous reserve and settlement was not marked, and occasionally both indigenous and peasant hunters accused the other of treading on th eir territory. This is also an area where outsiders from urban areas (principally MÃ¢nc io Lima) came to hunt. Residents of PDS SÃ£o Salvador had denounced this practice of non-resident hunting to IBAMA, but it was difficult for the authorities to intercept ille gal hunting activity as it was occurring, and few of these illegal hunters were ever prosecuted. Inhabitants of the settlement had also been threatened with retaliation by these i llegal hunters if they were summoned to the offices of IBAMA in Cruzeiro do Sul, resu lting from complaints lodged by settlement residents. Hunters from the settlement indicated that in the past when the population density in the area was not as great, there was little n eed to trek deep into the forest to capture wild game, as a short distance from oneÂ’s house a hunter would inva riably encounter a potential meal for his family. Seu Paolo of Boa Vista, pointing to an area around 100 meters from his front porch, lamented the f act that just a decade ago, bands of queixada would sometimes run through the area, and re sidents of the community would quickly grab their guns to kill their meal for the ne xt few days. Hunters would invariably shoot 1 The term o centro is a generic phrase used in Amazonia indicati ng an area that is in the interior away from the river (as distinct from a beira , the riverside), but not necessarily in the center of anything. In this case, though, when hunters in PDS SÃ£o Salvador spoke of walking to reach o centro they were, in fact, walking to the southern center of the settlement
164 as many animals as they had bullets availabl e, providing a feast for the entire community. A study conducted by Fragoso et al. (2002) of w ildlife and hunting pract ices in the area suggested that some animal species had beco me locally extinct as a result of hunting pressure exerted on wild game by the resident s of PDS SÃ£o Salvador. Trekking to these far away hunting areas widened the range in the seringal where game was harvested and might help replenish stocks of wild animal s close to settlement areas that were being rapidly depleted through excessive hunti ng. Conversely, hunting within a greater geographical range also increased the area where wild game was hunted, which might deplete animal resources throughout a wider area. These extended hunting trips normally occu rred during the rainy season (especially January to early May) when the demands of agricultural work were minimal, and men could afford to leave their families for several days at a time. Also, during the late winter months of February through April, food was harder to procure for households in the seringal, and many trees in the forest were fruitin g, attracting wild animals, thus making hunting an attractive alternative fo r households in the seringal. In the months of February and March in 2004, hunters in the community of Boa Vista told me that they were preparing to make overnight hunting trips be cause there was no food available in the community. Generally several hunters would join for ces on these overnight hunting trips into the forest. Most hunters in a group walked al ong paths that they ar e familiar with to hunting sites where they camped for several days. Hunters from the community of Boa Vista, though, used a compass as a guide duri ng these outings. A base camp was erected by the group, from which these men ventur ed out during the day, returning before
165 nightfall. Hunters generally hunted individua lly during the day, joining for a lunch break, and returned to the base camp as daylight fe ll. Hunters packd farinha to eat, and salt for their kills, during these hunts. Hunters would also sometimes bring an armadilha with them to set at night in the forest when they were sleeping. Some hunters, although it is illegal according to comm unity rules, brought their dogs with them on these hunting outings. Fr equently hunters would capture prey with dogs in instances where they might not have otherwise, as a result of their dogs chasing and confining animals in the forest. Dogs can corner prey such as paca and tatu, preventing them from burrowing into a dow ned log or hole in the ground. In one community in the settlement there was anger fr om one familial group directed at another, with strong accusations of hun ting with dogs levied towards the offending family. A hunter from the community of Girrassol blamed the paucity of animals in the settlement on the use of hunting dogs.2 He said that after the fall of the rubber economy in the area by the early 1990s, households turned to the sale of wild game as the principal means of renda familiar (household income), with hunters pr incipally using their dogs in the seringal. As a result of the success in hun ting prey with dogs, there were fewer animals in the forest in 2005. The hunting of wild life with dogs was a long-standing practice within the seringal (Melo 2000), but was sl owly waning through the education program directed at conserving wildlife in the area, under the auspices of PESACRE, a Brazilian NGO operating within the state of Acre. Group hunts almost always consisted of me n from different households within a community that were related by kin. Often th ese were brothers or brother-in-laws from 2 Other hunters said th at dogs scare away ( espanta ) prey and therefore, they di d not use them when hunting.
166 two or more households. Generally elder me n of the household (over 40 years of age) did not participate in this type of hunting outing. This manner of hunting required the strength and stamina of younger men who were required to carry on their backs all of the material needed to sustain themselves for seve ral days to a week, and then return to their households with their prey on their backs.3 Hunters cut off the heads of large animals captured during these trips, and gutted their prey, leaving the heads and entails in the forest to lighten the load they had to carry back to their households. These hunting trips afford the opportunity for binding friendships amongst men of different households within the community. Often men who traded days of labor in agriculture with one another ( troca dias ) were the same ones who made these extended hunting trips together. Households that trade days of labor with one another were family related. A Pearson correlation s hows that there was a statisti cally significant (positive) relationship between hunting frequency and trad ing labor days with other households in the community (r = .309, p .017). These trips also impor tantly provided food for the families of the hunters for several days duri ng a time when food resources were difficult to attain in the seringal. Generally when hunters returned to th eir communities after several days of overnight hun ting, they brought with them e nough wild game to feed their families (and those of relatives in the comm unity with whom they exchanged meat) for several days. These extended hunting trips were a chan ce for both husband and wife to have individual time separate from their spouse (and the community).4 Wives of the hunters 3 None of the households in the settlement have burros or horses. 4 Male hunters in Michigan also stat ed that an added benefit of hunting was that they were able to escape the women (mainly their wives) in their lives (Fine 2000).
167 would frequently travel to the city while their husbands were hunting in the forest, returning to their homes in the settlement wh en their husbands were scheduled to arrive from the forest. Several hunting households in two commun ities, Boa Vista and Rio Azul, were sampled for seven consecutive (randomly se lected) days per m onth over a consecutive four month period from January 2004 th rough April 2004, to measure their hunting activity Â– hunting frequency, hunting met hod (solo, group, extended over-night), and number and kind of animal kills. Four hunti ng households were followed in Boa Vista, and three hunting households were sampled in Rio Azul. Figure 5.1 shows the number of hunting outings per household by commun ity for this four month period. 0 1 2 3 4 5 6JanFebMarAprJanFebMarAprJanFebMarAprJanFebMarAprJanFebMarAprJanFebMarAprJanFebMarAprNumber of Hunting OutingsHousehold 1 Household 5 Household 4 Household 3 Household 2 Household 6 Household 7B oa Vi sta ( 1-4 ) Rio Azul ( 5-7 ) Figure 5-1: Number of H unting Outings per Month Hunting activity peaked in March for both communities and was lowest in April for both communities. Mid-way through the samp led week of April 2004 in the community of Rio Azul, the community members suddenly caught a large quantity of fish on a daily
168 basis, and subsequently all hunting activity st opped during this period. The fishing take was much greater than the wild game from th e forest that hunters ha d been returning with to their households, so the hunter s joined their wives in pulling fish from the Moa River. Six of the seven hunting households reported th at they hunted more frequently in the winter months than during the dry season. H unting activity overall for this four month, winter period was greater th an the yearly, monthly averag es reported by each of these seven households. This is consistent with wh at has been reported for peasants in the JuruÃ¡ Extractive Reserve in western Acre, wh ere hunting activity also peaks during the rainy season (da Cunha and de Almeida 2002) . One hunter (househol d #2 in Boa Vista) stated that he only hunts in the winter when he is not burdened with agricultural work. Hunting output was different between these hunting households in the two communities of Rio Azul and Boa Vista (see Table 5.1). The three hunting households in Rio Azul hunted more during the four m onth period (34 total hunting outings) and showed a much greater projected frequency of hunting/month than did the four hunting households in Boa Vista (14 hunting outings). Table 5-1: Frequency of Hunti ng per Household by Community Community Household Total Hunting Outings (28 days) Projected Hunting Outings (30 days) Boa Vista 1 3 3.2# 2x 3 3.2 3 5 5.4# 4 3 3.2# (sub-total) 14 3.8 Rio Azul 5 12 12.9 6 8 8.6# 7 15 15# (sub-total) 34 12.2 x Only hunted in winter. # Hunted more in winter.
169 The method of hunting also varied betw een the two communities as shown in Figure 5.2. Hunters in Boa Vista hunted more often with members of other households (8 hunting outings) than they did solo (6 hun ting outings), whereas the opposite was true in Rio Azul (25 solo hunting outings and 9 group hunting outi ngs). Only hunters in Boa Vista made overnight, extended stay hunting trips. Group hunts in both communities and overnight hunting trips in Boa Vista were more frequently made with kin from other households in the community than with nonkin hunting partners (groups hunts = 12 of 17 hunting outings with kin, 70.6%; over-night = 4 of 5 hunting outings with kin, 80%). 1 0 2 3 9 5 11 2 3 3 0 3 3 30 2 4 6 8 10 12 14 161234567HouseholdNumber of Hunting Outings Group Solo Boa Vista ( 1-4 ) Rio Azul ( 5-7 ) Figure 5-2: Method of Hunting per Household by Community Figure 5.3 shows hunting success per hunting household by community over the four month period. The trend was for hunti ng success to decline over the four month period in both communities, with the month of April the l east successful hunting month in both. April though, was also the month with the lowest frequency of hunting activity, so one would expect hunting yields to dec line as hunting activity was declining. The
170 greatest success of hunting kills occurred in the months of Ja nuary and February in both communities. 0 2 4 6 8 10 12 14JanFebMarAprJanFebMarAprJanFebMarAprJanFebMarAprJanFebMarAprJanFebMarAprJanFebMarAprNumber of Animals HuntedHousehold 1 Household 5 Household 4 Household 3 Household 2 Household 6 Household 7Boa Vista ( 1-4 ) Rio Azul ( 5-7 ) Figure 5-3: Hunting Success pe r Household by Community Another trend was that the overwhelming majority of successful hunting outings entailed the kill of small animals ( embiara ) in both communities: Boa Vista = 34 of 43 kills (79.1%); Rio Azul = 53 of 57 kills (93%). See Figure 5.4 below. There was a difference in the type of anim al that was successfully hunted in each community. In the community of Boa Vista, 31 mammals (72.1% of all kills), 11 birds (25.6%), and one reptile (2.3%) were hunte d. The opposite situation occurred in Rio Azul with hunters taking more birds (30 = 52.6% of all kills), than mammals (26 = 45.6%), and one reptile (1.8%).
171 2 15 13 4 15 17 21 0 0 9 0 0 4 00 5 10 15 20 251234567 HouseholdNumber of Animals Hunted Large Animal Embira Boa Vista ( 1-4 ) Rio Azul ( 5-7 ) Figure 5-4: Size of Animals Hunt ed per Household by Community The difference in the types of animals hunted between the two communities may be explained by the difference in the method of hunting. Hunters in Boa Vista that made overnight, extended-stay hunting journeys save d their bullets for larger animals, often refusing to shoot at smaller prey birds that they encountered while away from their households for several days. These extendedstay, hunting journeys were taken with the idea to return to the community with enough wild game to feed the huntersÂ’ households for several days, so using the few bullets th at hunters carried with them hunting birds defeated this hunting agenda. On the ot her hand, hunters in Rio Azul did not take overnight hunting trips, instead making opportuni stic single day hunting trips, shooting at the first animals that they encountered (whate ver the size of the animal). They were more focused on returning with that nightÂ’s meal for the household rather than on capturing several days worth of food.
172 Overall, the hunters in Boa Vista had better success measured by kills/hunting outing (43 kills/14 hunting outings = 3.1 kills/ outing) than did hunters in Rio Azul (57 kills/34 hunting outings = 1.7 kills/outing). Hunters in Boa Vista also took a much greater percentage of larger an imals (9 of 43 kills = 20.9% of all kills) than did hunters in Rio Azul (4 of 57 kills = 7% of all kills). Only two hunting households in this sample successfully hunted large animal s: house #3 in Boa Vista and house #6 in Rio Azul. In addition, over-night, extended stay, group hunt ers brought a much la rger bounty of game meat/hunting outing (31 kills/5 hunting outings = 6.2 kills/outing) than did day hunts (65 kills/43 hunting outings = 1.5 kills/outing). Differences in the success and size of game successfully hunted in the two communities probably were a function of the location of hunting and method of hunting used by hunters from the different communitie s. As mentioned above, hunters in the community of Boa Vista that made overnight, extended-stay hunting tr ips hiked six to ten hours to an uninhabited area of the settlement called o centro an area that had greater stocks of large game than did areas closer to community centers where the majority of hunting occured for hunters in the community of Rio Azul. These multi-day hunting trips afforded hunters the opportunity to come in to contact with larger animals for longer periods of time. This, combined with their intent to capture large animals to provide meat for their families for several days, meant that hunters in Boa Vist a returned to their households with a greater percentage of large animals than did hunters from the community of Rio Azul. In addition, these hunting trips lasting seve ral days were taken with the sole objective of returning to the co mmunity with a large quantity of game meat, and hunters were unencumbered with other work and social obligations. With single day
173 hunting outings, hunters would sometimes l eave their homes in the early morning, returning by mid aftern oon to resume other work obligati ons (agriculture or fishing). Therefore, the amount of time per day devoted to hunting was less in the community of Rio Azul than in Boa Vista over the samp led time period. Table 5.2 shows hunting success by househol d. Household #2 of Boa Vista was the most successful hunting household in the sample based on the number of animals killed/hunting outing (5 kills/out ing). Household #3 in Boa Vista had the most kills (22) of all of the hunting househol ds in the two communities. Household #6 was the most successful hunting household in Rio Azul in te rms of number of animals killed (21) and hunting kills/outing (21 kills/8 outings = 2.6 ki lls/outing). Lote, of household #6, was the acknowledged best hunter in the community; another adult male from another household in the community told me it was rare for him to return from a day of hunting in the forest empty handed. Table 5-2: Hunting Success/Household Community Household No. of Kills No. of Outings No. of Kills/outing Boa Vista 1 2 3 .67 2 15 3 5.0 3 22 5 4.4 4 4 3 1.3 (sub-total) 43 14 2.8 Rio Azul 5 15 12 1.3 6 21 8 2.6 7 21 14 1.5 (sub-total) 57 34 1.8 Meat Exchange Meat exchange was an integral part of the set of social relations that existed between households in the comm unity. This form of exchange extended, and helped maintain, a householdÂ’s social contact with neighbors in the commun ity. This form of
174 non-market exchange is a public spectacle th at is contrasted with the often secret exchange found in the marketplace (Gudeman a nd Rivera 1990). Market transactions that entail the transfer of money for goods and/or services , frequently involve only the buyer and seller, whereas the exchange of ga me meat between rura l households is an open transaction that many diffe rent actors from different ho useholds view and partake in. The ability of a household to give meat to other households signifie s first, prestige to the producer of the wild game (the hunter and his household), second, that social obligations between households have been met by the giving household, and third, that the receiving household has publicly received a Â“giftÂ” and must reciprocate at some future date to the giving house hold. Further, as Douglas (1 990) states, these informal exchanges operate where the market is absent. This was the duality of peasant existence in PDS SÃ£o Salvador that in tegrated non-market and market transactions within the livelihood system of peasant households. As Marcel Mauss (1990) wrote of the ex change of gifts between households, the exchange of meat between households involves a three-part process Â– to give, to receive, and to reciprocate. Participants in this revo lving process are first ob ligated to give game meat to kin or other households in the commun ity. Households are required to share their bounty of wild game with other members of the community, and the hording of meat is despised by neighbors. Households in PDS SÃ£o Salvador were also expected to accept wild game that is offered to them by others in the community. It is considered an offense to the offering party if a Â“giftÂ” of meat is refused. Refu sing an offering of wild game from another household in the community signifies that one is refusing to engage in social interaction
175 with other members of the community. This is looked upon as a breach of community trust, and the offending party is viewed as appearing to stand outsi de of the cultural norms of the community. This very infr equently occurs, though, as the receiving household is generally happy to accept a tasty re ward of wild game fr om their neighbors. The receiving household, in a sense, is consuming the material rewards of the other. Game meat is given to households that are expected to reciprocate in the partitioning of wild game in the future. Failure to reciprocate is a breach of this informal contract, as is the failure to receive, mentioned above. Therefore, the system operates with both households acting in all roles (giv ing, receiving, and reciprocating) within this social system of exchange. Meat exchange he lps to insure the survival of all members of the community as this rotating system of c onsumption feeds different households within the community. This system also fits within ScottÂ’s moral economy of the peasant (1976) that acts as a leveling mechanism in wealth and consumption between households in the community. If a wild game receiving household in PDS SÃ£o Salvador was not a hunting household, there were other means by which this non-hunting household could reciprocate its obligations to hunting househol ds. This might include returning fish resources, or giving part of the meat from a domesticated animal (normally chickens or ducks)5 to the hunting household that earlier ga ve part of its hunt to the non-hunting household. A householdÂ’s marketable agricu ltural products were rarely exchanged 5 Larger domesticated animals such as pigs and cattle we re generally viewed as a type of investment by the household and as such, the meat from these animals was normally not distributed to other households as payment of a householdÂ’s obligations. Cattle and pigs were usually slaughtered for sale, or during festas (parties) for the church, a school, a wedding, or a birthday.
176 between households, signifying the division th at existed between market and non-market products and transactions. Most households partitioned game meat to other households in the community based on three factors: 1) individuals involved in the hunting party, 2) the size of the kill, and 3) kin relationship between households. Any successful kill from a hunting outing was first equally divided amongst the hunters in a hunting party if there was sufficient meat to spread around. Then, if there was enough meat left to give to other households in the community, meat was further subdivided to selected households that were nearly always related by blood or marriage. Most hunting households stipulated that th e amount of meat they gave to other households depende do tamanho do bicho (depended on the size of the animal). If a single individual hunter returned to the community with a small animal (called embiara ) such as cutia ( Dasyprocta fuliginosa ), paca ( Agouti paca ), or quatipuru ( Sciuridae sp. ), or one of the commonly hunted birds in the seringal, including jacu ( Penelope jacquaou ), jacamim ( Psophia leucoptera ), papagaio ( Psittacidae sp. ), or nambu galinha ( Tinamus guttatus ), his family would keep the entire hunt . These hunts provide d a single meal for the family, but often the hunter had to return to the forest to hunt on the following day to put food on the table. With larger animals including queixada ( Tayassu pecari ), veado ( Mazama americana ), catitu ( Tayassu tajacu ), and capibara ( Hydrochaeris hydrochaeris ), the household of the successful hunter would either keep uma banda (onehalf of the animal) or um quarto (one-fourth of the animal). Whether the household kept one-half, or one-fourth, of the meat of the hunted animal depended upon how many households in the community were connected to the meat-giving household within the
177 system of reciprocal exchange of wild game . Several households re plied that they kept the same amount of meat that they parti tioned to other households, so that every household had the same amount of meat. One household stipulated that they gave the same quantity of meat to other house holds that they received in return. Kin were usually the channel of meat exch ange and thus, an important part of the channel of social relations within the co mmunity. Table 5.3 shows which householdÂ’s game meat was exchanged within the comm unity. Eighty percent of households that hunted in the survey (39 of 49 hunting househol ds) exchanged meat with households that were related through kin. Only 7 hunting hous eholds (14.3%) reported that (in addition to exchanging with kin relate d households) they shared wild game with households that were not related by kin.6 The one hunting household that did not share game meat did not have kin residing within the community. Table 5-3: Meat Sharing by Relation to Giving Household Kin Related Any Household Not Share No Response Given 39 (80.0%) 7 (14.3%) 1 (2%) 2 (4.1%) Women in peasant Amazonian communities are important actors in this system of inter-household relationships (Harris 2000). Although it is me n who hunt wild game and bring it back to their household, women cut, clean, cook, and partition meat to theirs, and frequently, other households in the communit y. Of the 49 households that hunted in the sample data, women participated in the shari ng of wild game to ot her households in the community in 35 (71.4%) of the hunting hous eholds. Table 5.4 shows the gender breakdown of meat division within hunting hou seholds interviewed in the study. Only 6 Three of these seven households did not have kin of either the husband or wife that lived in the community, yet they still exchanged wild game with their neighbors.
178 one hunting household (2% of hunting households ) did not share game meat with other households. Table 5-4: Sharing of Wild Game by Gender within H unting Households Men Only Women Only Both Does not Share 13 (26.5%) 12 (24.5%) 23 (46.9%) 1 (2.0%) Some individuals in PDS SÃ£o Salvador co mplained that other households did not share wild game with them. In one community in PDS SÃ£ o Salvador, the two different family groups, though related through inte r-marriage, were feuding and did not reciprocate in the partitioning of wild game with the other family. One woman in this community said that the othe r family group sometimes hid their successful hunt from her family group (thus a breach of the public spectacle of meat exchange within the community) so as to not be burdened with shar ing meat with their ne ighbors. In another community, a sister from one household, and a brother from another household, had been arguing for a period of a few days. One eveni ng the brother returned from a hunt with a few birds and a small monkey. Meat was given to the households of his other sister and three brothers who lived in the community, but not to the household of his sister with whom he was feuding. The spouse of the br other who hunted the birds and monkey, in support of her husband, said of her sister-in-la w, Â“Ela nÃ£o vai comer hoje.Â” (She is not going to eat today.) A common manner of showing displeasure and of punishing another household (or individual) was to refuse to share hunted game with the offending household (or individual).
179 Figure 5-5: Woman Cleaning Paca ( Agouti paca ) Hunted by Her Husband. It has been reported elsewhere in Amazoni a that men from different households would sometimes give meat to their female lovers living in other households within the community (Chagnon 1968; Siskind 1973; Kensi nger 1981). Siskind (1973) called this the reciprocal exchange of meat for sex. In one community in PDS SÃ£o Salvador, residents complained that th is occurred between a married man of one household and his married female lover from another household. Family of the spouses of the adulterous couple stated that the adulterous man woul d only hunt when his female lover requested that he bring her meat. They said that he was generally lazy, but quickly responded to the demands of his female lover. The adulter ous couple was eventually forced out of the community by the families of the jilted wife of one household and jilted husband of the other.
180 Inter-household Hypotheses on Hunting As is seen above, the practice of hunting in the settlement was not only determined by relationships within indivi dual households, but in additi on was critically affected by inter-household community norms, obligations , and relationships. Group hunts joined men from several different households in th e community, and overnig ht hunting activity necessitated men from several households bei ng away from their communities for up to a week at a time. Most of the men particip ating in these types of hunting activity were linked by relations of kin across households in the community. Further, meat exchange between households was based on cultural re sponsibilities of community, informally enforced by members of the community. An acknowledgment of these practices was critical to understanding how relationships between households influenced the practice of the hunt within the peasant communiti es of the seringal. Here we test the two hypotheses stated in Chapter 1 to account for the influence of inter-household relationships on the practice of hunting in PDS SÃ£o Salvador. Both hypothesis 8 and 9 were based on the assumpti on of cultural norms that join households together, as were reflected in the practic e of the hunt. Hypothesis 8 stated that households that were grouped together in cl usters within the community would hunt with greater frequency than households that were not grouped together within the community. This is based on the sharing of resources (specifically game meat) between households within the community that is the norm for Amazonian peas ant households (Harris 2000). Households that are in close, daily contact interact with greater frequency in the public display of hunting and reci procal meat exchange, enc ourage hunters to repay their obligations to other households in an effort to maintain their standing within the community.
181 During the rubber era along the Moa River, hous es were widely spaced, situated in the interior of each householdÂ’s colocaÃ§Ã£o, a nd often members of different households did not come into contact on a daily basis. With the collapse of the rubber economy in the area, households moved out of the forest to the margins of the waterways that cut through the seringal, and within close proximity of other houses (nor mally kin related to one another). Households began to cluste r into small single-family groups which only very recently coalesced into communities of several family groups. In this study, houses that had a clear line of sight to other houses in the community were considered to be clusters of households. Thirty-five of the 59 house holds sampled in this study were situated within eyesight of another house in the community. A T-test was performed to test the mean values of clustered houses in relation to houses that were not grouped together. The relationship indicated that hous es that were not grouped toge ther actually hunted with a greater frequency/month (2.25 times/month) than did houses that were clustered together (1.97 times/month). The relationship of clustere d houses did not prove to be statistically significant (n = 35, df = 57, p .399) and LevineÂ’s test showed that equal variances were assumed between the two populations (F = 1.104, p .298). Hypothesis 9 affirmed that households that had family living within the community will hunt more often than households that did not have family living within the community. As we have seen above, group hunt ing and meat exchange were frequently kin based; therefore, having family reside in the community should exert increased influence on the output of hunting by individu al households in order to maintain and nourish these inter-house hold relationships. Hunting is a form of non-market exchange
182 between households that joins family from different households, as opposed to market transactions that increasingly fall within the realm of indi vidual households. Non-market products (wild game, fish, small domesti cated animals) are exchanged between households, whereas market products (farinha, large domesticated animals) generally are not. A T-test was run to test whether or not a household had family residing in other households in the community affected their output of hunting. Since it was men who hunted together in the settlement, the mean va lues of households having kin of the men of the household were used in the comparison of hunting frequency. Thirty-six households fit this criteria (having male kin in other households), but the relationship showed that non-kin households hunted more often (2.30/mont h) than households th at have kin living in the community (1.94/month). The relations hip was not statistica lly significant (n = 36, df = 57, p .276) and equal variances within th e two populations were assumed (F = .147, p .702). It is possible that because of the reci procal exchange of wild game between households that is centered on kin relati onships within the community, isolated households (not in clusters) and households that did not have kin living close by were further removed from the transaction of shar ing meat between house holds and therefore, had to hunt with greater frequency to suppl y meat for their families. In conclusion, although neither hypotheses 8 or 9 proved to be statistically significan t, there were other indicators of inter-household influences that affected an individua l householdÂ’s output of hunting that did hold statisti cal significance. Some households in the seringal hunted with men from other households in the comm unity, while others chose not to. We might
183 expect that households that hunted together would do so w ith greater frequency due to inter-household bonds that were sustained be tween households in the community (partly through inter-household hunting re lationships) as well as r eciprocation between these households such as occurred in meat exchange explained above. Households that did hunt with members of other households did prov e to be highly significant in predicting hunting frequency (n = 35, df = 57, p .000). In addition, as stated above, often households that traded work days in agricultural labor were households that also hunt togeth er. The mean values of the 41 households trading labor days confirmed that households who did trade labor days hunted with greater frequency/month (2.39) than did househol ds that did not trade labor days (1.54). Further, the comparison was highl y significant (n = 41, df = 57, p .017). Modeling Inter-household Hunting Relationships As was the case in the previous chapter, multivariate linear regression was used to model the affects of inter-hous ehold variables on the depende nt variable, frequency of hunting. Tests for multicollinearity of the independent inter-household variables (Â“see housesÂ”, Â“kin in communityÂ”, Â“trade work days Â”, and Â“hunt with othe rsÂ”) were run before preceding with the regression models. A Chi-square test showed a high degree of association between the variables Â“trade work daysÂ” and Â“hunt with othersÂ” ( 2 = 10.681, p .001). T-tests run on the mean values of both of these variables also showed high statistical significance. I th eorize that kin relationships between households in the community that involve trading labor days between different households in the community lead to households hunting more fr equently with other households. Trading labor days and hunting with others are exampl es of social exchange between households and both are forms of non-market transactions . It is also possible, though, that this
184 relationship works in the opposite directi on with households that hunt together influencing labor exchange. Multiple regression was used to model the effects of the inter-household variables on the dependent variable Â“frequency of hunti ng.Â” The variables Â“see housesÂ”, Â“kin living in other householdsÂ”, and Â“hunt with othersÂ” were entered in Model 1. Model 2 included the variables Â“see housesÂ”, Â“kin livi ng in other householdsÂ”, and Â“trade work days.Â” Table 5.5 presents the re sults of the regression models. Table 5-5: Multivariate Regression Modeling of Inter-household Influence on Frequency of Hunting Model 1: Model 2: Variable -coefficients -coefficients See houses -.219** -.127 Kin in community -.150 -.209 Hunt with others .594*** Trade work days .333** Observations 59 59 Adjusted r2 .351*** .101** * Significant at p 0.10; ** significant at p 0.05; *** significant at p 0.01. In Model 1, the independent variable, Â“ hunting with othersÂ” proved to be highly significant (p .01). The variable Â“see housesÂ”, meas uring the clustering of households within the community, was significant, but in the opposite di rection that was hypothesized. The -coefficient value of -.219 for the variable Â“see housesÂ” means that as this variable decreased by 21.9%, the freque ncy of hunting increased by a factor of 1. The variable Â“kin in communityÂ” also show ed a corresponding nega tive relationship with the independent variable, likewise the opposit e of what was hypothesized in chapter 1. Model 1 did prove to be highly significant with 35.1% of the variance in the independent variable, Â“frequency of hunting,Â” explained by the independent variables that comprised the model (r2 = .351, p .01). Model 2, interchanging the va riable Â“trade work daysÂ” for
185 Â“hunt with others,Â” had much less explanator y power of the variance in the frequency of hunting (r2 = .101, p .05) than did Model 1. Overall, it appears from the data that inter-household hunting relationships are highly influenced by whether or not households hunt with othe r households (social) in the community. The variable Â“hunt with othersÂ” was the only variable th at indicated a high degree of statistical significance (p .01) while also greatly in fluencing the overall fit of the model. The high degree of association of the variable Â“trade work daysÂ” with the variable Â“hunt with othersÂ”, might indicate th at this effect was largely due to households that hunted together rather th an households that traded la bor days. The two variables Â“see housesÂ” (measuring the clustering affect of households in the community) and Â“kin living in the communityÂ” (in other househol ds) proved to have the opposite influence (negative) on the frequency of hunting than was expected. As expl ained above, this may be the result of meat exchange patterns between households wher eby households further removed from this reciprocal transaction (i solated households and households without kin in the community) were forced to hunt with greater frequency to provide meat for the household. Modeling Intra-househol d with Inter-household Effects on Hunting The hypotheses introduced in chapter 1 pred icted that a householdÂ’s frequency of hunting was the result of the effects of rela tionships and processe s occurring within the individual household joined with the influe nce of other households in the community. Here we combine the intra-household indepe ndent variables tested in the regression models of the previous chapter with the inte r-household variables test in the regression models above.
186 Model 1 from the inter-household linear regression models (see page 186) explained 35.1% of the variance in the depende nt variable, whereas the best fit of the intra-household regression mode ls (model 3 Â– page 154) ex plained only 24.9% of the variance in a householdÂ’s frequency of hunti ng. Here we combined inter-household independent variables with intra-househol d independent variables in multivariate regression modeling. The multivariate models tested here included only those variables that consistently atta ined statistical significance throughout the range of intra-household and inter-household regression models previously tested.7 To avoid the problems of multicolinearity, the variables Â“trade work daysÂ” and Â“hunt with othersÂ” were measured separately in Models 1 and 2. Model 3 include d those variables from Models 1 and 2 that reached statistical significance of at least p 0.05. Table 5.6 presents the results of these regression models. Table 5-6: Multivariate Regression Models (Intra-household and Inter-household) Model 1: Model 2 Model 3: Variable -coefficients -coefficients -coefficients Male-headed .181 .184* Hunting age boys .254** .115 .053 Emotional commitment .435*** .315*** .315*** to hunting Trade work days .170 Hunt with others .442*** .462*** Observations 59 59 59 Adjusted r2 .263*** .396*** .375*** *Significant at p 0.10; **significant at p 0.05; ***significant at p 0.01. The output of the regression models that combined intra-household with interhousehold independent variables as shown above in Table 5.6 was consistent with the models tested individually of the intr a-household variables (T able 4.11) and inter7 These variables included Â“male-headed househol dÂ”, Â“hunting age boysÂ”, Â“emotional commitment to huntingÂ”, Â“trade work daysÂ”, and Â“hunt with othersÂ”.
187 household variables (Table 5.5). It appears that a few of the variables had a large impact in explaining the variance of a householdÂ’s fr equency of hunting. Within an individual household, the Â“emotional commitment to huntingÂ”, and between households in the community, the variable Â“hunting with ot her households,Â” had highly statistically significant effects on the dependent vari able. Whether or not a household had domesticated animals or hunting age boys showed minor effects in explaining the variance in the dependent variable. The -coefficient values for the variables Â“maleheaded householdÂ” and Â“trade work daysÂ” in dicated that these variables also had a positive influence on a householdÂ’s hunting frequency. Regression modeling lends support to the s uggestion that households in PDS SÃ£o Salvador hunt primarily because that en joy the activity of hunting (ie. they are Â“emotionally committed to huntingÂ”) and because of the social relationships between households that are created and maintained with the practice of the hunt (Â“hunting with other householdsÂ”). This is an important fi nding indicating that th e livelihood systems of poor rural households are not pr edicated solely on the necessi ty of meeting the material needs of household members, but reflect the ag ency of rural house holds (that choose to participate in the practice of hunting) within the repertoire of po ssible livelihood options. While many households in PDS SÃ£o Salvador are increasingly turn ing away from hunting as the culture of farinha take s hold in the area, other hous eholds continue to actively engage in the practice of hunting as the pe rsonal enjoyment and social relationships between households derived from this livelihood activity are integral elements of the cultural identity of these rural people.
188 Conclusion Hunting is a key method of maintain ing a householdÂ’s kin networks in the community. Although others report that ki nship contact across households is womenÂ’s work (di Leonardo 2000), within PDS SÃ£o Salvador, both men and women of the household participated in the maintenance of kin relationships thr ough the hunting ritual. Men fostered links with men from othe r households in the community through the practice of hunting wild game in the forested areas of the settlement. This was largely accomplished through group hunting efforts (singl e-day hunts or overnight, extended-stay hunts) comprised of men from two or more different households in the community. As shown above, the majority of these group hunts joined men related by familial ties. Women also maintained kin networks through the partitioning of game meat to other households in the community. Though th e roles of men and women were gender stratified in the performance of the hunt, hunting as practiced within the settlement allowed both men and women to actively engage in the creation and maintenance of kin networks vital to a householdÂ’s (and communityÂ’s) survival. Refusing to share wild game and play a ro le in the reciprocal obligations between households was an affront to cultural norms in the community. Households were chastised for being stingy if they did not reciprocate wild game with others in the community. Further, since the majority of households shared w ild game with other households with which they were joined t ogether through familial ties, the unwillingness to partake in the role-p laying of meat exchange signified a transgression of kin relations. Kin relations were vital to the sustenance of households in the seringal as hunting and other forms of non-market exchange amongst households (i.e. trading labor days) are
189 integrated within the socio-economic fram ework of peasant livelihoods. Therefore, alienating familial bonds with other households was a risky venture. The multiple regression models indicated that a householdÂ’s hunting frequency reflected the influences of both intra-househol d and inter-household variables. The more salient of these variables were those of th e Â“emotional commitment of huntingÂ” of the individual household, whether or not househol ds Â“hunted with other householdsÂ” in the community. The inter-household variable Â“trading work daysÂ” also had a positive influence on the variance of a householdÂ’s hunting frequency, as did the intra-household variables Â“hunting age boysÂ” a nd Â“male-headed householdsÂ”. The variables Â“see housesÂ” and Â“domesticated animalsÂ” had a negative e ffect on the frequency of hunting, indicating that hunting frequency dropped as the number of a householdÂ’s domesticated animals increased.
190 CHAPTER 6 CONCLUSION In 1989, Parque Nacional Serra do Diviso r (PNSD) was established along the border with Peru in western Acre. This area contains BrazilÂ’s thir d highest diversity of primates, the greatest variety of palms in th e world, and has been designated an area of highest biological value in the Amazon region, yet in 2004 was populated by 522 families with a population of around 3100 people (IEB 2004).1 Brazilian Federal Law 9985/2000 prohibits people living within areas designated by the conser vation category of National Park and therefore, an attempt was made to relocate residents of the area outside the bounds of the National Park. A multidisciplinar y team of researchers from Brazil and the United States conducted fieldw ork in 1999 along the Moa River (the current site of PDS SÃ£o Salvador) that borders PNSD, looking for a potential site to move families that resided within the park. The State Â“discoveredÂ” that the Moa River was already settled with peasant families and that there was not sufficient land to support additional families from PNSD. This area was deemed a critical site in the buffer management of PNSD. Projeto Desenvolvimento SustentÃ¡vel (PDS) SÃ£ o Salvador was thus established in 2000 in an attempt to conserve th e natural resources of PNSD. The agenda of the newly created sustaina ble development settlement project (PDS SÃ£o Salvador), the first of its kind within Brazi l, was to manage natural resources in order to enhance the ecological and social sustai nability of the region. The management of 1 The Government of Acre in 2000 (ZEE 2000b) estimated the number of inhabitants living in PNSD to be approximately 5785 people. The lower population estimate of IEB might be accounted for by migration of individuals out of PNSD.
191 fauna, which has a critical function in the li velihood strategies of residents of PDS SÃ£o Salvador, was recognized by policy makers and local inhabitants as crucial in meeting the goals of sustainability for pres ent and future generations within the settlement and in the surrounding areas including PNSD . Animal resources must be managed sustainably in order to meet ecological need s of the forest ecosystem as well as human needs derived from wild game. Furthermore, wildlife will onl y be conserved if hunting is sustainable. Millions of peasants in the Amazon basi n rely on wild game to supply a large percentage of the protein intake in their diets. Rural households often do not have substitutes for the meat from wild game and therefore, must rely on their hunting prowess to supply the necessary calories and protein in take for their families. Hunting continues to be an integral part of the livelihood strategy of ru ral peasant households throughout Amazonia, yet there is a paucity of literatu re pertaining to hunting by the peasantry in Amazonia. This dissertation grew out of an attemp t to understand the motivations of hunting by peasant households living along the Moa Rive r in western Acre. The household is the appropriate locus of hunting by Amazonian peas ants as the practice of the hunt does not reside with the individual hunter himself/ herself, but results from a negotiation of interests and needs by family members. Furt hermore, it is here, at the level of the household, that livelihoods are created and main tained by rural peoples. It is argued here that the practice of hunting by the peasan try in Amazonia reflects the nature of relationships amongst individuals within th e household (intra-hous ehold) and between households in the community (inter-household) . Intra-household rela tionships reflect the material realities of the household, ge nder dynamics within the household, and the
192 emotional commitment to hunting by the household. Inter-house hold effects on hunting result from social obligations between house holds in the community, often joined by kin relations. Working with 59 households in PDS SÃ£o Sa lvador, in six different communities, I employed a variety of methods in the attemp t to understand the pr actice of the hunt by households in the settlement. Household surv eys, productive/reproductive surveys of the work of adult household members, spot sampling of adult activities, hunting outing surveys, and participant observation were th e tools used to unc over hunting practice by households sampled in the study. This diss ertation began with a review of how the livelihoods of residents of th e settlement have rapidly ch anged within the past two decades. The movement from rubber tapping to small-scale agricultural production centered on the production of farinha combine d, with the near simultaneous designation of the land base as a sustainable developmen t project, have had important implications for the practice of the hunt by households in the area. Next, I tested hypotheses to examine intra-household and inter-househol d effects on hunting practice by households of the settlement. In this chapter, I summar ize the principal findings of this study, discuss their implications for conservation and de velopment in PDS SÃ£o Salvador, and the implications of the study findi ngs for gender studies. Summary of Key Findings and Conclusions In Chapter 2, I examined the history of the Moa River in the past century and the changes in livelihood that have taken place fo r residents of PDS SÃ£o Salvador over the past several decades. I noted that household livelihood strategies in this part of western Acre have passed from the culture of rubber to one based on the culture of farinha. This change from livelihoods based on the extrac tion of the NTFP, rubbe r, necessitating the
193 maintenance of the forest ecosystem, to one based on small-scale agriculture and the conversion of forested land for crop producti on, has had important implications for the practice of hunting in the region. The output of hunting by households in PD S SÃ£o Salvador has decreased during this livelihood transformation in the past two decades. Household livelihood strategies, increasingly centered on sm all-scale agricultural producti on, entailed the clearing of forestland where the majority of hunting activit y occurred in the settlement. Therefore, agricultural pursuits removed men, the hunters, from the spaces in the settlement where animals congregated and hunting activity prev iously transpired. The work and time demands of agricultural producti on have further limited the ability of most men to pursue the activity of hunting in the forested areas of the settlement. Habitat for animals of the seringal has also been increasingly lost as trees are cut, land is cleared, brush is burne d, and crops are planted. Anim als were forced to retreat further from agricultural areas deeper into the forest to seek ecosystems that were favorable to their existence. Hunting nece ssitated wondering throughout the forest in the search of wild game. Hunting increasingly became a tiring, all-day pursuit, instead of a quick little jaunt close by oneÂ’s hous e to capture food for the family. Probably though, the biggest change for the inhabitants of PDS SÃ£o Salvador has been the social and political changes resulting from the chan ge in livelihoods from rubber to farinha. During the rubber era, households were widely scattered throughout the forest located on each householdÂ’s colocaÃ§Ã£o. As farinha took hold as the prominent livelihood strategy for inhabitants of the area by the 1990s , households began to move out from the forest environment to locations along the wate rways of the seringal. Households grouped
194 together in tight familial clusters as is the norm for peasant communities in Amazonia. With the establishment of the seringal as the settlement PDS SÃ£o Salvador in 2000, communities were created as a means to receive credit offered by INCRA. The social relations between households c ontained in the performance of the hunt may have also witnessed a transition as households moved together to create communities. Households that were in close proximity to each other (measured in this study by their visibility with one another with the variable Â“see housesÂ”) may have decreased their hunting output as the result of the increase in the sharing of wild game between these households. The reciprocal na ture of meat sharing, which was culturally inscribed in the settlement, potentially afforded clustered households the opportunity to decrease their frequency of hunting because their neighbors would supply them with meat on days when they did not hunt. The groupi ng of households in the settlement as a political strategy to obtain governmental cr edit combined with the breakdown of the rubber system of isolated houses scattered th roughout the forest has increased the social exchange between households through the practice of hunting. Members of the household desire meat to eat, and the shar ing of the product of the hunt, wild game, cements important social bonds between households. The grouping of households into communitie s has also led to a decrease in the number and size of animal encountered by hunters close by the comm unities. Residents of the settlement remarked how just a decade or so ago, animals would sometimes wander through their communities and men woul d grab their guns from their homes to kill their meals for the night (or next severa l nights), but that today hunters must hike several hours to capture wild game. As noted in the prev ious chapter, some hunters
195 turned to extended-stay, overnight hunting outin gs in response to the lack of wild game nearby settled areas. By 2005, larger game wa s mostly absent from close contact with settled areas, although this was not always th e case, with most da ily hunting excursions, if successful, resulting in the take of small animals (embiara) only. With the declaration of the PDS SÃ£o Sa lvador as a sustainable development settlement, the federal government granted leas e of the land base and use of its resources (within the appropriate federal laws and guidelines) to the re sidents of the settlement. The abandonment of the area by the previous la nd owners in the late 1980s (the Morais family) created a vacuum of uncertainty in te rms of ownership of the land and the rights to the natural resources contained within th is territory. When the federal agencies INCRA and IBAMA stepped in to formally de signate and manage the natural resource base, a new political system was installed. T hus, in the decade and a half from the late 1980s to the early 2000s, the rural inhabitant s that had lived along the Moa River for several generations experienced not only th e rapid transformation of their livelihoods from ones based on the extraction of rubber to small scale agricultu re centered on the production of farinha, but in conjunction, a pol itical transformation in the area from the patron-client relationship of the aviamento r ubber system, to a leadership vacuum in the seringal, to ownership and management by the State. Not surprisingly, there have frequently been hotly contested debates ove r the appropriate use of natural resources within the area amongst settlement resident s themselves, between residents and INCRA and IBAMA, and within these different federa l agencies as well. The issue of hunting and the use of wildlife by residents of PDS SÃ£o Salvador has been one of the most contentious and important issues to settlement inhabitants.
196 Rules were established for the management of natural resources in the sustainable development settlement with the joint partic ipation of the federal agencies INCRA and IBAMA, the Brazilian NGO PESACRE, and local residents. The management of fauna was the most hotly contested resource and cont inues to be a flashpoint of tension within many of the ten communities of the settleme nt. Regulations outlawing hunting with dogs have been mostly successful in limiting this traditional hunting techni que, yet the practice continues with some hunting households. Nonresidents of PDS SÃ£o Salvador hunt wild game in the area, and the threats to resident s of the settlement fr om these individuals continue to be a serious problem. There is also the uncertainty of the boundary and hunting te rritory between the Nukini Indigenous Reserve and the community of TimbaubÃ¡ in the northwestern portion of the settlement. The issue of territoriality and the owne rship of (hunting) spaces and natural resources (inc luding wild game) are integral to the friction between different cultural groups that live along the Moa Rive r. The northwestern part of PDS SÃ£o Salvador, an uninhabited 25,000 hectare region of the settlement, borders the Nukini Indigenous Reserve and was an area of la nd that the Nukini had sought to have incorporated within their indigenous reserve. INCRA, though, added this piece of land to the previously existing land base of PDS SÃ£o Salvador. The boundary between the land ownership of these two groups was not mark ed on the ground resulting in frequent disputes between the Nukini and residents of PDS SÃ£o Salvador regarding the ownership and use of resources in this area. In a ddition, the hunting of wildlife in this area by outsiders from the urban cente rs in the region has led to third-party conflicts over
197 resource use. There has been, though, a cau tious movement between leaders of the two communities to reconcile differences in use rights of wildlife. Further, the fact that wildlife is a Â“mobile resourceÂ” that freely crosses political boundaries adds to the uncertainty of management of this resource. Both the Nukini and the residents of PDS SÃ£o Salv ador have federal protecti on for their claims to the ownership of the land base that they each o ccupy, and the use of the resources within these land bases. A key difference is that the Nukini were granted ownership of their territory through the Brazilian federal agency responsible for indigenous affairs, FundaÃ§Ã£o Nacional do Ãndio (FUNAI), whereas th e residents of PDS SÃ£o Salvador were granted renewable 20-year lease rights from INCRA. This would seem to suggest that the Nukini have a stronger supporting federa l legal statute protecting their land rights than do the inhabitants of PDS SÃ£o Salvador . This may have implications for hunting and the use of wildlife along th e Moa River. In future hun ting disputes between these two groups will these federa l agencies (and potentially the courts) lean towards supporting the claims of the N ukini, or those of the reside nts of PDS SÃ£o Salvador? Herein lies an opportunity for state and federal agencies in tandem with the NGO, PESACRE, which has worked with settlement re sidents for nearly a d ecade, to participate in an ongoing dialogue with these diverse rural communities to efficiently manage resource conflicts while effectively managing the regionÂ’s resource base. In Chapter 4, I investigated intra-house hold effects on the practice of hunting in PDS SÃ£o Salvador. Although the overwhelmi ng majority of households interviewed hunted, less than one-half of households reported enjoying hunting as part of the
198 householdÂ’s repertoire of livelihood activ ities. Most hunters reported hunting for utilitarian reasons to supply meat for their families. Most hunters reported that they hunted with greater frequency in the wet winter months (November into May) than they di d in the dry summer months. This can be explained by three factors. First, a house holdÂ’s agricultural obligations, which peak in the dry season, preclude men from wandering the forest in search of wild game. As agricultural output dras tically slows during th e rainy season, men have more time to devote to hunting. Second, the winter months of February through April are described as mau de rancho when food resources are increasingly scarce in the seringal, forcing men to hunt with greater frequency to provide f ood for their families. Third, hunters stated that the wet forest floor in wi nter allowed them to stalk thei r prey in the forest without drawing undue attention to themselves and s caring away the animals they were pursing. Time allocation of adult behavior revealed that households are primarily preoccupied with productive pursuits in the dry, summer months, and with insuring that their reproductive needs were met in the wet, wint er months. Production did not increase in the settlement during winter to accommodate the decrease in food resources, but as ChayanovÂ’s model predicts, household cons umption dropped. A householdÂ’s material reali ties played an important function in hunting output. Hypothesis 1 stated that households with lots of children would hunt more than households with fewer children. A t-test in dicated statistical si gnificance for greater hunting frequency of households with five or more children. Hypothe sis 2 also proved to be statistically significant as households without domesticated animals hunted more frequently than households that did have domesticated animals. This indicates that the
199 meat of domesticated animals may act as a s ubstitute for the meat of wild game. I did find however, that an outside income source (wages, retirement checks) for the household did not have a statistically significant effect on hunting outpu t, and therefore, hypothesis 3 was rejected. Gender relationships within the househol d also had a critical role in the performance of the hunt in PDS SÃ£o Salvador . The forest environment, where most hunting activity occurs, has trad itionally been defined as a masculine space. Part of the socialization process of boys becoming men in the community was tied to learning how to hunt. Adolescent boys were taught how to hunt by their fath ers and older brothers in the household, and by their uncle s residing in other households within the community. Cultural norms of behavior limited women in the settlement from engaging in this activity. I argue that the decision to hunt does not rest solely with the hunter himself, but is dependent on the combined productive and re productive efforts of both men and women in the household. First, th e reproductive needs of men ar e cared for by women in the household, allowing men the opportunity to part icipate in the perfor mance of the hunt. Second, the work activities of men and wo men in the household are affected by the behavior of the other. Women make importa nt contributions to the food supply of the household, as is demonstrated by their part icipation in fishing and agricultural production. WomenÂ’s role in providing fish for the household influences the hunting output of men in the household. A hunter of the settlement remarked that when his household did not have fish, he was forced to hunt to provide food for his family. In
200 addition, in one community when residents suddenly starte d pulling large quantities of fish from the river, all hunting ac tivity momentarily ceased. In 49 of the 59 households surveyed in th is study men hunted, whereas in only one household did a woman report that she also hunted. Men spent 7.1% of their time devoted to hunting while womenÂ’s participati on in the pursuit of wild game within the forest was negligible (0.1% of their time ). In general, men focused on productive pursuits (hunting, agriculture, care of large animals) a nd women took care of the reproductive necessities of the household (ch ildcare, cooking, cleaning, care of small animals) while also helping with productive activities (production of farinha, fishing). Men were said to work in the fore st, and women within the house. Hypothesis testing of the influence of gender on hunting output was based on the fact that hunting was defined as a masculine activity in the seringal, one in which women infrequently partook. A set of three hypotheses (H4 Â– H6) were predictors of the influences of gender on a householdÂ’s hunti ng frequency. Whether or not the household head was male or female (H4) did not prove to statistically in fluence the householdÂ’s hunting output. Households with hunting ag e adolescent boys (H5) were a slightly significant indicator of a hous eholdÂ’s frequency of hunting. Finally, whether or not women of the household participat ed in fishing (H6) did not prove to be a statistically significant influence on hunting frequency. The practice of hunting continued to hold strong cultural ties for many hunters in the settlement. For some, hunting was more than just a means to provide a meal for the family, but was in addition, a means of rela xation from the daily pressures of the household and larger community. Hunters spok e of the calming nature of the forest as
201 they pursued wild game and how hunting a llowed them to escape the Â“noiseÂ” of the household. Others said that hunting allowed them to pass the enti re day without getting upset and that hunting was Â“a breakÂ” from work (in the roÃ§ado producing farinha). Hunting also afforded husband and wife tim e to Â“escapeÂ” from one another, allowing each an opportunity, uninterrupted by their s pouses, to socialize with members of their own gender. The variable Â“emotional commitment to hun tingÂ” was used as a proxy to measure a householdÂ’s willingness to continue to hunt as an integral part of the livelihood strategy for the household. Hypothesis 7 predicted th at as a householdÂ’s emotional commitment to hunting increased so w ould the householdÂ’s frequency of hunting. A Pearson correlation confirmed hypothesis 7 indicating a statistically si gnificant relation between a householdÂ’s emotional commitment to hunting a nd hunting frequency. In addition to intra-household influen ces on hunting behavior, in Chapter 5 I explored the relationships betw een households in the community that critically affected the performance of the hunt. Group hunts with members of different households created bonds between men from different households in the community. These hunting outings were normally organized along the lines of kin relationships between households. Community norms of behavior and ob ligations between households were intertwined with the hun ting ritual. Meat exchange of wild game with other households in the community was a culturally coded respons ibility for households in the settlement. The sharing of meat was a public display of meeting oneÂ’s obligations with other community members. The failure to participat e in the three-part r eciprocal process of sharing meat across households was an affr ont to cultural norm s of behavior of
202 community. Because households in the commun ity gave, received, a nd reciprocated wild game, this rotating system of consump tion fed different households within the community, helping to insure the survival of all members of the community. Both men and women actively engage d in creating and maintaining kin relationships across households through the performance of th e hunt. Men did so through their actions in pursing wild game with kin from other hou seholds in the community, and in the killing of wild game that was brought back to the household for redistribution to kin-related households. Women were the loci of the reciproc al sharing of wild game as they were the ones who partitioned the anim als captured by their husbands and sons to other households in the community. Although hunting activity was a highly gender stratified activity with men in the seringal defined as the hunters, women of th e household were integral to the system of hunting of the peasantry. As we have s een above, the reproduc tive and productive roles of women helped to transfor m the practice of the hunt by men of the household. Hunting also afforded both men and women the opportunity to play their roles in the shaping of kin relationships through m eat exchange between househol ds in the community. The decision to hunt was not, thus, made by men on their own, but reflected the negotiation of men and women of the household in th e allocation of labor an d other resources. Inter-household hypotheses on hunting freque ncy (H8 and H9) were based on the obligations and ties between hous eholds within the community that are reproduced in the performance of the hunt. Neither hypothesi s 8, predicting that households that were clustered together would hunt with greater frequency than households that were not grouped together, or hypothesis 9, stating that households th at had kin living in other
203 households within the community would hunt more often than households lacking kin in the community, proved to be significant. In fact, in both cases the exact opposite of what was predicted proved to be true: isolated hous eholds hunted with gr eater frequency than clustered households, and households lacki ng kin hunted with greater frequency than households that did have kin living in the community. It is possible that isolated households and households that did not have kin living close by hunted more often because they are further removed from the transaction of sharing meat between households and therefore, had to hunt with greater frequency to supply meat for their families. Two other inter-household variables, though, proved to be highly significant predictors of hunting behavior. Househol ds that hunted with members of other households in the community hunted with sta tistically significant gr eater frequency than did households that did not hunt with memb ers of other households. Also, households that traded labor days ( troca dias ) with other households hunted with greater frequency than did households that did not participate in trade work days with other households. Both of these variables reflect the nature of kin relationshi ps between households within the community. Multivariate regression models joining the set of intr a-household variables with inter-household variables indicate d that two variables in partic ular explained much of the variance of a householdÂ’s hunting output. The variable Â“emotional commitment to huntingÂ” within the household, and between hous eholds in the community, the variable Â“hunting with other householdsÂ” had highly statistically signifi cant effects on the dependent variable, Â“frequency of hunting.Â” Whether or not the household was male-
204 headed, if the household had hunting age boys, and if the household traded work days with other households s howed minor positive influence on a householdÂ’s hunting frequency. The variable Â“domesticated an imalsÂ” had a slight negative influence on hunting output by households in the study, signify ing that households with domesticated livestock hunted less than households th an lacked domesticated animals. In areas where wild game is mainly hunted for subsistence use and not for sale in the market as occurs in PDS SÃ£o Salvador , hunting persists in peasant economies not only because of the influence of material necessities of the household (although food security is important), but primarily because of other factors. The results of this study indicate that the greatest in fluences on the hunting output of the household were due to the desire to continue with hunting as a key component of the householdÂ’s livelihood strategy (emotional commitment to hunting), a nd because of the social bonds that were produced with the performance of the hunt. The hunting lifestyle, on one hand, reflects the choice of individual households in the re pertoire of livelihood possibilities. Some households hunted because they like to hunt and desired to continue wi th this practice. Hunting, further, reflects upon the social na ture of relationships between households, specifically the practice of hunting in kin gr oups, and the clustering of linked households. Peasant hunting, then, also persists because of the inter-household re lationships that it produces in the community. Peasant livelihoods reflect the mixture of market with non-market transactions and as such, are fundamentally different from cap italist livelihoods that are predicated on the market. Hunting plays an important role in the creation of non-market exchanges of the peasantry. Meat sharing, hunting with other households, and trading labor days are all
205 non-market transactions between households that are an esse ntial part of peasant hunting strategy in PDS SÃ£o Salvador. Whereas the market often involves indi vidual household li velihood strategies, peasant hunting in PDS SÃ£o Salvador enta ils communal livelihood strategies joining several households in the community. The pe rformance of the hunt helps to reinforce familial bonds across households as the social engagement of hunting between households (almost always) joins households li nked via kin relationshi ps. The cultivation of kin networks is extremely important in peasant communities that are marginalized from political and economic benefits that exist for other actors in society. These linkages act as safety valves in times of need, such as occurs with meat sharing during the winter months known as mau de rancho in PDS SÃ£o Salvador when food resources are difficult to acquire, helping to sust ain individual households, wh ile simultaneously creating a sense of community. Peasant hunting may also lead to a lesseni ng of further market integration and thus the Â“proletarizationÂ” of the peasantry. The ability of the peasantry in PDS SÃ£o Salvador to substitute non-market exchanges (meat sharing, trading labor days) for market transactions (purchasing meat) allowed them to refrain from further integration with the market that marginalized their contribution a nd role within the system. The practice of the hunt as a key form of non-market excha nge was thus, vital to the construction of peasant livelihoods that had some agency in terms of their engagement with the market system. Implications of Findings for Anthropology This study indicates the diversity that exis ts within peasant communities in western Amazonia. Even within comm unities that consisted of just a few households closely
206 joined by kin relationships in PDS SÃ£o Salv ador, there was a range of livelihood choices that reflected the different op tions of livelihood strategies an d resource use that peasant households chose to employ. Diversity occurs not only across wide social, political, and cultural boundaries and communities, but also within relatively small geographical and cultural spaces. Some rural households in PDS SÃ£o Salva dor continued to move towards closer integration with the market system, seen in the production of products for sale, notably farinha and livestock. The culture of rubbe r based on the extraction of forest products within a standing forest had been replaced by the culture of farinha that necessitated increased deforestation, with th e removal of individuals from the intact forested areas in the seringal in the transition from rubber ta pping to agricultural production, therefore, decreasing their hunting output as the frequency of contact with animals in the forest was similarly decreased. Other households in the settlement retained a forest culture that integrated hunting as an activ ity of fundamental importance w ithin the repertoire of the householdÂ’s livelihood strategy, in the process av oiding the necessity of increased market transactions. It is not that these hunting households were pr oactively resisting the market system, but that they avoided the market b ecause of the non-market exchanges that were reproduced and reciprocated, of ten contained within the pr actice of the hunt, between different households in the community. The social exchanges comprised in the performance of the hunt in PDS SÃ£o Salvador , including sharing of wild game between households (mostly aligned through relationshi ps with kin in the community) and hunting with kin in other households, were essential feat ures of the culture of the residents of the settlement.
207 Another result of this study, which was alluded to above, revolves around the peasantryÂ’s claims to ownership of territory, and the natural resources that lay within and that move through (wildlife) this land base. Along the Moa River in western Acre, there are four groups of peoples or institutions th at lay claim to land and the management of resources contained therein: the peasants in PDS SÃ£o Salvador, the inhabitants of the Nukini Indigenous Reserve, federal gove rnment agencies (INCRA, IBAMA, and FUNAI), and private landowners. The ag endas of each of these groups, although not inherently mutually exclusive, were also frequently in competition. If the ownership of land and the use of its resources ar e determined in terms of historical longevity of residence on a piece of land, the NukiniÂ’s claims supersede those of the other actors in the regi on. The Nukini were further pr otected by the State with the designation of their (current) land base of 27,000 hectares as an Indigenous Reserve in 1994. The rural inhabitants of PDS SÃ£o Sa lvador had lived along the Moa River for several generations, the area th at was declared a sustainabl e development settlement in 2000. Their land ownership rights were contai ned with the 20-year renewable lease that was granted to individual house holds. The federal governmentÂ’s first major impact in the debate over land tenure in the region occurr ed in 1989 with the d eclaration of Parque Nacional Serra do Divisor (PNSD). Residents who lived within the park boundaries prior to its designation as a national park have cont inued to resist their removal from this land that they claimed prior ownership of. Sin ce this time, INCRA, IBAMA, and FUNAI had become important players in the system of la nd tenure and resource use in western Acre. For the residents of PDS SÃ£o Salvador, ag ricultural plots were operated on an individualized household tenure sy stem, whereas the resources of the settlement that were
208 not contained within these household plots were managed on a co llective (community wide) basis.2 The sustainable management of natu ral resources including wildlife thus required the participation of the diverse hous eholds that comprised the ten settlement communities. As previously mentioned, the transition from familial-based political groupings of kin related households in the region to INCRA imposed community-based political units that united two or more ki n groups, had led to numerous conflicts. Infighting over resources within the settle ment, combined with boundary and resource disputes across political bounda ries (Nukini with settleme nt residents, settlement residents with neighboring private landowners, settlement residents with Â“visitingÂ” urban hunters, residents of the nationa l park with federal agencies), has increased tension within and between different cultural groups including the State, while increasing the need for a clear understanding of the owne rship of land and resource us e rights by all actors in the region. The Â“forest governmentÂ” of Acre has b een proactive in its commitment to integrating sustainable resource management policies while acknowledging the cultural, social, and economic contributions of rural dw ellers of the forest environs throughout the state (Kainer et al. 2003). The peasantry has been a key player in the agenda of the forest government that sought to encourage rural Acr eanos to continue to create sustainable and prosperous livelihoods, and refr ain from migrating to urban ar eas. The rural residents of PDS SÃ£o Salvador have staked their claim to the land and resource base along the areas of the Moa River that they have occupied fo r several generations w ith state and federal 2 To further complicate matters, in many instances, the maps that marked the boundaries of land tenureship in this region of western Acre were geographically inconsistent with what actually was the case on the ground (Salisbury 2002).
209 government support. Now is the opportunity for increased partnership with federal, state, and local organizations to ma nage land and natural resource s in a sustainable manner as was the intent with the declaration of Brazi lÂ’s first sustainable development settlement, PDS SÃ£o Salvador. Implications of Findings for Conservation and Development PDS SÃ£o Salvador was created in the search for alternatives to stem the tide of destructive land-use practices th at have plagued much of Amaz onia in the past. In doing so, a new strategy that joined governmental agencies with Brazilian NGOs and local residents of the settlement sought to account for the ecological diversity of the region while also respecting th e cultural and social histories and livelihoods of the inhabitants of the area. Community meetings lasting seve ral years established rules for the use of water, botanical, and animal resources in the settlement. Today the challenge is in the implementation of these regulations while cont inuing to search for ecological and social sustainability for future generations. This is a daunting task given th at the inhabitants of the settlement are among the poorest in the poor state of Acre, relying on the bounty of nature for their sustenance, and are a young popul ation with population pressure only to increase in the years to come. This research project unfolded as an atte mpt to provide information to a Brazilian NGO, PESACRE, to help the organization in the provision of conservation and development projects within the settlement. Wi ldlife was a critical resource for the rural poor that live along the Moa River in wester n Acre (and throughout Amazonia), yet prior to this study little was know n about the motivations of p easant households in terms of their hunting and the use of wild life. Animals of the seringal played a key role in the structure and functioning of the forest eco system, and the removal of animal species
210 through hunting might have negative consequenc es for the environment. As wild game was an important source of protein and food resources for the inhabitants of the settlement, the loss of animals might furthe r impoverish an already marginalized and poor group of people. Wildlife will only be co nserved if hunting is sustainable, and hunting output must be monitored. The findings of this study are mixed in te rms of their contribution to conservation and development in the region. Some of th e results of the stu dy are suggestive of a resistance to changing hunting practices that may be detrimental to ecological sustainability of wildlife use in the settle ment, while other results indicate possible substitutes exist for the hunting of wild game . Hunting in the settlement was principally motivated by the livelihood preference of the household and the social bonds that were created across households. These factors may contribute to the resistance of hunting households to alterations in their hunting pr actice. As mentioned in Chapter 4, some hunters stated that they would continue with hunting even if they had other sources of meat available to feed their families. Thes e hunters were largely resistant to decreasing their hunting output. Second, th e hunting of wild game strengthens communal relations across households, and limiting hunting output may lead to more individualistic livelihood strategies, lesseni ng the importance of inter-hou sehold social bonds that hunting helps to create and ma intain in the community. Inter-household networks were critical to the survival of individual households and th e community as a whole, and chipping away at these bonds created in th e performance of the hunt may change the socio-economic framework in the seringal. There may be, though, some alternatives to the use of wild game by households in the settlement.
211 The domestication of livestock was shown to lessen the frequency of hunting, as presumably households substituted home-grown meat for that of wild game. Generally it was the smaller animals (ducks, chickens) th at are raised for home consumption while larger animals (pigs, cattle) are kept as hous ehold investments. Women of the settlement were largely responsible for the care and maintenance of ducks and chickens that were critical food sources for the household, esp ecially in the winter months when food resources were more difficult to obtain. Small animal projects could provide an alternative source of meat for families while also potentially raising the status of women of the household who are invested in the rais ing of these animals. Because meat was a more highly valued source of food than were plant foods in the seri ngal, if women of the household had a greater stake in the provision of meat for the household, then the status of women in the household might rise as well. Another potential alternative is to increase the stock of fi sh in the settlement as a substitute for the meat of animals of the fo rest. Providing opportunities to capture more fish could limit the necessity to hunt by residents of the area.3 This could be accomplished through the promotion of small household fish ponds that would supply a year-round, easily accessible supply of fish for the household. While I was conducting my research in the settlement, two house holds were experimenting with constructing small fish ponds ( aÃ§udes ) next to their houses. They ha d done so without any assistance from the governmental and non-governmental agen cies that work in the area. I also 3 Although the participation of women in the activity of fishing in this study did not prove to influence the hunting frequency of men, this measured participation only, and not the catch of fish. In Rio Azul, when community members suddenly stated pulling large quantities of fish from the river, all hunting activity momentarily ceased. Therefore, if sufficient fish resources were available to the household, hunting frequency might decrease.
212 witnessed several households in the city of MÃ¢ncio Lima that had aÃ§udes providing fish for the whole family. This promotion of aÃ§ udes is potentially an important tool in limiting the hunting pressure on wildlife in the settlement. Anxious residents of the area, who were largely ignored by the state for decades, were initially preoccupied with the status of their lands, fearful that families from the Serra do Divisor National Park would be move d to SÃ£o Salvador. When studies deemed that it was not wise to move additional families to the area, the first several years of the settlement process were devoted to creat ing a management framework for the area through a participatory process of dialogue between government agencies and local inhabitants. The Conselho Gestor, cons isting of representatives from the ten communities that comprised PDS SÃ£o Salv ador, was established to promote and disseminate information and knowledge amongst residents of the settlement. Community regulations for the use of natural resources within the settlement were established, and PESACRE took the lead in educa tion programs for residents. As each household was assigned agricultura l lots of around 20 hectares by INCRA, and credit was extended to these designees, the emphasis on the management of resources waned within the communities and amongst the governmental agencies invested with the management of natural resources. Wildlife education programs for the residents of PDS SÃ£o Salvador must, though, continue if wild life in the settlement is to be managed sustainably. The settlement had reached the ne xt stage in its development as land tenure issues receded and management issues took cen ter stage. The challenges that lay ahead entailed how to manage the resources of the area in the face of growing population pressure to provide safe, su stainable and healthy futures for the environment and the
213 peoples that live in the region. Wildlife will continue to play an important role in the creation and maintenance of liv elihoods for residents of the settlement, and creative management alternatives must be developed to avoid the Â“empty forestÂ” syndrome. Implications of Findings for Gender Studies Traditionally, in hunting studies the role of men has held prominence, while that of women has largely been overlooked (Servi ce 1966; Lee and Devore 1968). In the evolution of human culture, men were depicted as the hunters of wild game while women were classified as gatherers of plant foods. The productive work of men in the provision of food resources for their families was held to be primary, with womenÂ’s role in food production seen as supplemental to the effort of men in the household. Women writers challenged this notion, with some arguing that gathering by wo men, not hunting by men, played a more important role in the de velopment of human so ciety (Dahlberg 1981). I have argued throughout this study that the roles of men and women of the household are both important in the performa nce of the hunt in peasant societies. Although the roles of men and women in hunting are highly gender stra tified, the practice of the hunt results from the negotiation between male and female memb ers at the level of the household. The productive efforts of men in the pursuit of wild game are dependent on the productive pursuits of women in pr ocuring food for the household and vice-aversa. WomenÂ’s role in fishing often de termined whether men of the household would pick up the gun to hunt wild game or not. As one male hunter from PDS SÃ£o Salvador remarked, Â“When there are no fish, one must hu ntÂ”. Another man in the settlement had given up hunting altogether because his wife was such a successful fisherperson who provided ample quantities of fish on a regular basis for the household.
214 The end point of the hunt, wild game, is tied to the rituali zed performance of reciprocity between peasant households that requires the participation of both men and women. Men bring wild game to their wi ves who distribute the product to (other households in) the community. The practice of the hunt provides a space for both men and women of the household to actively create and maintain relations with kin in the community through the reciprocal division of ga me meat with other households. In areas largely devoid of social servic es, the maintenance of kin rela tions is critical in insuring the survival of peasant households.
215 APPENDIX A HOUSEHOLD SURVEY Data:_________________ Entrevistador:__________ Comunidade:__________ Casa#:_________________ [CondiÃ§Ã£o da casa: Pintada?_____ Telhado de alumÃnio?_____ Pode ver outras casas na comunidade?______ Localizado na beira do rio/de ntro na mata/na capim?____________] QUESTIONÃRIO 1. Onde nascerem vocÃªs? (EstadoÂ…) Ele:_____________________ Ela:_________________ Seus filhos?_______________ 2. Onde nascerem seus pais? (EstadoÂ…) (Ele) Pai:_________________ (Ela) Pai:__________________ MÃ£e:________________ MÃ£e:_________________ 3. Quantos anos vocÃªs moram no seringal? Ele:__________ Ela:__________ Quantos anos vocÃªs moram nesta comunidade? Ele:___________ Ela:___________ Moram aqui o ano todo? Ele:_______________ Ela:_______________ Filhos:_______________ 4. Quantos anos vocÃªs tem? Ele:________; Ela:________ Quantos filhos vocÃªs tÃªm na casa? (Marcar as idades de cada umÂ…)
216 Filhos: _____ Filhas: _____ _____ _____ _____ _____ _____ _____ 5. TÃªm famÃlia morando prÃ³ximo a vocÃªs? (Quem, e onde?)_____________________________ Em MÃ¢ncio Lima?________(Quem?)_____________ Em Cruzeiro do Sul?_______(Quem?)_______________ 6. TÃªm filhos ou/e famÃlia morando fo ra do seringal que manda dinheiro a vocÃªs?____________ VocÃªs mandam dinheiro a eles?_________________ 7. VocÃªs tÃªm criaÃ§Ã£o de animais domÃ©sticos?__________(Quantos?) Boi:_________ Porcos:_______ Galinhas:_____ Pato:_________ Coelho:_______ Outros:________ 8. VocÃªs vendem estes animais as outras famÃlias na comunidade?________________ Vendem nos mercados de MÃ¢ncio Lima?_________ Cruzeiro do Sul?_________ 9. Trabalham com diÃ¡rias na terra das outras famÃlias no seringal? (Quem na casa?Â…)_________ VocÃªs tÃªm emprego fixo em MÃ¢ncio Lima ou Cruzeiro do Sul? (Quem?)______________ 10. Recebem cheques de aposentadoria, defici Ãªncia, dinheiro para ensinar no seringal, agente de saÃºde, salÃ¡rio maternid ade, ou soldado de borracha? (Quem e qual das coisas?Â…) 11. AlguÃ©m na casa tem conta no banco em MÃ¢ncio Lima ou Cruzeiro do Sul? (Quem?)_______ Motoserra?_________ Barco com motor?_______ Radio?________ 12. VocÃªs trocam dias de trabalho co m outras casas na comunidade?___________(Com quem?)
217 13. VocÃªs caÃ§am animais silvestres?___________Por que sim ou nÃ£o? 14. VocÃªs caÃ§am sozinhos?____________ CaÃ§am com outras pessoas da casa? (Quem?)_________________ CaÃ§am com outras pessoas na comunidade? (Quem?)_________________ 15. TÃªm cachorros?__________CaÃ§am com cachorros?___________ Outras casas da comunidade caÃ§am com cachorros?____________________ 16. TÃªm espingarda?__________Quantos tÃªm?_________ 17.Quantas vezes por mÃªs vocÃªs saem para caÃ§ar?__________________ JÃ¡ houve Ã©poca no passado que vocÃªs caÃ§avam mais do que hoje?_______(Quando?) 18.VocÃªs gostam de caÃ§ar?_______________Por que sim ou nÃ£o? 19. Como decidem quando Ã© a hora de sair para caÃ§ar? 20. CaÃ§ar Ã© algo que vocÃªs gostam de fazer ou Ã© uma coisa que vocÃªs fazem porque Ã© necessÃ¡rio? (ExplicaÂ…) 21. VocÃªs pensam que caÃ§ar Ã© mais difÃcil do quer trabalhar na agricultura?_______________ Mais difÃcil do que a venda do produto?______________________ Mais difÃcil do que cuidar dos filhos?_________________________ Mais difÃcil que a pesca?___________________________________ (*Por cada um, indica porquÃª sim ou nÃ£o Ã© mais difÃcilÂ…) 22. CaÃ§am durante o ano todo?_________ CaÃ§am mais durante a est aÃ§Ã£o de chuva ou seca?_____________(Por que?) 23. VocÃªs acham que a caÃ§a faz parte da re sponsabilidade do homem para com a famÃlia?_______ (Por que sim ou nÃ£o?) 24. VocÃªs acham que sÃ³ os homens devem caÃ§ar?_____________(Por que sim ou nÃ£o?) 25. Que idades vocÃªs comeÃ§aram caÃ§ar no seringal? Pai:__________ Filhos:_________ 26. Quem ensinarem vocÃªs a caÃ§ar? Pai:_____________
218 Filhos:________________ 27. Se vocÃªs tivessem outras carnes disponÃ veis para comer e nÃ£ o precisasse caÃ§ar, seguiriam caÃ§ar ou nÃ£o?_______________(Por que?) 28. Onde caÃ§am no seringal? Na floresta?_______ No roÃ§ado?_______ Na beira do rio e os igarapÃ©s?_________ Outros lugares?_________________ 29. Tem gente que mora fora do seringal que vem ao seringal para caÃ§ar?______________ Para pescar tambÃ©m?_________________________ 30. Se vocÃªs voltam a casa com carne da mata, dividem a carne com outras casas da comunidade?___________________ Com quem? 31. Tem outras casas na comunidade que dar a carne da mata a vocÃªs?____________(Quem?) 32. Como decidir a quantidade de carne da mata que vocÃªs dividem com outras casas na comunidade? 33. Quem na casa decide repartir a carne da mata com as outras casas? 34. VocÃªs vendem carne da mata as outras casas na comunidade?___________ Vendem em MÃ¢ncio Lima ou Cruzeiro do Sul?_______________ Tem outras casas na comunidade que vendem carne de mata as outras casas na comunidade?________________ Eles vendem em MÃ¢ncio Lima ou Cruzeiro do Sul?________________ 35. VocÃªs compram carne da mata das outras casas na comunidade?______________ Compram a criaÃ§Ã£o dos animais domesti cados, como a galinha, pato, coelho, o porco, das outras casas na comunidade?_________________ 36. VocÃªs compram carne de lata em MÃ¢ncio Lima ou Cruzeiro do Sul?____________ O que carne compra? 37. Quem na casa pesca? 38. VocÃªs vendem o peixe que vocÃªs pescam, dentro da comunidade?______________ Vendem em MÃ¢ncio Lima ou Cruzeiro do Sul?______________ 39. VocÃªs dividem peixe que vocÃªs pescam com outras casas na comunidade?________
219 (Com quem?) 40. Tem outras casas na comunidade que dar peixe que eles pescam a vocÃªs?______________ (Quem?) 41.De quanto Ã© quanto tempo vocÃªs baixam para a cidade? Baixam juntos? Ele:__________ Ela:__________ Filhos:__________________ 42. VocÃªs receberem o credito do INCRA? NÃ£o As vezes Sempre/sim A. Gostamos de caÃ§ar? B. A caÃ§a Ã© fÃ¡cil? C. CaÃ§a nÃ£o Ã© uma perda de tempo mesmo se voltar para casa sem care da mata? D. Gostamos de caÃ§ar mais de fazer outro trabalho? E. Continuam caÃ§ar mesmo que nÃ£o precisse? F. Prefeririam caÃ§ar do que compra carne? G. A caÃ§a nÃ£o mexe/atrapahla com nossa outro trabalho? H. CaÃ§am para socializar com os filhos, amigos, e vizinhos?
220 APPENDIX B PRODUCTIVE/REPRODUCTIVE HOUSEHOLD SURVEY Data:____________ Entrevistador:____________ Communidade:_____________ Casa#:____________________ Casados/Viudo(a)/S oltero(a):__________ Persoas na casa (com idades): Filhos:________________________ Filhas:________________________ Outras persoas:__________________ ATIVIDADES DA GENTE DA CASA Atividade Marido Esposa Filhos Filhas Outra persoa (AvÃ´, AvÃ³) Cuidar as crianÃ§as Cozinhar Lavar os pratos Costurar Lavar a roupa Colher agua Colher lenha Estudar Cuidar os animais: Porcos Galinhas Vacas Outro? Pescar CaÃ§ar Trabalho com diÃ¡rias Trocar dias Serra madeira Corta o capim Brocar Plantar: Milho FeijÃ£o Mandioca
221 Atividade Marido Esposa Filhos Filhas Outra persoa (AvÃ´, AvÃ³) Legumes Outra? Fazer farinha Vender os produtos Coher dentro da floresta: AÃ§ai Mel de abelha Buriti CipÃ³ Outra? Ensinar/Professor Cheque de aposentadoria SalÃ¡rio de maternidade Outra coisa?
222 APPENDIX C TIME ALLOCATION Data:_________________ Hora:_________________ Sol/Chuva:_____________ Entrevistador:__________ Communidade:__________ Casa#:_________________ AMOSTRA NA HORA Atividade Marido Esposa Filhos Filhas Outra Persoas? (AvÃ´, vizinho..) Cozinhar Comer Passear (Com quem?) Descansar Dormir Costurar Cuidar as crianÃ§as Lavar roupa/pratos (Qual?) Colhe agua Colhe lenha Cuidar os animais (Quais animais?) Pescar CaÃ§ar Plantar (O que?) Cortar a capim/brocar Serra madeira Colhe dentro da floresta Vender produtos Fazer farinha Estudar Ensinar/Professor Jogar/brincar Viajar afora Outra coisa? NinguÃ©m esta na casa
223 APPENDIX D HUNTING OUTINGS Comunidade:____ _________ Etrevistador:______________ SAIDOS DE CAÃ‡AR Sol/Chuva SaÃa CaÃ§ar (Onde?) #Casa CaÃ§a Sozinho/ Com Outras #Tiros Animais Matados Quem Matou (Onde?)
224 LIST OF REFERENCES Allegretti, Mary 1995 The Amazon and Extracting Activities. In M. ClÃ¼sener-Godt and I. Sachs, eds. Brazilian Perspectives on Sust ainable Development of the Amazon Region. 157-174. New York: The Parthenon Publishing Group Ltd. Almeida, Anna Luiza Ozorio de 1992 The Colonization of the Amazon. Au stin, Texas: The University of Texas Press. Alvard, Michael 1995 Intraspecific Prey Choice by Amazonian Hunters Current Anthropology 36(5): 789-818. Alvard, Michael, John Robinson, Kent Redford, and Hillard Kaplan 1996 The Sustainability of Subsisten ce Hunting in the Neotropics. Conservation Biology 11(4): 977-982. Anderson, Anthony 1990 Alternatives to Deforestation: Steps toward Sustainable Use of the Amazon Rain Forest. New York: Columbia University Press. Anderson, Leslie 1994 The Political Ecology of th e Modern Peasant: Calculation and Community. Baltimore: The Johns Hopkins University Press. Aspelin, Paul 1979 Food Distribution and Social Bonding among the MamaindÃª of Mato Grosso, Brazil. Journal of Anth ropological Research 35(3): 309-327. AugÃ©, Marc 1995 Non-Places: Introduction to an Anthropology of Supermodernity. New York: Verso. Bakx, Keith 1988 From Proletarian to Peasant: Ru ral Transformation in the State of Acre, 1870-1986. Journal of Development Studies 24(2): 141-160.
225 Barham, Bradford and Oliver Coomes 1994 Wild Rubber: Industrial Orga nisation and the Microeconomics of Extraction during the Amazon Rubber Boom (1860 Â– 1920). Journal of Latin American Studies 26(1): 37-72. Beckerman, Stephen 1994 Hunting and Fishing in Amazoni a: Hold the Answers, What Are the Questions? In, Anna Roosevelt, ed. Am azonian Indians from Prehistory to the Present: Anthropological Pe rspectives. 177-200. Tucson: The University of Arizona Press. BenerÃa, Lourdes 1979 Reproduction, Production and the Sexual Division of Labour. Cambridge Journal of Economics 3: 203-225. Bennett, Elizabeth 2002 Is There a Link Between Wild Meat and Food Security. Conservation Biology 16(3): 590-592. Bennett, Elizabeth, E. Milner-Gulland, M. Bara rr, H. Eves, J. Robinson, and D. Wilkie 2002 Hunting the WorldÂ’s Wildlife to Extinction. Oryx 36: 328-329. Bernstein, Henry 1982 Notes on Capital and Peasantry. In, John Harris, ed. Rural Development: Theories of Peasant Economy a nd Agrarian Change. 160-177. London: Hutchinson University Library. Bodmer, Richard 1991 Strategies of Seed Dispersal a nd Seed Predation in Amazonian Ungulates. Biotropica 23: 255-261. Bodmer, Richard 1989 Frugivory in Amazonian Artiodactyl a: Evidence for the Evolution of the Ruminant Stomach. Journal of Zoology 219: 457-467. Bodmer, Richard, Tula Fang, and Ibanez, L. 1988 Ungulate management and conservation in the Peruvian Amazon. Biological Conservation 45: 303-310. Bodmer, Richard, Etersit Lozano, and Tula Fang 2004 Economic Analysis of Wildlife Us e in the Peruvian Amazon. In, Kirsten Silvius, Richard Bodmer, and Jose Fragoso, eds. People in Nature: Wildlife Conservation in South and Central America. 191-207. New York: Columbia University Press. Bodmer, Richard, P. Puertas, L. Moya and Tual Fang 1994 Estado de las Poblaciones del Ta pir en la AmazonÃa Peruana: En el Camino de la ExtinciÃ³n. BoletÃn de Lima
226 Boserup, Ester 1970 WomanÂ’s Role in Economic Devel opment. New York: St. MartinÂ’s Press. Bouquet, Mary 1984 WomanÂ’s Work in Rural Southwest England. In, Norman Long, ed. Family and Work in Rural Societie s: Perspectives on Non-wage labor. 142-159. London: Tavistock Publications. Branford, Sue 1985 The Last Frontier: Fighting Over Land in the Amazon. London: Zed Books Ltd. Calouro, Armando 1995. CaÃ§a de SubsistÃªncia: Susten tabilidade e PadrÃµes de Use entre Seringueiros Ribeirinhos e NÃ£o-ribe rinhos do Estado do Acre. BrasÃlia. DissertaÃ§Ã£o de Mestrado em Ecologia. Universidade de BrasÃlia. CÃ¢mara, Evandro Leal 1999 DiagnÃ³stico SÃ³cio-EconÃ´mico e EcolÃ³gico. SÃ©rie Assentamento SustentÃ¡vel SÃ£o Salvador. Volume I. Rio Branco, Acre: PESACRE. Campbell, Connie 1996 Out On the Front Line but Sti ll Struggling for Voice: Women in the Rubber TappersÂ’ Defense of the Forest in Xapuri, Acre, Brazil. In, Dianne Rocheleau, Barbara Thomas-Slayter a nd Esther Wangari, eds. Feminist Political Ecology: Global Issues and Local Experiences. 27-61. New York: Routledge. Campbell, David 2005 Land of Ghosts: The Braided Lives of People and the Forest in Far Western Amazonia. Boston: Houghton Mifflin Company. Chagnon, Napoleon 1968 YanomamÃ¶: The Fierce People. Ca se Studies in Cultural Anthropology. Fort Worth, Texas: Harcourt Brace College Publishers. Chayanov, A.V 1986 The Theory of Peasant Econom y. Madison: The University of Wisconsin. Chibnik, Michael 1994 Risky Rivers: The Economics and Politics of Floodplain Farming in Amazonia. Tucson, Arizona: The University of Arizona Press. Church, G. E. 1904 The Acre Territory and the Caoutchouc Region of South-western Amazonia. The Geographical Journal 23(5): 596-613.
227 Comunidade do Projeto de Desenvolvi mento SustentÃ¡vel SÃ£o Salvador 2003 Plano de Use dos Recursos Na turias do Projeto de Desenvolvimento SustentÃ¡vel SÃ£o Salvador. Ri o Branco, Acre: PESACRE. Coomes, Oliver and Bradford Barham 1994 The Amazon Rubber Boom: Labor Control, Resistance, and Failed Plantation Development Revisited. Th e Hispanic American Historical Review 74(2): 231-257. Cronkleton, Peter 1998 Landownership Turnover and Family Farm Survival in an Amazon Resettlement Project. PhD Dissert ation, University of Florida. Cunha dos Santos, Magna 2003 MÃ¢ncio Lima: PopulaÃ§Ã£o Ã Margem de Terras Federais. In Fabiano Toni and David Kaimowitz, org. MunicÃpios e GestÃ£o Floresta na AmazÃ´nia. 295-335. da Cunha, Manuela Carneiro and Maur o Barbosa de Almeida, organizers 2002 EnciclopÃ©dia da Floresta: O Alto JuruÃ¡: PrÃ¡ticas e Conhecimentos das PopulaÃ§Ãµes. SÃ£o Paulo, Brazil: Companhia das Letras. da Cunha, Manuela Carneiro and Mauro de Almeida 2000 Indigenous people, Traditional People, and Conservation in the Amazon. Daedalus 129(2): 315-338. Dahlberg, Frances, ed. 1981 Woman the Gatherer. New Ha ven: Yale University Press. Davis, Shelton 1977 Victims of the Miracle: Deve lopment and the Indians of Brazil. Cambridge: Cambridge University Press. Dean, Warren 1987 Brazil and the Struggle for Rubbe r: A Study in Environmental History. Cambridge: Cambridge University Press. Deere, Carmen Diana and Magdalena LeÃ³n 2001 Empowering Women: Land and Property Rights in Latin America. Pittsburgh: University of Pittsburgh Press. Deere, Carmen Diana and Alain de Janvry 1979 A Conceptual Framework for the Empirical Analysis of Peasants. American Journal of Agricult ural Economics 61(4): 601-611. Descola, Phillipe 1996 In the Society of Nature: A Native Ecology in Amazonia. Cambridge: Cambridge University Press.
228 Diegues, Antonio Carlos 1992 The Social Dynamics of Defore station in the Braz ilian Amazon: An Overview. Discussion Paper No. 36: Un ited Nations Research Institute for Social Development. Douglas, Mary 1990 Foreward. In, Marcel Mauss. The Gift: The Form and Reason for Exchange in Archaic Societies. vii-xviii. New York: W. W. Norton. Dwyer, Daisy and Judith Bruce, eds. 1988 A Home Divided: Women and In come in the Third World. Stanford: Stanford University Press. Ellis, Frank 2000 Rural Livelihoods and Diversity in Developing Countries. Oxford: Oxford University Press. Ellis, Frank 1993 Peasant Economies: Farm Hous eholds and Agrarian Development. Cambridge: Cambridge University Press. Emmons, Louise 1999 Neotropical Rainforest Mammals: A Field Guide. Second Edition. Chicago: The University of Chicago Press. Fearnside, Philip 1997 Environmental Services as a St rategy for Sustainable Development in Rural Amazonia. Ecological Economics 20: 53-70. Fearnside, Philip 1984 BrazilÂ’s Amazon Settlement Schemes: Conflicting Objectives and Human Carrying Capacity. Habitat International 8(1): 45-61. Fine, Lisa 2000 Rights of Men, Rites of Passage: Hunting and Masculinity at Reo Motors of Lansing, Michigan, 1945-1975. Journa l of Social History 33(4): 805823. Fragoso, JosÃ©, Magna Cunha dos Santos, and VÃ¢ngela Nascimento 2002 A Fauna Silvestre e as PrÃ¡ticas de CaÃ§a. Rio Branco, Acre, Brazil: PESACRE. Friedman, Harriet 1980 Household Production and the National Economy: Concepts for the Analysis of Agrarian Formations. J ournal of Peasant Studies 7(2): 158184.
229 Galetti, Mauro, Marco AuÃ©rlio Pizo and PatrÃcia Morellato 2003 Fenologia, Frugivoria e DispersÃ£o de Sementes. In Laury Cullen, Jr., Rudy Rudran, and Claudio Valladares-Padua , organizers. MÃ©todos de Estudos em Biologia da ConservaÃ§Ã£o and Mane jo da Vida Silvestre. Curitiba, Brazil: FundaÃ§Ã£o O BoticÃ¡rio de ProteÃ§Ã£o Ã Natureza. Godelier, Maurice 1986 The Making of Great Men: Ma le Domination and Power among the New Guinea Baruya. Cambridge: Cambridge University Press. Godoy, Ricardo 2001 Indians, Markets, and Rainforest s: Theory, Methods, Analysis. New York: Colombia University Press. Godoy, Ricardo, David Wilkie, and Jeffrey Franks 1997 The Effects of Markets on Neot ropical Deforestati on: A Comparative Study of Four Amerindian Societie s. Current Anthropology 38(5): 875878. Gomes, Valeiro 2001 Dynamics of Land-use in an Amaz onian Extractive Reserve: The Case of the Chico Mendes Extractive Reserve in Acre, Brazil. MA Thesis. Gainesville: University of Florida. Goodland, R. and H. Irwin 1975 Amazon Jungle: Green Hell to Red Desert? Amsterdam: Elsevier Scientific Publishing Company. Government of Acre. http://www.ac.gov.br/. Accessed November 8, 2004. Grinker, Roy Richard 1994 Houses in the Rainforest: Ethnicity and Inequality among Farmers and Foragers in Central Africa. Berkeley : University of California Press. Gross, Daniel 1984 Time Allocation: A Tool for the Study of Cultural Behavior. Annual Review of Anthropology 13: 519-558. Gudeman, Stephen and Alberto Rivera 1990 Conversations in Colombia: Th e Domestic Economy in Life and Text. Cambridge: Cambridge University Press. Guerra, Raissa 2002 Ã‰ PossÃvel Atingir a Sustentab ilidade nos Assentamentos de Reforma AgrÃ¡raia na AmazÃ´nia Legal? O Caso do PDS SÃ£o Salvador no Estado do Acre. MA Thesis. Universi dade de BrasÃlia, Brasil.
230 Hall, Anthony 1989 Developing Amazonia: Deforesta tion and Social Conflict in BrazilÂ’s CarajÃ¡s Programme. Manchester: Manchester University Press. Hames, R. 1979 A Comparison of the Efficien cies of the Shotgun and the Bow in Neotropical Forest Hunti ng. Human Ecology 7(3): 219-252. Harris, Mark 2000 Life on the Amazon: The Anthropology of a Brazilian Peasant Village. Oxford: Oxford University Press. Harris, Mark 1998 The Rhythm of Life on the Amaz on Floodplain: Seasonality and Sociality in a Riverine Village. The Journal of the Royal Anthropological Institute 4(1): 65-82. Hawkes, Kristen 1993 Why Hunter-Gatherers Wor k. Current Anthropology 34(4): 341-361. Hawkes, Kristen, Kim Hill, and James OÂ’Connell 1982 Why Hunters Gather: Optimal Foraging and the AchÃ© of Eastern Paraguay. American Ethnologist 9(2): 379-398. Hecht, Susan and Alexander Cockburn 1990 The Fate of the Forest: Developers, Destroyers and Defenders of the Amazon. New York: Harper Perennial. Hecht, Susan, Anthony Anderson and P. May 1988 The Subsidy From Nature: Shifting Cultivation, Successional Palm Forests and Rural Development. Human Organization 47: 25-35. Hegmon, Michelle 1991 The Risks of Sharing and Shari ng as Risk Reduction: Interhousehold Food Sharing in Egalitarian Societies. In, Between Band and States. Susan Gregg, ed. 309-329. Center for Archaeol ogical Investigations, Occasional Paper No. 9: Southern Illinois University. Hell, Bertrand 1996 Enraged Hunters: The Domain of the Wild in North-western Europe. In Philippe Descola and GÃsli PÃ¡lss on, eds. Nature and Society: Anthropological Perspectives . 205-217. London: Routledge. Henrich, Joseph 1997 Market Incorporation, Agricultura l Change, and Sustainability among the Machigueuga Indians of the Peru vian Amazon. Human Ecology 24(2): 319-351.
231 Hill, Kim and Hillard Kaplan 1993 On Why Male Foragers Hunt and Share Food. Current Anthropology 34(5): 701-706. Hill, Kim, Hillard Kaplan, Kristen Hawkes, and Ana Hurtado 1985 Menâ€™s Time Allocation to Subsis tence Work among the Ache of Eastern Paraguay. Human Ecology 13(1): 29-47. Homma, Alfredo, A. Contro, A. Carlos, and R. Carvalho 1992 A Pecuria na Transamaznica. O Liberal. Belm. 30 July 1992. 6. Hurtado, Ana, Kristen Hawkes, Kim Hill, and Hillard Kaplan 1985 Female Subsistence Strategies among Ache Hunter-gatherers of Eastern Paraguay. Human Ecology 13(1): 1-28. Hyden, Goran 1980 Beyond Ujamaa in Tanzania: Underdevelopment and an Uncaptured Peasantry. Berkeley: Universi ty of California Press. Institutio International de Educao do Brasil (IEB) 2004 O Divisor que Nos Une. 40 min. Director: Maristela Bernardo. Instituto Nacional de Coloniza o e Reforma Agrria (INCRA) 30 July 1998 No Title. Rio Branco, Acre. Instituto Nacional de Coloniza o e Reforma Agrria (INCRA) 6 September 1979 Comisso Especial de Discriminao de Terras Devolutas da Unio no Estado do Acre. Cruzeiro do Sul, Acre. Jochim, Michael 1988 Optimal Foraging and the Divisi on of Labor. American Anthropologist 90(1): 130-136. Jorgenson, Jeffery 2000 Wildlife Conservation and Game Harvest by Maya Hunter in Quintana Roo, Mexico. In, John Robinson and E lizabeth Bennett, eds. Hunting for Sustainability in Tropical Fore sts. 251-266. New York: Columbia University Press. Kabeer, Naila 1994 Reversed Realities: Gender Hi erarchies in Development Thought. London: Verso. Kainer, Karen, Marianne Schmink, A. Leite, and M.J. da Silva Fadel 2003 Experiments in Forest-based De velopment in Western Amazonia. Society and Natural Resources 16: 1-18.
232 Kaplan, Hillard and Kim Hill 1985 Food Sharing Among Ache Foragers : Tests of Explanatory Hypothesis. Current Anthropology 26(2): 223-239. Kellert, Stephen and Joyce Berry 1987 Attitudes, Knowledge, and Behavi ors toward Wildlife as Affected by Gender. Wildlife Society Bulletin 15(3): 363-371. Kensinger, Kenneth 1983 On Meat and Hunting. Current Anthropology 24(1): 128-129. Kent, Susan 1993 Variability in Faunal Assemblage s: The Influence of Hunting Skill, Sharing, Dogs, and Mode of Cooking on Faunal Remains at a Sedentary Kalahari Community. Journal of Anthropological Archaeology 12: 323385. Kent, Susan 1989 Cross-cultural Perceptions of Fa rmers as Hunters and the Value of Meat. In Susan Kent, ed. Farmers as Hunters: The Implications of Sedentism. 117. Cambridge: Cambridge University Press. Leach, Melissa 1994 Rainforest Relations: Gender a nd Resource Use among the Mende of Gola, Sierra Leone. Washington: Smithsonian Institution Press. Lee, Richard and Irven DeVore, eds. 1968 Man the Hunter. Chica go: Aldine Publishing Company. Lee, Rob 2000 Impact of Subsistence Hunting in North Sulawesi, Indonesia, and Conservation Options. In, John Robi nson and Elizabeth Bennett, eds. Hunting for Sustainability in Tr opical Forests. 455-472. New York: Columbia University Press. di Leonardo, Micaela 2000 The Female World of Cards and Holidays: Women, Families, and the Work of Kinship. In Caroline Brettell and Carolyn Sargent, eds. Gender in Cross-Cultural Perspective. 379-389. Upper Saddle River, NJ: Prentice Hall, Inc. Linares, Olga 1976 Â“Garden HuntingÂ” in the Amer ican Tropics. Human Ecology 4: 331-349. Linkie, M., D. Matryr, J. Holden, A. Yanuar, A. Hartana, J. Sugardjito, and N. Leader Williams. 2003 Habitat Destruction and Poaching Th reaten the Sumatran Tiger in Kerinci Seblat National Park, Sumatra. Oryx 37: 41-48.
233 Lisansky, Judith 1985 Migrants to Amazonia: Spontaneous Civilization in th e Brazilian Frontier. Boulder: CO: Westview Press. Little, Paul 2001 Amazonia: Territorial Struggle s on Perennial Frontiers. Baltimore: The John Hopkins University Press. Loker, William 1993 The Human Ecology of Cattle Raisi ng in the Peruvian Amazon: The View from the Farm. Human Organization 52: 14-24. Mauss, Marcel 1990 The Gift: The Form and Reason for Exchange in Archaic Societies. New York: W.W. Norton. Meijaard, Erik, Douglas Sheil, Robert Na si, David Augeri, Barry Roesnbaum, Djoko Iskandar, Titiek Setyawati, Martjan La mmertink, Ike Rachmatika, Anna Wong, Tonny Soehartono, Scott Stanle y, and Timothy OÂ’Brien 2005 Life after logging: Reconcil ing Wildlife Conservation and Production Forestry in Indonesian Borne o. Jakarta, Indonesia: CIFOR. Melo, Helio 2000 O Caucho e a Seringuieira, HistÃ³rias da Amazonia, Os MistÃ©rios da Mata, Os MistÃ©rios dos RÃ©pteis e dos Pe ixes, A ExpÃªriencia do CaÃ§ador, os MistÃ©rios dos PÃ¡ssaros e Via Sacr a na AmazÃ´nia. FundaÃ§Ã£o Elias Mansour. Governo do Estado do Acre. Rio Branco, Acre. Moore, Henrietta 1988 Feminism and Anthropology. Minneapo lis: University of Minnesota Press. Moran, Emilio, ed. 1983 The Dilemma of Amazonian Developm ent. Boulder, CO: Westview Press. Moran, Emilio 1981 Developing the Amazon. Bloomi ngton: Indiana University Press. Netting, Robert 1989 Smallholders, Householders, Fr eeholders: Why the Family Farm Works Well Worldwide. In Richard Wilk, ed. The Household Economy: Reconsidering the Domestic Mode of Production. 221-234. Boulder, CO: Westview Press. Nissy, Charles 1966 Acre: An Amazonian Frontier of Br azil. PhD dissertation: University of Florida.
234 Nugent, Stephen 1993 Amazonian Caboclo Society: An Essay on Invisibility and Peasant Economy. Oxford: Berg Publishers. Olwig, Karen Fog 1998 Epilogue: Contested Homes: Home-making and the Making of Anthropology. In Nigel Rapport and A ndrew Dawson, eds. Migrants of Identity: Perceptions of Home in a World of Movement. 225-236. New York: Berg. Pallant, Julie 2003 SPSS Survival Manual: A Step by Step Guide to Data Analysis using SPSS for Windows (Versions 10 a nd 11). United Kingdom: Open University Press. Pantoja, Mariana Ciavatta 2004 Os Milton: Cem Anos de Hist Ã³ria nos Seringais. Recife: FundaÃ§Ã£o Joaquim Nabuco, Editora Masangana. Parker, Eugene, ed. 1985 The Amazon Caboclo: Historical and Contemporary Perspectives. Williamsburg: Department of Anthropology, College of William and Mary. Peres, Carlos 1991 Seed Predation of Cariniana micrantha (Lecythidacae) by Brown Capuchin Monkeys in Central Am azonia. Biotropica 23(3): 262-270. Peres, Carlos 1990 Effects of Hunting on West ern Amazonian primate communities. Biological Conservation 54: 47-59. PESACRE 1998 Sustainable Settlement in the Western Amazon: Preliminary Surveys of the SÃ£o Salvador Resettlement Project in Acre, Brazil. Proposal submitted to the W. Alton Jones Foundation. Peters, Charles 1996 Ecology and Management of NonTimber Forest Resources. World Bank Technical Paper No. 332. Washi ngton D. C.: The World Bank. Peters, Charles 1994 Sustainable Harvest of Non-tim ber Plant Resources in Tropical Moist Forest: An Ecological Primer. Biodi versity Support Program. Washington D. C.
235 Popkin, Samuel 1979 The Rational Peasant: The Polit ical Economy of Rural Society in Vietnam. Berkeley: University of California Press. Posey, Darrell and William BalÃ©e, eds. 1989 Resource Management in Amazoni a: Indigenous and Folk Strategies. Advances in Economic Botany, Volume 7. New York: The New York Botanical Garden. Raffles, Hugh 2002 In Amazonia: A Natural History. Pr inceton: Princeton University Press. Rao, Madhu and Philip McGowan 2002 Wild-meat Use, Food Security, Livelihoods, and Conservation. Conservation Biology 16(3): 580-583. Rapport, Nigel and Andrew Dawson 1998 Home and Movement: A Polemic. In Nigel Rapport and Andrew Dawson, eds. Migrants of Identity: Perceptions of Home in a World of Movement. 19 Â– 38. New York: Berg. Redford, Kent 1992 The Empty Forest. BioScience 42(6): 412-422. Redford, Kent and Christine Padoch, eds. 1992 Conservation of Neotropical Forest s: Working from Traditional Resource Use. New York: Columbia University Press. Redford, Kent and John Robinson 1987 The Game of Choice: Patterns of Indian and Colonist Hunting in the Neotropics. American An thropologist 89(3): 650-666. Robinson, John and Elizabeth Bennett 2000 Hunting for Sustainability in Tropical Forests. New York: Columbia University Press. Robinson, John and Kent Redford 1986 Body Size, Diet, and Population De nsity of Neotropical Forest Mammals. The American Naturalist 128(5): 665-680. Rocheleau, Dianne, Barbara Thomas-S later, and Esther Wangari, eds. 1996 Feminist Political Ecology: Gl obal Issues and Local Experiences. New York: Routledge. Rosenblatt, David 1992 Regional versus Federal Interests in the Development of BrazilÂ’s Amazon Region. PhD Dissertation: Universi ty of California at Berkeley.
236 Ross, Eric 1987 An Overview of Trends in Diet ary Variation from Hunter-Gatherer to Modern Capitalist Societies. In, Marv in Harris and Eric Ross, eds. Food and Evolution. 7-55. Philadelphia: Temple University Press. Ross, Eric 1978a Food Taboos, Diet, and Hunting Strategy: The Adaptation to Animals in Amazon Cultural Ecology. Current Anthropology 19(1): 1-16. Ross, Eric 1978b The Evolution of the Amazon P easantry. Journal of Latin American Studies 10(2): 193-218. Sachs, Carolyn 1996 Gendered Fields: Rural Women, Ag riculture, and Environment. Boulder, CO: Westview Press. Safa, Helen 1995 The Myth of the Male Breadwinne r: Women and Industrialization in the Caribbean. Boulder, CO: Westview Press. Salisbury, David 2002 Geography in the Jungle: Investiga ting the Utility of Local Knowledge for Natural Resource Management in the Western Amazon. MA Thesis. Gainesville: University of Florida. Scheper-Hughes, Nancy 1992 Death Without Weeping: The Vi olence of Everyday Life in Brazil. Berkeley: University of California Press. Schmink, Marianne 1999 Conceptual Framework for Gender and Community-Based Conservation. MERGE. Center for Latin American Studies. University of Florida. Schmink, Marianne and Charles Wood 1992 Contested Frontiers in Amazonia. New York: Columbia University Press. Schwartzman, Stephan 1992 Land Distribution and the Social Co sts of Frontier Development in Brazil: Social and Historical Context of Extractive Reserves. Advances in Economic Botany 9: 51-66. Scott,James 1976 The Moral Economy of the Peasant. New Haven: Yale University Press. Service, Elman 1966 The Hunters. Englewood Cliffs , New Jersey: Prentice-Hall, Inc.
237 Shaeff, Gary 2002 AvaliaÃ§Ã£o da ComercializaÃ§Ã£o. SÃ©rie Assentamento SustentÃ¡vel SÃ£o Salvador. Volume IV. Rio Branco, Acre: PESACRE. Shaw, James 1991 The Outlook for Sustainable Harvest of Wildlife in Latin America. In John Robinson and Kent Redford, eds. Neotropical Wildlife Use and Conservation. 24-34. Chicago: The University of Chicago Press. Siskind, Janet 1973 To Hunt in the Morning. New York: Oxford University Press. Skidmore, Thomas and Peter Smith 1992 Modern Latin America. New York: Oxford University Press. Slater, Candace 1994 Dance of the Dolphin: Transf ormation and Disenchantment in the Amazonian Imagination. Chicago: University of Chicago Press. Slocum, Rachael, Lori Wichhart, Dianne Rocheleau, and Barbara Thomas-Slater 1995 Power, Process and ParticipationÂ–Tools for Change. London: Intermediate Technology Publications. Smith, Nigel 1996 The Enchanted Amazon Rain Fore st: Stories from a Vanishing World. Gainesville, Florida: Universi ty Press of Florida. Smith, Nigel 1982 Rainforest Corridors: The Tran samazon Colonization Scheme. Berkeley: University of California Press. Smith, Nigel, Emanuel SerrÃ£o, Pa ulo Alvim, and Italo Falesi 1995 Amazonia: Resiliency and Dynamism of the Land and Its People. Tokyo: United Nations University Press. de Souza, Carlos Alberto Alves 2002 HistÃ³ria do Acre: Novos Temas, Nova Abordagem. Rio Branco, Acre. Speth, John 1988 Nutritional Aspects of Food Shari ng in Egalitarian Foraging Societies. In, Brenda Kennedy and Genevieve LeMoine, eds. Diet and Subsistence: Current Archaeological Perspectives. Proceedings of the Nineteenth Annual Conference of the Ar chaeological Association of the University of Calgary. 108-114. Calgary, Canada. Stearman, Allyn 1991 Making a Living in the Tropical Forest: YuquÃ Foragers in the Bolivian Amazon. Human Ecology 19(2): 245-259.
238 Stearman, Allyn 1989 YuquÃ Foragers in the Bolivian Am azon: Subsistence Strategies, Prestige, and Leadership in an Acculturating Society. Journal of Anthropological Research 45(2): 219-244. Stearman, Allyn and Kent Redford 1995 Game Management and Cultural Survival: The YuquÃ Ethnodevelopment Project in Lowland Bolivia. Oryx 29: 29-34. Stedman, Richard and Thomas Heberlein 2001 Hunting and Rural Socialization: Contingent Effects of the Rural Setting on Hunting Participation. Rural Sociology 66(4): 599-617. Tambs, Lewis 1966 Rubber, Rebels, and Rio Branco: The Conquest for the Acre. The Hispanic American Historical Review 46(3): 254-273. Thiel, Barbara 1994 Further Thoughts on Why Men Shar e Meat. Current Anthropology 35(4): 440-441. Toni, Fabiano 2003 La GestiÃ³n Forestal en los Munici pios de la Amazonia BrasileÃ±a. In, LyÃ¨s Ferroukhi, ed. La GestiÃ³n Forestal Muni cipal en AmÃ©rica Latina. 146-178. Treece, David 1990 Indigenous Peoples in Brazilian Amazonia and the Expansion of the Economic Frontier. In, David Goodm an and Anthony Hall, eds. The Future of Amazonia: Destruction or Sustainable De velopment. 264-287. London: Macmillan Press Ltd. Tsing, Anna 1993 In the Realm of the Diamond Queen. Princeton: Princeton University Press. Vickers, W. 1991 Hunting Yields and Game Co mposition Over Ten Years in a Amazon Indian Territory. In J. Robinson a nd Kent Redford, eds. Neotropical Wildlife Use and Conservation. 53-81. Chicago: The University of Chicago Press. Wadley, Reed, Carol Colfer, and Ian Hood 1997 Hunting Primates and Managing Fore sts: The Case of Iban Forest Farmers in Indonesian Borneo. Human Ecology 25(2): 243-271. Wagley, Charles 1976  Amazon Town: A Study of Ma n in the Tropics. London: Oxford University Press.
239 Wallace, Richard 2004 The Effects of Wealth and Mark ets on Rubber Tapper Use and Knowledge of Forest Resources in Acre, Braz il. PhD Dissertation, University of Florida. Weinstein, Barbara 1985 Persistence of Caboclo Culture in the Amazon: The Impact of the Rubber Trade, 1850-1920. In Eugene Parker, ed. The Amazon Caboclo: Historical and Contemporary Perspectives. 89114. Williamsburg: Department of Anthropology, College of William and Mary. Wilk, Richard 1997 Household Ecology: Economic Ch ange and Domestic Life among the Kekchi Maya in Belize. Dekalb, IL: Northern Illinois University Press. Wilkie, D. S. and J. F. Carpenter 1999 Bushmeat Hunting in the Congo Basin: An Assessment of Impacts and Options for Mitigation. Biodiversity and Conservation 8: 927-955. Wolf, Eric 1982 Europe and the People without Hist ory. Berkeley: Univer sity of California Press. Wolff, Cristina Scheibe 1999 Mulheres da Floresta: Uma Hist ria Alto Juru, Acre (1890). So Paulo, Brazil: Editora Hucitec. The World Conservation Union (IUCN). http ://www.redlist.org/. Accessed 8 June 2005. Wunderle Jr., J. M. 1997 The Role of Animal Seed Di spersal in Accelerating Native Forest Regeneration on Degraded Tropi cal Lands. Forest Ecology and Management 99: 223-235. Yost, J. and P. Kelly 1983 Shotguns, Blowguns and Spears: The Analysis of Technological Efficiency. In R. Hames and W. Vi ckers, eds. Adaptive Responses of Native Amazonians. 189-224. New York: Academic. Zoneamento Ecolgico-Econmico (ZEE) do Acre 2000a Aspectos Socioeconmicos e Ocupao Territorial. Volume II. Acre. Governo do Estado do Acre. Pr ogram Estadual de Zoneamento Ecolgico-Econmico do Estado do Acre. Zoneamento Ecolgico-Econmico (ZEE) do Acre 2000b Indicativos para Gest ao Territorial do Acre. Volume III. Acre. Governo do Estado do Acre. Program Estadual de Zoneamento Ecolgico-Econmico do Estado do Acre.
240 BIOGRAPHICAL SKETCH Eric Minzenberg was born in Los Angele s, California. He attended St. Bernard High School in Playa Del Rey, CA, graduati ng in 1981. In 1986, he graduated with a Bachelor of Science degree from the Universi ty of California at Berkeley, majoring in forestry and natural resource management. A few years after the completion of his BS degree, he opened a garden design, landscap ing and maintenance business in the San Francisco Bay area. After four years of su ccessfully operating his own small business, Eric entered the Peace Corps in 1995 spending 2+ years working with rural communities promoting sustainable conservation and deve lopment projects in the western tropical forests of Ecuador. Upon his return to the United States, Eric entered the masterÂ’s program in Latin American Studies at San Diego State Univ ersity. In 2000, he graduated, magnum cum laude, with a Master of Arts in Latin American Studies. Later that same year, he entered the doctoral program in anthropology at the Un iversity of Florida. Eric pursued an interdisciplinary course of studies with a focus on peasant and gender studies in the lowland tropics of South America integrati ng anthropology curriculum with the tropical conservation and development program housed in the Center for Latin American Studies at UF. In the future, Eric plans to continue his work with peasant communities in western Acre.