<%BANNER%>

Factors Conditioning the Development of a Community Forestry Coalition in Western Amazonia, Brazil


PAGE 1

FACTORS CONDITIONING THE DEVELOPMENT OF A COMMUNITY FORESTRY COALITION IN WESTERN AMAZONIA, BRAZIL By FRANKLIN PANIAGUA A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS UNIVERSITY OF FLORIDA 2005

PAGE 2

Copyright 2005 by Franklin Paniagua

PAGE 3

In memory of Lubin Villalobos and Guillermo Marin

PAGE 4

ACKNOWLEDGMENTS I feel very fortunate to have the chance to thank the people whose love and care have made this document possible. My parents, brother and sister, were my strongest supporters. The teaching and guidance of my committee (Marianne Schmink, Karen Kainer and Tom Ankersen) gave me the opportunity to learn and grow both personally and professionally. I am deeply grateful for Mariannes extraordinary patience and faith in my work. I thank Jon Dain and Bob Buchbacher for their insightful comments and for becoming de facto committee members. I would like to acknowledge Magna Cunha (in Brazil) for her openness and dedication to work. Magna was the person who suggested the Grupo de Produtores Florestais as a research topic. Through her, a network of friends and colleagues facilitated my work. Among them are: Nivia Marcondes, Pedro Bruzzi, Luis Meneses, Renata Texeira and Carlos Vicente; the community forestry leaders from Peixoto, Cachoeira, Porto Dias and Cautrio. Finally, heartfelt thanks go to my friends and family-of-practice at UF: Hannah, Margarita, Wanda, Myrna (CLAS-TCD); Lin (Editor-in-Chief), Alfredo, Claudia (Senior Editor), Meredith, Wendy-Lin (Excited Editor), Tita (y Pablo), Ana Cristina, Richard, John, Christine, Amy, Maria, Emilio, Daniel and Adriana. In Costa Rica: Roberto, Sergio, Luis, Carlos, Llamil, Gaby, Ale, Andrea, Rolo and nutelly to Vivi. iv

PAGE 5

TABLE OF CONTENTS Page ACKNOWLEDGMENTS.......................................................................................iv LIST OF TABLES...............................................................................................viii LIST OF FIGURES...............................................................................................ix LIST OF ABBREVIATIONS..................................................................................x ABSTRACT..........................................................................................................xi CHAPTER 1 INTRODUCTION...........................................................................................1 Sociopolitical Context.....................................................................................4 Natural Resource Management Institutions.............................................7 History of the GPF.................................................................................10 Research Design and Methodology.............................................................16 Research Design...................................................................................16 Hypothesis.............................................................................................18 Sites.......................................................................................................18 Instruments............................................................................................19 Fieldwork...............................................................................................20 Data Analysis.........................................................................................21 Synthesis.....................................................................................................21 2 ANALYTICAL FRAMEWORK......................................................................23 Natural Resources Management Coalitions.................................................24 Approaches to Analyze Community Forestry Coalitions..............................26 Framework to Analyze Community Forestry Coalitions................................31 Internal Factors......................................................................................34 External Factors.....................................................................................42 Design Factors.......................................................................................49 Integrating the Analysis.........................................................................53 Analytical Framework Limitations..........................................................58 Synthesis.....................................................................................................60 v

PAGE 6

3 INTERNAL FACTORS.................................................................................61 The GPF Coalition Members........................................................................62 Porto Dias Project..................................................................................65 Centro de Trabalhadores da Amaznia...........................................67 Associao de Seringueiros de Porto Dias.....................................69 Peixoto Project......................................................................................70 Empresa Brasiliera de Pesquisa Agrcola.......................................73 Association of Rural Producers of Peixoto......................................74 Cachoeira Project..................................................................................75 Associao de Moradores e Produtores do PAE Chico Mendes....78 Cachoeira Projects External Partners.............................................79 Cautrio Project.....................................................................................82 Ao Ecolgica Guapor.................................................................84 Associao dos Seringueiros do Vale do Guapor.........................84 Comparative Analysis..................................................................................86 Participation...........................................................................................87 Leadership.............................................................................................89 Management Capacity...........................................................................92 Project Relationships.............................................................................96 Projects Training Focus........................................................................99 Projects Appropriation by the Community Associations......................107 Synthesis...................................................................................................113 4 EXTERNAL FACTORS..............................................................................117 Law and Policies........................................................................................119 Land Tenure........................................................................................119 Forest Management.............................................................................122 Social Organization..............................................................................124 Markets and Community Timber Projects..................................................127 External Stakeholders................................................................................129 Federal Government............................................................................130 Acre State Government.......................................................................131 World Wildlife Fund..............................................................................135 Furniture Company AVER...................................................................136 External Stakeholder Map...................................................................136 Cross-scale Interactions.............................................................................137 Extractive Reserves.............................................................................138 Political Dilemma.................................................................................139 Synthesis...................................................................................................141 vi

PAGE 7

5 DESIGN FACTORS...................................................................................144 Concept and Design...................................................................................145 Perceptions about the GPF........................................................................148 Partner Organizations..........................................................................148 Community Associations.....................................................................150 External Stakeholders..........................................................................151 Perception Matrix.................................................................................152 Process Design..........................................................................................154 Innovation............................................................................................155 The GPF Timetable.............................................................................155 Innovation Process..............................................................................157 Synthesis...................................................................................................162 6 CONCLUSIONS........................................................................................164 Results as Related to the Hypothesis........................................................165 Hypothesis.................................................................................................167 Methods.....................................................................................................168 Emerging Issues........................................................................................168 Economic and Cultural Transition........................................................169 Community Enterprises.......................................................................171 Future of the GPF......................................................................................173 LIST OF REFERENCES..................................................................................175 BIOGRAPHICAL SKETCH...............................................................................183 vii

PAGE 8

LIST OF TABLES Table Page 1.1: Historical Timeline.........................................................................................6 2.1: NRM Collaborative Organization Analytic Variables....................................28 2.2: CFC Analytical Framework..........................................................................33 2.3: Elements of the Innovation Process............................................................52 2.4: Integrating Matrix.........................................................................................54 3.1: Comparative Analysis Variables..................................................................87 3.2: Task Distribution by Projects.....................................................................108 3.3: Project Appropriation by the CAs...............................................................112 5.1: GPF Perception Matrix..............................................................................152 5.2: GPF Coalition Timetable............................................................................156 6.1: Integrating Matrix with Results...................................................................166 viii

PAGE 9

LIST OF FIGURES Figure Page 1.1: GPF Problem Tree......................................................................................13 1.2: Drivers of the GPF Initiative.........................................................................14 2.1: Hierarchical System of Factors Conditioning the GPF.................................34 2.2: CFC Internal Agents....................................................................................35 2.3: Timber Project Community of Practice........................................................37 2.4: Communities of Practice Model...................................................................38 2.5: CFC External Agents...................................................................................43 2.6: Historic Trajectory Modeled on the Adaptive Cycle.....................................46 3.1: Stakeholder Map.........................................................................................64 3.2: Task Distribution by Project.......................................................................109 3.3: Task Distribution Overall............................................................................111 4.1: Policies Supporting Timber Extraction and the GPF Coalition...................127 4.2: Community Timber External Stakeholders................................................137 4.3: Hierarchical Web of Innovations................................................................140 5.1: GPF Coalition Structure.............................................................................146 5.2: GPF Functional Structure..........................................................................160 ix

PAGE 10

LIST OF ABBREVIATIONS CA Community Association CFC Community Forestry Coalition CIFOR Center for International Forestry Research CNS National Rubber Tappers Council CTA Amazonian Workers Center EMBRAPA Brazilian Institute of Agricultural Research FSC Forest Stewardship Council GPF Forestry Producers Group IBAMA Brazilian Institute of Environment IMAC Environmental Institute of Acre INCRA National Institute of Colonization and Agrarian Reform NRM Natural Resource Management NTFP Non-Timber Forest Products PESACRE The Acre Group for Agroforestry Research and Extension PO Partner Organization SEATER Secretariat of Rural Extension and Technical Assistance SEFE Secretariat of Forestry and Extractivism WWF World Wildlife Fund x

PAGE 11

Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Arts FACTORS CONDITIONING THE DEVELOPMENT OF A COMMUNITY FORESTRY COALITION IN WESTERN AMAZONIA, BRAZIL By Franklin Paniagua August 2005 Chair: Marianne Schmink Department: Latin American Studies My study examines how forest communities might be linked to markets on their own terms, analyzing particularly the development of a community forestry coalition in western Amazonia, Brazil. I built on growing interest in pro-grassroots coalitions that integrate organized communities, NGOs, government and international agencies as a strategy to successfully link communities to markets. I used a multi-layered assessment that integrates internal, external, and design factors conditioning the development of a coalition among a set of four community timber projects from the states of Acre and Rondnia, in the Brazilian Amazon region. Developing a second-degree organization (such as the coalition) represents a self-organized response to the social and organizational challenges of executing community timber extraction projects. A major challenge was to try to meet market demands while responding to local social and political constraints. The coalition followed an NGO-project xi

PAGE 12

pattern instead of a collaborative organization pattern, creating a gap between the original idea of the coalition and the actual process. My findings support the need to re-adjust the role that social learning plays in building an autonomous alliance among community forestry projects, and contributes to the development of instruments for analyzing collaborative organizational innovations. Because this coalition was still at a very early stage compared to cases in other regions, such as southern Mexicos Sociedad de Ejidos, it is still difficult to know whether the coalition is a secure structure for equitable market integration. xii

PAGE 13

CHAPTER 1 INTRODUCTION How can communities be linked to the market on their own terms? (Schmink, 2004), This powerful question lingers as an emerging challenge for community forestry. Responses from the tropics reflect changing trends in policy that have opened a window of opportunity to build more equitable relationships between the rural poor and the market. Among the changing trends: almost 25% of the global forest estate is owned or administered by indigenous and rural communities (White & Martin, 2002); democratization is stimulating an increase in local peoples participation in decision making; and global markets are increasing the demand for tropical timber with a premium on certified social and ecological products 1 (Scherr et al., 2002). Nevertheless, these trends have diverse effects on the tropics as regions undergo transformation in different ways and at different paces. Some authors suggest that a key factor in successful market strategies is the development of pro-grassroots coalitions that integrate organized communities, NGOs, government and international agencies (Alcorn & Toledo, 1998; Brown & Rosendo, 2000; Fisher et al., 2001). 1 A certified product is one for which certain production characteristics (conditions, procedures, byproducts) are overviewed by a third party who guarantees to the consumers of the product, that certain production standards are maintained. In the case of timber, the system used is the Forest Stewardship Council (FSC). This is an international body that establishes the standards to certify wood based on environmental and social considerations. FSC has a body of certifiers, conducts the assessments, and provides the credential to the production unit. 1

PAGE 14

2 The state of Acre in Brazil has undergone intense transformation over the past 15 years since the initial struggle of the seringueiros (rubber tappers) (led by Chico Mendes and others) drew international attention, and watershed major political changes encouraged forest-centered development (Kainer et al., 2003). The result of these policy struggles was the creation of the Extractive Reserves (RESEX) and the Agro-Extractivist Projects (PAEs) that secured land tenure rights for the rubber-tapper communities. Currently, the Forest Government, with national and international support, is engaged in generating a suitable policy environment for sustainable community-centered development. In an effort to diversify traditional non-timber forest products (NTFPs) such as Brazil nut and rubber, a number of agencies have set up community timber extraction projects. Each project paired an NGO or government agency with a local community association. Until the summer of 2002, these projects had had partial success, experiencing difficulties consolidating production and identifying stable markets for their timber products. The projects did not have the capacity to produce sufficient quantity or adequate quality for the national and international markets interested in sustainable community timber. In response to this situation, several groups proposed the creation of a consortium of community forestry producers (Grupo de Produtores Florestais Comunitarios, GPF). This second-degree organization was to act as a platform for joint marketing of the projects timber. According to the proposal, the community associations participating in the timber projects would lead the coalition. The aggregated volume of wood from the different projects would

PAGE 15

3 make it more attractive to sustainable timber buyers from cities like So Paulo and Rio de Janeiro in southern Brazil. Other regions with community forestry development, such as Central America (Segura et al., 1997), Nepal (Shrestha et al., 1998) and especially southern Mexico (Bray et al., 2003) have experienced successful second-degree organization development 2 These larger-scale inter-community innovations have, in turn, secured and strengthened timber production throughout their areas of influence (Colchester et al., 2003). Protected areas and watershed management are also fields where this phenomenon of stakeholder collaboration has grown in recent years (Wondolleck & Yaffee, 2000) The capacity to build a regional coalition and to have it render positive feedback to the local timber activity is dependent on a complex set of factors. These factors range from the social capital and organizational capacity of the potential member groups, to state and federal policies that shape the technical, and the organizational context. The purpose of my study was to untangle the web of factors that condition the success of this organizational innovation. I used a hierarchical approach that analyzed three categories of factors: internal, external and design. The analysis of these factors was then integrated through a series of matrices, to contrast and compare the effects of each category in the coalition development process. The information resulting from this research is meant to contribute to the ongoing process of providing a picture of the emergent 2 A second-degree organization is a formally established group made up of other organizations. Typically the member organizations will be of a lower scale of order. For example, a federation is a second-degree organization of associations. Associations could come from specific communities while the federation would represent an entire region.

PAGE 16

4 constraints of the GPF initiative, and to give insights towards alternative paths for overcoming these constraints. Sociopolitical Context The state of Acre has an area of 153.149 km 2 and a population of approximately 600,595 in 2000. The administrative structure of Acre is divided into three major provinces, 5 regions, and 22 municipalities. Forest conversion in the state has been fast, grouped and dominated by the establishment of cattle ranches. Yet by 2001, Acre still retained 89% of its original forest cover. In the 1880s Acre was colonized as part of the first rubber boom to supply rubber to ports in Manaus and Belm. This boom was instigated by both Charles Goodyears discovery of the vulcanization process, through which rubber is made chemically stable by exposure to sulfur and heat (1839); and demand for tires for bicycles and cars. Between the years 1840 and 1860, the price of rubber nearly tripled (Bakx 1988), at the height of the rubber boom in 1899, the territory of Acre supplied 60% of the rubber from the Amazon (Hall 1997). During this time, rubber barons and laborers from northeastern Brazil heavily colonized Acre. In 1903, Acre was officially annexed to Brazil after a war called the Acre Revolution. The fall of the rubber trade occurred around 1910, when rubber trees grown in Southeast Asia began to produce. Increased production in Asia caused rubber prices to fall in Brazil and led to the eventual collapse of Brazils export market (Hall 1997). A smaller, second rubber boom in Brazil during World War II when the Allies lost control of the Asian rubber plantations. In 1942, the United States and Brazil signed a five-year agreement to support increased rubber production

PAGE 17

5 in Brazil and ensure their access to Brazilian rubber (Schmink and Wood 1992), which revived the traditional rubber tapping system known as the seringal. Because of the declining rubber market between the 1950s and 70s, many rubber estate owners abandoned their landholdings, leaving local rubber tappers to begin tapping rubber for subsistence and trade purposes (Kainer et al. 2003). With another influx of land speculators in the 1970s, under the support of Brazils military dictatorship, many rubber tappers were thrown off their lands. In response, they began to organize, under the leadership of such figures as Chico Mendes, and fight for their land. The creation of extractive reserves in 1989 was an outcome of this movement, following the fall of Brazils military dictatorship and the support of national and international environmental groups, who saw extractive reserves as an ideal grassroots conservation and development initiative. Table 1.1 summarizes the main historical events that have molded the current socioeconomic and political situation in Acre. The table pairs the events from Acre with those at the national level (Brazil). The table follows the rubber boom-and-bust cycle from the mid eighteen hundreds to the mid nineteen eighties. Starting in the late eighties, signs of a re-organization process emerged, with the transformation of the old rubber estates into common property extractive reserves. This reorganization phase continued through the nineties and to the present. The social and political struggles of the 1980s paved the way for most of the resource management institutions were created. The degree of violent

PAGE 18

6 conflict, especially in the Alto Acre region of Xapur and Brasilia, was significant although not as extreme as in the eastern Amazonian state of Par. Wilson Pinheiro and later Chico Mendes were among the rubber tapper leaders killed at the hands of cattle ranchers. Table 1.1: Historical Timeline Year Acre Brazil 1852 Annexed to the province of Amazonas 1877-1880 Large immigration from the northeast to the Amazon Aviamento system Drought in the northeast 1898 Bolivia creates the Departamento de Acre 1903 Acre Revolution: armed conflict between Brazil and Bolivia that resulted in the creation of the State of Acre as part of the Brazilian Federation. Prior to the conflict the territory as part of the Bolivian Republic 1904 Petropolis treaty, creation of the state of Acre of Brazil. 1910-20 Decline of the rubber economy due to competititon from Southeast Asia rubber plantations. 1920-45 Dormant" economic period 1945-49 Resurgence of the rubber economy and the Seringal system Treaty between Brazil and USA governments to supply rubber to allied forces during WWII 1950-70s Building of rubber tappers social organization; deforestation; development Military dictatorship in Brazil (1964-1984). Land struggles in Amazonia 1980 Land tenure conflicts in the southwest region of the state Gradual democratization 1989 First extractive reserves Legalization of extractive reserves 1998 Jorge Vianas Forest Government is elected 2001 BR-317 pavement by the state government) Avana Brazil: infrastructure for the Amazon It is interesting to note that while the same violence that affected Acre happened in other places (Rondnia Par) the result was significantly different. A number of factors transformed the path of violence in an unexpected way for Acrean history. Instead of cattle ranchers expelling other social groups and taking over their land, they were defeated through the support of the government.

PAGE 19

7 Cattle expansion was halted in Acre through the creation of extractive reserves (RESEX). Although originally presented as an alternative means for rubber tappers to secure land tenure, extractive reserves became an innovative conservation and development strategy that was championed internationally. Along with the leadership and organizational capacity of the rubber tapper movement, the international support for the RESEX was key in changing attitudes of the Brazilian government. With the creation of the RESEX, violence de-escalated, and an opportunity for reorganization emerged. The success of this social movement led Jorge Viana to win the gubernatorial race in 1998. Under the banner of the Governo da Foresta (the forest government), Vianas progressive government has institutionalized most of the ideas that flourished during the conflict years. Extractivist activities and community forest management are strongly supported by the state government. Currently, there are three extractive reserves in Acre, including the Chico Mendes Extractive Reserve, which is the largest in Brazil, and eight agro-extractivist projects (PAEs). The success of extractive reserves has come under scrutiny in recent years, due to increased deforestation within their boundaries, largely due to the fact that extraction of non-timber forest products, alone, is often not sufficient for making ends meet. Cattle ranching and logging have increased in Acre, and with the paving of BR-317 in 2001, supported by Acres Governo da Floresta, the landscape will continue to change. Natural Resource Management Institutions In Acre, natural resources are managed through a complex web of individual, communal, state and federal level institutions. At the base of this

PAGE 20

8 management lies the property rights system. RESEX and PAE (Agro-extractivist settlements) land is owned by the Brazilian government at the national or state level, and granted to a community as a concession, normally for thirty years. Since extractive reserves are managed by IBAMA, the Brazilian environmental protection agency, and PAEs by INCRA (Land Reform Agency), communities are required by IBAMA and by INCRA to create plans for the sustainable management of natural resources within the reserves. Property rights are held by individuals, as well as communally in extractive reserves, with rubber trails providing the base for private access to natural resources. As rubber prices decline, and the value of harvesting Brazil nuts increases (along with the illegal harvest of timber or conversion of forested land to pasture within the reserves), the allocation, use and control of forest resources within this land tenure system is bound to change. Land cover change decisions promoted by landowners must be approved both by IBAMA and IMAC, the federal and state environmental agencies. Forest extraction goes through a particular process called licenciamento, or licensing, which requires the development of a Management Plan that must be approved by a forest engineer. Timber, once harvested, requires special transportation permits and for the producer to pay a series of state and federal taxes. The cost and duration of all these legal procedures is beyond the reach of most small landholders, who then turn to illegal logging practices. In extractive reserves the alternative has been to obtain this legal authorization as a group, through the development of collective timber extraction projects. The banner community

PAGE 21

9 timber extraction project is in the community of Cachoeira, just outside of the city of Xapuri, in the Alto Acre Region. Illegal logging is controlled by both IMAC and IBAMA, but the actual legal investigation corresponds to the Ministrio Pblico (Attorney General). Illegal logging cases do not commonly come up. Of the few cases that are brought to court, even fewer get sentenced because in most cases there is not sufficient evidence. A particular problem at the Ministrio Pblico is the issue of the reserva legal. This is a minimum percentage of property that must remain forested, according to federal land tenure laws. In the Amazon region 80% of each private property must remain in forest cover (90% for RESEX). This limitation has also become an incentive for illegal logging. A significant alignment of policy actions has been occurring in the years since the transformation generated by the rubber tapper social movement and emergence of extractive reserves. These elements are The zoneamento econmico-ecolgico This state-wide economic and environmental planning strategy brought together a large set of information sources into a comprehensive document on the status of the environment in Acre. This study represents a significant baseline for tracking changes in land-use and establishing sustainability thresholds. IMAC this state environmental agency represents a significant source of information and decision making that was less active before the establishment of the governo da floresta The consolidation of UFAC, the Federal University of Acre as a source of environmental research and training of professionals. UFAC researchers have been key in the development of the MAP (Madre de Dios, Peru; Acre, Brazil and Pando, Bolivia) network that later became a large-scale interstate and international effort The development of learning networks between research organizations like UFAC and EMBRAPA (Brazilian Institute of Agricultural Research) with development NGOs (PESACRE, the Acre Group for Agroforestry Research and Extension; CTA, Amazonia Workers Center), local organizations and government agencies on issues such as fire management, forestry and

PAGE 22

10 agro-forestry. These networks are also linked to international as well as national research centers and funding sources. History of the GPF The emergence of the Grupo de Produtores Florestais (GPF) must be understood within the broader process of community timber extraction development in Western Amazonia.The fifteen-year period since the establishment of extractive reserves (RESEX) and agro-extractivist settlements (PAEs) has been fruitful but challenging at the same time. The struggle of the rubber tappers social movement has rendered not only the necessary legal institutions to secure their land tenure, but also a significant improvement in the political role and social stature of the extractivist communities in Amazonia and in Brazil, in general. This has resulted in small but significant improvements in the quality of life of the rubber tapper families. However, the opportunity provided by secure land tenure came with its own set of problems, some of them due to changes in the broader political and economic context. Others were unintended consequences of the social movements success. Community timber extraction can be considered a bitter fruit of this process. It is a fruit because the regained land rights allowed extractive communities to take advantage of this highly valuable resource. It is bitter because the original intention of RESEXfar from cutting treeswas to conserve the forest for traditional extractive activities (e.g., Brazil nuts, rubber). The failure of traditional extractive products to sustain a viable livelihood motivated the search for alternative activities and land uses, among them cattle ranching and timber extraction.

PAGE 23

11 Cattle ranching came almost naturally to rubber tapper families, as it already formed part of their livelihood strategies. Surplus from other activities would be invested in cattle as a secure savings. Increasing the number of head of cattle per family requires changing land cover from forest into pasture, a process that has increased in recent years (Gomez, 2001; Salisbury, 2002). Although legally each family has a right to clear 10% of their land for pasture and/or agriculture, the increased economic role of cattle drives some families to deforest beyond that threshold. The cumulative effect of larger numbers of families following this cattle strategy is uncertain, but it does call into question the existence of RESEX as a means of maintaining forest cover, since forest and pastures are mutually exclusive land covers. On the other hand, timberif sustainably harvestedcould be complementary to the RESEX model. However, it would require a major institutional and cultural change in Western Amazonia. The process of directing this change was spearheaded by NGOs that had played a supporting role to the rubber tappers movement through the eighties and nineties. These groups transformed themselves into forest managers to lead the development of experimental community timber projects. These experiments began around 1996-97. The projects struggled with a variety of obstacles, including direct opposition from significant parts of the rubber tapper community (Kainer et al., 2003). Interest in the activity grew with the election of Jorge Vianatrained as a forest engineeras governor of the state of Acre in 1998, and has increased significantly with his reelection in 2002. Expectations regarding the benefits of

PAGE 24

12 timber extraction for extractivist communities have increased to the point of becoming a mainstream idea for the future of RESEX and PAE communities. Economically, however, timber extraction has not shown itself to be as successful for the rubber tapper families as they had expected. Until 2002, most of the projects were far from generating stable profits or handing over complete operational control to the communities (Stone, 2003). The cultural, political, and institutional transformations required for timber extraction to serve as a viable economic alternative for RESEX are still in their developmental stages. Timber project leaders recognized this situation and proposed an articulation among the different timber projects as a strategy to cope with the increasing complexities of the timber harvest activities (CTA, 2001) The timber project leaders presented this proposal for the Central de Associaes de Produtores Florestais do Acre, a union of the timber producing associations, at the first meeting of community timber projects of Acre and Rondnia in December of 2001. The justification for the idea was presented in the manner described in Figure1.1. The Central, later called Grupo de Produtores Florestais (GPF), would vertically integrate the current community timber projects to facilitate timber sales. This integration would alleviate the different marketing problems experienced by some of the projects, as well as channel the increase in production from new or existing community timber projects. A vertical integration of projects would take advantage of larger volumes of woods to link regional, national, and international markets. Control of the Central would remain in the hands of the community-based associations,

PAGE 25

13 guaranteeing that profits would return directly to improve the livelihood of the extractivist families. Marketing limitations ofcommunity timberprojects GPFCoalition Better livelihoods Selling Timber Project Success Figure1.1: GPF Problem Tree The Amazon Workers Center (CTA) 3 proponents of the GPF ideadetailed the reasons and expectations of the coalition, in the manner illustrated in Figure 1.2. The main motivation for the proposal came from the problems projects experienced in selling their timber. These problems were related to issues such as: (1) the legal limitations on community-based associations to sell products 4 (2) the imminent reduction in funding from donors, and (3) the general difficulty of demonstrating significant success in achieving the groups sustainability goal. 3 Centro dos Trabalhadores da Amazonia is an NGO based in Rio Branco, the capital of the state of Acre. Created in 1981and with a long history of accompanying the rubber tappers social movement, the group originally conducted education programs. In 1996, the organization started the first Brazilian Amazonia community timber project in the agro-extractivist settlement (PAE) of Porto Dias. 4 Community-based associations are legally limited to conduct social activities, not commercial enterprises. For this reason timber sales can only be signed off by individual association members. Normally it is the president of the association who is in charge of signing the sale.

PAGE 26

14 Marketing limitationsof community timberprojects GPFCoalition TimberCertification Sustainability? Legal Restrictionsto sell timber byCommunity Assoc. Reduction Funding Economic Ecological Social Need of largervolume for Nat.and Int.Markets As a coop or abusiness the GPFcan sale timber EntrepostoSelling post Figure 1.2: Drivers of the GPF Initiative While these other problems were independent of the issue of marketing, solving the problem of selling the timber would bring about a solution to them, according to the GPF proposals supporters. The GPF was envisioned mainly as a selling post (entreposto da venda) that would receive timber from all the projects and be responsible for marketing and selling. According to the plan, the GPF would be organized as a cooperative of the different community associations and, in this way, solve the legal limitations impeding the sale of products by the social organizations. The entreposto would manage more volume than the individual projects could, and would therefore be able to sell to larger markets and take advantage of the FSC forest certification promoted by WWF-Brazil 5 (CTA, 2001) International and national buyers interested in 5 WWF: World Wildlife Fund Brazil is a Brazilian NGO, part of an international network of WWF. In the case of Acre and Rondnia, it has supported the Cautrio community timber project in Rondnia, and has promoted the certification of the community timber projects and the creation of the GPF.

PAGE 27

15 certified wood would be interested in obtaining it from the community only if they could guarantee a significant and stable volume. In 2001, a large group of individuals involved in community timber management and forestry policy in Acre participated in an exchange with the ejidos of Quintana Roo, Mexico. This exchange provided participants with a vivid example of a successful coalition of community forestry enterprises. The GPF was modeled on the example of the Asociacin de Ejidos 6 It was this closer contact with the Mexican ejidos that helped to articulate the question of whether Western Amazonia had the necessary conditions to develop a similar coalition. Despite significant similarities such as the pre-existing extractivist culture, there were some equally important differences between southern Mexico and western Amazonia. One of the most important differences was the strong support provided by the Mexican Government and some international agencies through the Plan Piloto Forestal, for over fifteen years. Another difference was the strong social capital shared by Ejido communities, that strengthened local leadership and cultural identity. Recognizing these differences, the proponents of the coalition had the challenge to adapt the Sociedad de Ejidos model to the situation in western Amazonia. This research represented an effort to inform the GPF proponents about the existing factors conditioning the coalition initiative and the alternative paths of development. For this reason it was designed and conducted 6 This association of ejidos is presented and discussed in Chapter Two.

PAGE 28

16 collaboratively with the GPF proponents, as will be detailed in the following section on the design and methodology of the study. Research Design and Methodology Research interest in collaboration, as a catalyst for change in NRM, is relatively recent. Tools and approaches to study collaboration are still in the making. Researchers in fields such as communal property or political ecology have been more interested in the results or impacts of collaboration than in looking at its process (Alcorn & Royo, 1999; Mayer, 2002; Ostrom, 1990). Studies concerned with the way collaboration is developed are scarce. Most of the analysis corresponds to development and conservation organizations such as the United States Agency for International Development (USAID), Canadas International Development Research Center (IDRC), or the United Nations Food and Agriculture Organizations Forest, Trees and People Program (FTPP). Different conceptual and applied publications have resulted from their field projects (Buckles, 1999; USAID, 2002). Research Design The objective of this research is to yield insights for improving multi-stakeholder collaboration platforms, similar to that attempted in the Brazilian GPF initiative, the focal case of this thesis. The idea for the research developed from conversations with a forestry researcher involved in several of the community timber projects and also in the creation of the GPF. An immediate practical need was identified to understand the conditions present for the GPF, both within the groups and in the policy environment. The research would provide a deeper understanding of these factors.

PAGE 29

17 Based on literature review I organized these factors into three groups: internal, external and design 7 Internal factors refer to the organizational conditions (leadership, management capacity, participation) and collaborative practices at each of the community forestry projects. A total of four projects were considered in the research. Each community forestry project consisted of a community association (CA) and a partner organization (PO). Partner organizations were both governmental and NGO types. External factors refer to the policy and market environment in which the proposed initiative is embedded. These factors are separated into laws, policies and institutional behaviors. State and national (federal) laws and policies are of special interest, as well as the funding policies of international aid agencies and NGOs. Design factors refer to the conceptualization of the proposed coalition, and the way it has been adopted by the different timber communities and their NGO partners. The analysis of these factors was intended to provide a picture of the potential success of multi-stakeholder coalitions. To establish a reference for what would constitute a successful coalition process, I developed different indicators derived from a selection of similar studies. The main source for these indicators was the articles on The Union of Forest Ejidos (Union de Ejidos Forestales) of southern Mexico by Santos and Bray in their respective publications (Bray et al., 2003; Santos et al., 1998). Other studies done on sustainable businesses in the Amazon basin (A. Anderson & Clay, 2001; Clay, 2001). 7 A detailed description of the research framework is provided in Chapter 2.

PAGE 30

18 Hypothesis I hypothesized that my results would reveal a scenario in which both the design and external factors would be positive for the emergence of the coalition. By contrast, the internal factors would be the source of major constraints. I developed this hypothesis based primarily on the literature that suggested internal factors such as management capacity of the local groups and social capital as the weak factors leading to the failure of community timber projects, in analyses conducted by CIFOR and WRI-Forest Trends (Colchester et al., 2003; Scherr et al., 2002). I also based it on my personal experience working in the development of natural resource management coalitions. I had worked in coalition efforts around watersheds and protected areas in several countries of Latin America, such as Ecuador, Costa Rica, Mexico, Panama and Peru. Sites Four sites were defined for the study, corresponding to the four timber projects involved in the first GPF meeting held in December 2001. Three of them were in the state of Acre (Porto Dias, Peixoto and Cachoeira) and one was in the neighboring state of Rondnia (Cautrio). I visited these four sites at least once, during the five-week period defined for fieldwork. At the time of my arrival, a new person from outside the region was hired to spearhead the GPF initiative. She participated in the definition of the research plan and joined me for the field visits and interviews. We worked as a team, conducting all the interviews and analyzing each session together. However, the final analysis was done separately.

PAGE 31

19 Instruments Three open-ended questionnaires were prepared, as well as research guidelines for the analysis of relevant laws and policies. We were to interview the NGO or public agency project directors (or coordinators) and at least two members of community association boards from each of the four projects. Where available, more project staff and community members would be interviewed, as well as other relevant individuals not directly involved with the timber projects. The first questionnaire was directed to the group of external stakeholders that included Acre state government officials, NGO coordinators and other knowledgeable individuals not directly involved in the timber projects. The second was geared towards the PO project staff, and the third to the community association leaders. The first one focused on the perceptions of the government officers on three topics: community forestry in Acre, issues and future of current community timber projects (including the idea of the GPF), and the effectiveness of timber extraction as a conservation policy. These topics were laid out in a set of twenty questions. The other two questionnaires were oriented to sketch a profile of each organization (NGO, government agency or community association) involved in the project. These profiles included general basic information about the groups (foundation, membership, organization) and a particular look at their relationship with the partner organization in the timber project. This meant asking the association about the NGOs and vice versa. Both groups were asked about the management capacity of the associations and their degree of appropriation of the timber project. In this context,

PAGE 32

20 appropriation refers to the degree of control and operational autonomy the local group had over the timber activity. Perception of the GPF proposal was a crosscutting topic in all questionnaires. These two instruments had forty-two items for discussion with each respondent. Fieldwork Data gathering included other sources such as laws, policy briefs and organization bylaws, project reports and evaluations. Fieldwork extended from June 24 to August 16, 2002. The first four weeks were devoted to the initial contacts and research strategy design. From week four to week eight, interviews were conducted and documents were collected. A total of forty interviews were carried out, including sixteen community association leaders, fourteen NGO or agency officials from the timber projects, seven state or federal officials, and three researchers from the local Federal University of Acre (UFAC). Related personnel from NGO and agencies were selected according to their involvement in the timber projects. All project coordinators (or directors) were interviewed, along with at least one other field-staff personnel from each PO. From the associations, all presidents and treasurers (administrators) were consulted. Other association members were also interviewed. University researchers were included to gain a broader perspective on the role of community timber projects in western Amazonia development. Interviews took 1-2 hours on average; they were conducted as informal conversations. Guiding questions or topics were suggested according to the evolution of the discussion in a flexible manner, giving the most freedom to the interviewee to lead the dialogue. All discussions took place in a friendly and

PAGE 33

21 relaxed environment. All of the participants agreed willingly to be taped. Time constraints allowed for only one visit to each respondent; only in very few cases was there a chance for a second contact. Data Analysis Data from both primary and secondary sources were analyzed using qualitative methods, such as stakeholder maps and comparative matrices. Archival information was assessed to identify critical themes and develop timelines of keystone events and processes. These data were critical for cross-referencing comments from the interviews and the meeting notes. A stakeholder map was developed initially as an instrumental tool to guide the analytical process. Themes selected prior to the interviews were reviewed and used to organize the data into a series of matrices that will be presented and discussed throughout the thesis. Stakeholders external to the timber projects were separated and grouped in a single matrix. Both project personnel (from NGO and government agencies) and association leaders were brought together in three different matrices: one regarding the transfer of project decision making (appropriation); a second on the management capacity of the local associations; and a third clustering their perceptions on the relationship dynamics within each project. Perceptions about the GPF idea and its process so far were assembled into a matrix including all the stakeholders. Synthesis This chapter introduces the emergence of the GPF coalition as the issue of research of this thesis. It frames the Grupo de Produtores Florestais as a response to the broader challenge of how to justly integrate rural communities to

PAGE 34

22 global markets. The chapter describes the sociopolitical context of community forestry in western Amazonia. It describes how the creation of extractivist communal areas like RESEX and PAEs provided the opportunity for communal timber projects to be established in the mid nineties. Solving the marketing problems confronted by these projects was the main reason behind the initial GPF proposal. The history, of this initiative is detailed in this chapter. The last section of the chapter presents the structural pillars of this research. Beginning with the hypothesis it describes the research design and strategy that was supported by the theoretical framework presented in Chapter Two.

PAGE 35

CHAPTER 2 ANALYTICAL FRAMEWORK In 1993, Kai Lees seminal book Compass and Gyroscope introduced the concept of adaptive management, from the original ideas of C.S. Holling (Holling, 1978) and Carl Walters (Walters, 1986), to the wider conservation community. Lee compared the current challenge of dealing with environmental change to that of the sixteenth century explorers, who were acutely aware of the limited capacity of their navigation gadgets and their knowledge. Also, they dealt with uncertainty on a daily basis and generated practical knowledge, vital for their immediate survival (Lee, 1993). The situation of practitioners working in the development of multi-stakeholder collaboration efforts in natural resource management (NRM) is no different. Most of these NRM initiatives have started spontaneously, sparked by a major social or ecological disturbance. In other situations, a long history of failures or conflict has left parties with collaboration as a last option. In most of the cases, efforts have run on intuition and the goodwill of the parties involved, more than on systematized knowledge. This issue represents a limitation to collaborations potential to achieve effective conservation and development. This chapter aims at contributing to the crafting of that knowledge by providing an introduction to the study of collaboration in NRM, with particular emphasis on the study of community forestry multi-stakeholder collaboration in the form of formal second-degree coalitions. As sources for this review I have 23

PAGE 36

24 considered the available, but scattered, literature on the subject, along with works in related fields such as political ecology, adaptive management, social learning and innovations. I have woven this diverse literature into an analytical framework to assess the conditions for the emergence of multi-stakeholder coalitions. I detail the components of this framework in the latter part of this chapter and use it in the following chapters to evaluate the situation of the GPF initiative in the western Amazon Basin, Brazil. Natural Resources Management Coalitions Coalitions and other types of multi-stakeholder collaboration innovations in NRM have gradually drawn attention from researchers as the practice multiplies in a diversity of ecological contexts. In each site where multi-stakeholder dialog has grown into substantial collaboration and later into new forms of organization, motivation seemed to lie within each participants experiences and interests. The process of collaboration is an expression of the self-organizing quality of complex living systems (Holling et al., 1998). Neither theory nor enforced external policy seems to have played a direct or substantial role in the development of this trend (Brick et al., 2001; Wondolleck & Yaffee, 2000). The parties needs assessment plays a key role in crafting these collaboration efforts, but is accompanied in most cases by a nested set of institutions and external factors that both constrain and facilitate this bottom-up drive. These other factors, such as development projects (and their founding theories), governmental policies and institutions, and donor agendas, condition the expression and extension of collaboration. In many cases collaborative fora are platforms where actors representing formal social structure (NGOs, government agencies) come together with social

PAGE 37

25 movement actors. As such, these platforms run the risk of being interpreted solely as mediating devices to serve the powerful groups (Edmunds & Wollenberg, 2001; Kenney, 2000; Nygren, 2000). The emergence of a myriad of collaborative initiatives has occurred without any thoughtful scholarly analysis of the conditions under which such efforts would be successful. Many of the studies done thus far have been based on North American experiences (Brick et al., 2001; Wondolleck & Yaffee, 2000), with only a scattered number of case studies written on coalitions in the tropics (Fisher et al., 2001; Poats, 2001) and even fewer that look at community forestry networks (Rosendo & Brown, 1998; Shrestha et al., 1998). Conservation organizations such as the Biodiversity Support Program (BSP, 2000) and the Center for International Forestry Research (Buckles, 1999), as well as The Nature Conservancy (Fisher et al., 2001; Monge, 2001) and the International Union for the Conservation of Nature (Borrini-Feyerabend et al., 2001), have publications promoting the development of collaborative resource management, with documented cases and guides for establishing similar groups. Even within this small amount of emergent applied literature, the complexity of the phenomenon of collaboration is evident. The variety of titles (networks, alliances, co-management, coalitions, federations, etc) used to name the collaborative structures are as diverse as the contexts in which the networks arise. Fisher et al. underline the need for an objective and critical study of collaborative experiences, given the risks of promoting alliances, coalitions, etc., as the new magic bullet for conservation (Fisher et al., 2001).

PAGE 38

26 Community Forestry Coalitions (CFCs) According to the FAO, the potential of forests and their resources to reduce poverty depends upon, among other factors, support to participatory processes, multi-party learning and enhanced interagency collaboration (Thomson & Freudenberger, 1997). Community forestry, defined as the management of forest resources by local people for their own benefit (Hood et al., 1998; Veer et al., 1997), should rely strongly on developing the capacity to collaborate effectively. In fact, forestry has transformed itself from a technical activity, into an interdisciplinary field in which economic, social and political understanding is crucial (Colchester et al., 2003). Community forestry is among the most sensitive areas, where social and human skills play an important role. Community forestry practitioners have to deal with the forest issue as it encounters market pressures, land-right disputes, legally mandated participation and other current phenomena produced by the rapidly changing political environment (Alcorn & Royo, 1999; Escobar, 1999; Peluso, 1992). Approaches to Analyze Community Forestry Coalitions In this section I review different approaches that have been used to analyze collaborative efforts in natural resource management. Guided by the definition of CFCs from the previous section, I selected studies of collaborative efforts that shared similarities with CFCs, so that their methods and tools could be extrapolated for the analytical framework I designed to study the GPF in Brazil. From this review a scattered set of elements from different sources was identified, and three frameworks were compared to establish the final study variables.

PAGE 39

27 Researchers from a variety of disciplines have analyzed collaborative resource management. Anthropologists and sociologists interested in social movements and collective action have a long tradition of looking at peasants and rural social movements (L. Anderson, 1994). Political ecology brought this same attention to look at social groups and social movements with closer ties to their environment, such as indigenous and extractivist societies. In this field, collaboration and conflict between social actors at different scales remains an area of interest for research (Bryant, 1991; Edmunds & Wollenberg, 2001; McCay, 2000; Peluso, 1992; Schmink & Wood, 1987). Interest in linking social and ecological systems has also sparked an interest in collaboration and common property, leading to a set of interdisciplinary studies of collaborative NRM systems (Folke et al., 2002; Holling & Meffe, 1996; Lee, 1993; Royo, 1999). Social capital studies led by political scientists and economists also consider the role of collaborative organizations. Communitys organizational capacity, social networks and second degree organizations (federations) are studied as forms of structural social capital (Babbington & Carroll, 2003; Dasgupta, 2002; Pender & Scherr, 2002). These studies have applied a diversity of methodologies from quantitative to ethnographic. The focus of research has leaned towards the results and the impact of collaboration, not so much on the process. The conditions leading to collaboration, or the organizations for collaboration, have received little attention. Both social capital and common property (Kopelman et al., 2001; Ostrom, 2003)

PAGE 40

28 represent new frameworks for which process and organizational characteristics are increasingly the center of attention. The efforts to understand NRM collaborative processes also evolved from conflict management and environmental conflict management research (Buckles, 1999; Susskind et al., 1999). Conflict management researchers have mostly relied on case study methodologies to extract the collaborative process characteristics (Brick et al., 2001; Fisher et al., 2001; Margoluis et al., 2000; Yaffee et al., 1997). In Table 2.1 I compared three of these cases across the set of critical variables for collaborative research identified by each study. Table 2.1 NRM Collaborative Organization Analytic Variables Coughlin et al., 1999 Edwards et al., 2001 Fisher et al., 2001 Location Issues Participants Outcomes Decision Authority Connection to existing procedures Elements of process structure Scientific basis for planning, decision-making, implementation and monitoring Level of support/ opposition Level of experience/ knowledge Funding Time frame (when initiated/ meeting frequency Scale of projects Land ownership Function Membership Objectives Management Leadership Facilitation Information exchange Skill development Communication Resources Levels (scale of members) Goals and objectives Size and composition Geographic range Degree of formality Leadership Management Supportive policies and institutions Threats to conservation Process Structure Results The first case, from Coughlin et al, is itself a comparative study of conservation partnerships in the midwest of the United States. The second is from the East-Asia Community Forestry Network (RECOFT), based on characteristics identified by participants of a forestry network workshop in 2001.

PAGE 41

29 In the third, Fisher et al. provide a conceptual framework for conservation coalitions, based on protected areas cases in the United States and other tropical countries. For this comparison I use the work of Arun Agrawal (2001) comparing common property regime (CPR) institutions as a model. Agrawal concludes that developing lists of conditions for CPR institutional analysis is useless (Agrawal, 2001). Nevertheless, I believe that such an approach is appropriate because this analysis is targeted towards feeding back practical information, rather than employing the case study merely as an anchor for a theoretical analysis of institutional development. Moreover, as Agrawal points out, the list of categorized variables can be calibrated and used in cross-case analysis with the adequate database. Despite being geared towards different types of collaborative efforts the variables identified could be applied regardless of the type of structure (partnership, network or coalition) (Table 2.1). Biophysical aspects were included in variables such as location, geographical range, threats to conservation, and resources managed. Background and contextual aspects were incorporated in factors such as time frame, decision authority, scale of members and, supportive polices and institutions. All three approaches considered strategic aspects such as objectives and outcomes of collaboration. Operational aspects were common to all cases, as well (structure, leadership, management, resources, process). These overall variable categoriesbiophysical, contextual, strategic and operationalcan also be broken down into two more general areas: internal

PAGE 42

30 variables and external variables. Internal variables consider issues within the groups involved in collaboration, in addition to the process and purpose of the cooperation itself. External variables refer to all aspects outside the collaborative process that provide the surrounding environment to the effort, including biophysical, socioeconomic and political aspects. In order to adapt these variables to evaluate the GPF initiative, I had to take into account the issue of temporal change within collaborative efforts. While the majority of studies have examined already-consolidated coalition processes, the GPF initiative was still in a pre-emergence phase. References regarding the conditions leading towards collaboration were made in retrospect and only for successful cases (Colchester et al., 2003; Wondolleck & Yaffee, 2000). From the set of variables considered above, the external set could be applied directly to the GFP case, but the internal variables needed to be readjusted. Since there was no existing coalition for which to consider these variables (decision-making process, leadership or structure), I redirected these variables to analyze the experiences of the potential coalition members. The assumption was that the past and current organizational behavior of the potential members would inform the behavior of these groups in the future coalition. By analyzing current collaborative practices we could forecast their behavior in the future coalition. To do this, I relied on the concept of communities of practice (Wenger, 1988; Wenger & Snyder, 2000) as a guiding metaphor to study the potential coalition-members internal organizational issues, including factors such

PAGE 43

31 as internal conflicts, leadership, and decision-making and organizational capacity. To analyze the GPF in its pre-emergent state, I decided to consider it as an idea that was still being presented to parties and discussed. Therefore, I decided to incorporate a particular set of variables that would assess the process of design, diffusion and adoption. For this purpose, I adapted elements from the study of technological innovations (Douthwaite, 2002; Douthwaite et al., 2001; Rogers, 1995) that will be explained in a later section. First, however, I detail the analytical framework that resulted from the review of the literature on collaborative research discussed above. I go into more depth regarding particular aspects of the literature as they inform sections of the framework. This framework built on my previous work included coordinating a cross-site learning coalition project for an international conservation organization (cases in Panama, Mexico, Brazil, Peru and Panama). For this project I prepared a self-evaluation tool for coalitions, which is related to this framework. Previously, I had also participated in the documentation of three unrelated natural resource management coalition cases (Galapagos National Park, Ecuador; Osa Biological Corridor, Costa Rica; Serra do Divisor, Brazil) Framework to Analyze Community Forestry Coalitions Thus far, I have delineated the theoretical trajectory that led me to the design of the analytical framework I used in my research. I used a hierarchy theory approach (Allen, 2004) to structure my analysis of the factors providing conditions for the emergence of a CFC. These factors were drawn from the literature studying different types of conservation coalitions, and organized

PAGE 44

32 through a hierarchical approach. Hierarchy theory posits that any level of examination is constrained by the levels immediately above while controlling those immediately below; thus, all three levels must be considered in order to understand causes and consequences at the level of interest. Figure 2.1 presents the components and boundaries of the hierarchical system to study. The middle component (design factors) represents the main scale of interest. This represents the level where the organizational innovation is occurring. Design refers to the structure and process definition of the community forestry coalition. This scale contains the lower scale of internal factors that encompasses the groups that are interested in participating in the coalition. At the highest level is the external factors component. This scale contains the factors that constrain the activities of the interest groups (potential coalition members). By developing a hierarchical approach to CFCs, I pursue the integration of the seemingly paradoxical aspects of coalitions, such as the manifestation of both human agency and sociopolitical structure. Agency is contemplated within the internal factors and structure in the external ones. While the hierarchical approach allows the mapping of the GPFs significant components, its main dynamics might remain hidden. I focus on two key dynamics: first, on the cross-scale interactions that will be analyzed mainly in Chapter Four (external factors) and secondly, in the social learning process occurring within the timber projects and through the GPF innovation process. Chapter Three (internal factors) and Chapter Five (design factors) both discuss the learning dynamics.

PAGE 45

33 Table 2.2: CFC Analytical Framework Categories Factors (Variables) Indicators (Sources) Collection tools Analysis tools At the Group level: o Participation o Leadership o Management Capacity Documents of activities, results Organization members perception Questionnaires Interviews with key group leaders Archival research Stakeholder analysis and map Internal: At the Project Level: o Partner relationships o Capacity development o Appropriation by community Project documents: reports, proposals, evaluations Project participants perception Questionnaires Interviews with key project participants Archival research Stakeholder mapping Comparative Matrix to elicit the common collaboration pattern External: State and federal government agencies, policies and laws on conservation, forestry development and social organization NGO and Aid agencies Forest products markets Laws, policy documents, project proposals, regulations, donor policy documents, NGO reports, Key official perception about the issues Questionnaires Interviews with key project participants Archival research Stakeholder analysis and mapping Coalition Design: o Objectives o Structure o Leadership o Resources Documents detailing the coalition proposal Promoters vision and perception Questionnaires Interviews with key project participants Archival research Graphs visualizing coalition structure alternatives Design: Innovation Process o History o Perceptions o Adaptations Promoters vision and perception Potential participants perception External supporting agents perception Questionnaires Interviews with key project participants Archival research Innovation process timeline Coalition proposal perception matrix Table 2.2 shows the three main categories of factors: internal, external and design (column 1). The remaining four columns detail the factors within each category, the indicators for each of those factors, and the relevant collection and analytical tools 8 8 It should be noted that this matrix organizes the factor categories differently than Figure 2.1. The chosen sequence that starts with describing the actors (internal) then the context (external) and finally the innovation process (design) is meant to be more accessible to the reader. It is also a sequence that resembles more closely the research process that when from the actors to the context and finally looked at the coalition innovation.

PAGE 46

34 The next sections of this chapter will describe each of the three general categories, starting with each categorys objective and then moving horizontally across the table, discussing the variables, the indicators and the tools. A subsequent section integrates the three components. The final section presents some of the caveats and limitations of the analytical framework. External Factors: Design Factors:Diffusion of InnovationsPolitical Ecology Coalition process Communities of practice,Social learning Internal Factors: Figure 2.1: Hierarchical System of Factors Conditioning the GPF 9 Internal Factors The internal factors category summarizes the conditions of the organizations interested in the formation of a coalition. In the case of the CFC cases reviewed, these groups were mainly community-based associations directly involved in community timber enterprises. In this analytical framework, I consider not only these groups but also the NGOs or government agencies that jointly developed the timber activities with the communities (Figure 2.2). Although formally, CFCs have not included entities other than community-based 9 Stone (2003) uses a similar diagram in her analysis of local participation in two community projects in Acre.

PAGE 47

35 groups, I believe the supporting groups play a very important role in the emergence of this type of second-degree organization, particularly in the case of Acre and Rondnia. In these states, community timber activities are conducted through projects financed by international donors and executed by these non-community groups. Both the community associations and the partner organizations described in this figure conform the middle component of this studys hierarchical system. \000\000\000\000\000\000\000\000\000\000\000\000\000)TjETEMC /P <>BDC Qq174.17999 500.63989 m171.60001 500.51999 l169.2 500.34 l166.67999 499.91995 l164.34 499.37985 l162.17999 498.66 l160.2 497.81998 l158.16 496.85999 l156.42 495.84 l154.86 494.63989 l153.36 493.37985 l152.10001 492.06 l151.08 490.67999 l150.23999 489.12 l149.58 487.56 l149.16 485.94 l149.03999 484.25998 l149.16 482.57999 l149.58 481.01999 l150.23999 479.45999 l151.08 477.89993 l152.10001 476.51999 l153.36 475.13992 l154.86 474 l156.42 472.85999 l158.16 471.84 l160.2 470.87991 l162.17999 470.03998 l164.34 469.31998 l166.67999 468.78 l169.2 468.35999 l171.60001 468.12 l174.17999 468.06 l176.81999 468.12 l179.34 468.35999 l181.73999 468.78 l184.08 469.31998 l186.23999 470.03998 l188.15999 470.87991 l190.14 471.84 l191.94 472.85999 l193.5 474 l194.94 475.13992 l196.2 476.51999 l197.28 477.89993 l198.12 479.45999 l198.72 481.01999 l199.14 482.57999 l199.25998 484.25998 l199.14 485.94 l198.72 487.56 l198.12 489.12 l197.28 490.67999 l196.2 492.06 l194.94 493.37985 l193.5 494.63989 l191.94 495.84 l190.14 496.85999 l188.15999 497.81998 l186.23999 498.66 l184.08 499.37985 l181.73999 499.91995 l179.34 500.34 l176.81999 500.51999 l174.17999 500.63989 lW* n0 0 0 scnBT/T3_0 1 Tf1.92 0 0 -11.52 147.84 495.35999 Tm(\000\000\000\000\000\000\000\000\000\000\000\000\000)TjETEMC /P <>BDC Qq174.17999 500.63989 m171.60001 500.51999 l169.2 500.34 l166.67999 499.91995 l164.34 499.37985 l162.17999 498.66 l160.2 497.81998 l158.16 496.85999 l156.42 495.84 l154.86 494.63989 l153.36 493.37985 l152.10001 492.06 l151.08 490.67999 l150.23999 489.12 l149.58 487.56 l149.16 485.94 l149.03999 484.25998 l149.16 482.57999 l149.58 481.01999 l150.23999 479.45999 l151.08 477.89993 l152.10001 476.51999 l153.36 475.13992 l154.86 474 l156.42 472.85999 l158.16 471.84 l160.2 470.87991 l162.17999 470.03998 l164.34 469.31998 l166.67999 468.78 l169.2 468.35999 l171.60001 468.12 l174.17999 468.06 l176.81999 468.12 l179.34 468.35999 l181.73999 468.78 l184.08 469.31998 l186.23999 470.03998 l188.15999 470.87991 l190.14 471.84 l191.94 472.85999 l193.5 474 l194.94 475.13992 l196.2 476.51999 l197.28 477.89993 l198.12 479.45999 l198.72 481.01999 l199.14 482.57999 l199.25998 484.25998 l199.14 485.94 l198.72 487.56 l198.12 489.12 l197.28 490.67999 l196.2 492.06 l194.94 493.37985 l193.5 494.63989 l191.94 495.84 l190.14 496.85999 l188.15999 497.81998 l186.23999 498.66 l184.08 499.37985 l181.73999 499.91995 l179.34 500.34 l176.81999 500.51999 l174.17999 500.63989 lW* n0 0 0 scnBT/T3_0 1 Tf1.92 0 0 -11.52 147.84 483.84 Tm(\000\000\000\000\000\000\000\000\000\000\000\000\000)TjETEMC /P <>BDC Qq174.17999 500.63989 m171.60001 500.51999 l169.2 500.34 l166.67999 499.91995 l164.34 499.37985 l162.17999 498.66 l160.2 497.81998 l158.16 496.85999 l156.42 495.84 l154.86 494.63989 l153.36 493.37985 l152.10001 492.06 l151.08 490.67999 l150.23999 489.12 l149.58 487.56 l149.16 485.94 l149.03999 484.25998 l149.16 482.57999 l149.58 481.01999 l150.23999 479.45999 l151.08 477.89993 l152.10001 476.51999 l153.36 475.13992 l154.86 474 l156.42 472.85999 l158.16 471.84 l160.2 470.87991 l162.17999 470.03998 l164.34 469.31998 l166.67999 468.78 l169.2 468.35999 l171.60001 468.12 l174.17999 468.06 l176.81999 468.12 l179.34 468.35999 l181.73999 468.78 l184.08 469.31998 l186.23999 470.03998 l188.15999 470.87991 l190.14 471.84 l191.94 472.85999 l193.5 474 l194.94 475.13992 l196.2 476.51999 l197.28 477.89993 l198.12 479.45999 l198.72 481.01999 l199.14 482.57999 l199.25998 484.25998 l199.14 485.94 l198.72 487.56 l198.12 489.12 l197.28 490.67999 l196.2 492.06 l194.94 493.37985 l193.5 494.63989 l191.94 495.84 l190.14 496.85999 l188.15999 497.81998 l186.23999 498.66 l184.08 499.37985 l181.73999 499.91995 l179.34 500.34 l176.81999 500.51999 l174.17999 500.63989 lW* n0 0 0 scnBT/T3_0 1 Tf1.92 0 0 -11.52 147.84 472.31998 Tm(\000\000\000\000\000\000\000\000\000\000\000\000\000)TjETEMC /InlineShape <>BDC Q0.09399 0.827 0 scn149.03999 467.81998 50.40001 32.82001 ref/CS0 CS 0 0 0 SCN0.3 w 10 M 1 j 1 J []0 d 199.25998 484.31998 m199.25998 484.31998 l199.25998 493.32901 188.024 500.63989 174.17999 500.63989 c160.33601 500.63989 149.10001 493.32901 149.10001 484.31998 c149.10001 475.31104 160.33601 468 174.17999 468 c188.024 468 199.25998 475.31104 199.25998 484.31998 chS1 0.612 0.188 scn374.91 388.85999 m374.91 388.85999 l374.91 424.13303 325.49103 452.75998 264.60001 452.75998 c203.70901 452.75998 154.28999 424.13303 154.28999 388.85999 c154.28999 353.58701 203.70901 324.95999 264.60001 324.95999 c325.49103 324.95999 374.91 353.58701 374.91 388.85999 cf375.09 388.85999 m375.09 388.85999 l375.09 424.24902 325.59 452.97 264.60001 452.97 c203.61 452.97 154.11 424.24902 154.11 388.85999 c154.11 353.47101 203.61 324.75 264.60001 324.75 c325.59 324.75 375.09 353.47101 375.09 388.85999 chS0.09399 0.827 0 scn344.75998 353.63998 m344.75998 353.63998 l344.75998 362.53302 333.60501 369.75 319.85999 369.75 c306.11502 369.75 294.95999 362.53302 294.95999 353.63998 c294.95999 344.74701 306.11502 337.53 319.85999 337.53 c333.60501 337.53 344.75998 344.74701 344.75998 353.63998 cf344.97 353.63998 m344.97 353.63998 l344.97 362.63202 333.72101 369.92999 319.85999 369.92999 c305.99902 369.92999 294.75 362.63202 294.75 353.63998 c294.75 344.64801 305.99902 337.35001 319.85999 337.35001 c333.72101 337.35001 344.97 344.64801 344.97 353.63998 chS248.51999 426.53998 m248.51999 426.53998 l248.51999 435.46603 237.33801 442.70999 223.56 442.70999 c209.78201 442.70999 198.60001 435.46603 198.60001 426.53998 c198.60001 417.61401 209.78201 410.37 223.56 410.37 c237.33801 410.37 248.51999 417.61401 248.51999 426.53998 cf248.7 426.53998 m248.7 426.53998 l248.7 435.56503 237.43701 442.88995 223.56 442.88995 c209.68301 442.88995 198.42 435.56503 198.42 426.53998 c198.42 417.51501 209.68301 410.19 223.56 410.19 c237.43701 410.19 248.7 417.51501 248.7 426.53998 chS188.34 337.13998 50.52 32.82001 ref1 1 1 scn213.48 369.95999 m210.90001 369.84 l208.44 369.66 l205.92 369.23999 l203.64 368.70001 l201.42 367.97998 l199.44 367.13998 l197.45999 366.17999 l195.65999 365.10001 l194.10001 363.95999 l192.65999 362.70001 l191.40001 361.44 l190.31999 360.12 l189.48 358.5 l188.87999 356.94 l188.45999 355.38 l188.34 353.70001 l188.45999 352.01999 l188.87999 350.45999 l189.48 348.89999 l190.31999 347.28 l191.40001 345.95999 l192.65999 344.57999 l194.10001 343.32001 l195.65999 342.17999 l197.45999 341.10001 l199.44 340.20001 l201.42 339.35999 l203.64 338.57999 l205.92 338.10001 l208.44 337.67999 l210.90001 337.44 l213.48 337.38 l216.12 337.44 l218.64 337.67999 l221.03998 338.10001 l223.31998 338.57999 l225.53998 339.35999 l227.51999 340.20001 l229.5 341.10001 l231.3 342.17999 l232.86 343.32001 l234.36 344.57999 l235.62 345.95999 l236.64 347.28 l237.48 348.89999 l238.08 350.45999 l238.5 352.01999 l238.62 353.70001 l238.5 355.38 l238.08 356.94 l237.48 358.5 l236.64 360.12 l235.62 361.44 l234.36 362.70001 l232.86 363.95999 l231.3 365.10001 l229.5 366.17999 l227.51999 367.13998 l225.53998 367.97998 l223.31998 368.70001 l221.03998 369.23999 l218.64 369.66 l216.12 369.84 l213.48 369.95999 lf*EMC /P <>BDC q213.48 369.95999 m210.90001 369.84 l208.44 369.66 l205.92 369.23999 l203.64 368.70001 l201.42 367.97998 l199.44 367.13998 l197.45999 366.17999 l195.65999 365.10001 l194.10001 363.95999 l192.65999 362.70001 l191.40001 361.44 l190.31999 360.12 l189.48 358.5 l188.87999 356.94 l188.45999 355.38 l188.34 353.70001 l188.45999 352.01999 l188.87999 350.45999 l189.48 348.89999 l190.31999 347.28 l191.40001 345.95999 l192.65999 344.57999 l194.10001 343.32001 l195.65999 342.17999 l197.45999 341.10001 l199.44 340.20001 l201.42 339.35999 l203.64 338.57999 l205.92 338.10001 l208.44 337.67999 l210.90001 337.44 l213.48 337.38 l216.12 337.44 l218.64 337.67999 l221.03998 338.10001 l223.31998 338.57999 l225.53998 339.35999 l227.51999 340.20001 l229.5 341.10001 l231.3 342.17999 l232.86 343.32001 l234.36 344.57999 l235.62 345.95999 l236.64 347.28 l237.48 348.89999 l238.08 350.45999 l238.5 352.01999 l238.62 353.70001 l238.5 355.38 l238.08 356.94 l237.48 358.5 l236.64 360.12 l235.62 361.44 l234.36 362.70001 l232.86 363.95999 l231.3 365.10001 l229.5 366.17999 l227.51999 367.13998 l225.53998 367.97998 l223.31998 368.70001 l221.03998 369.23999 l218.64 369.66 l216.12 369.84 l213.48 369.95999 lW* n0 0 0 scnBT/T3_0 1 Tf1.92 0 0 -11.52 188.15999 380.16 Tm(\000\000\000\000\000\000\000\000\000\000\000\000\000)TjETEMC /P <>BDC Qq213.48 369.95999 m210.90001 369.84 l208.44 369.66 l205.92 369.23999 l203.64 368.70001 l201.42 367.97998 l199.44 367.13998 l197.45999 366.17999 l195.65999 365.10001 l194.10001 363.95999 l192.65999 362.70001 l191.40001 361.44 l190.31999 360.12 l189.48 358.5 l188.87999 356.94 l188.45999 355.38 l188.34 353.70001 l188.45999 352.01999 l188.87999 350.45999 l189.48 348.89999 l190.31999 347.28 l191.40001 345.95999 l192.65999 344.57999 l194.10001 343.32001 l195.65999 342.17999 l197.45999 341.10001 l199.44 340.20001 l201.42 339.35999 l203.64 338.57999 l205.92 338.10001 l208.44 337.67999 l210.90001 337.44 l213.48 337.38 l216.12 337.44 l218.64 337.67999 l221.03998 338.10001 l223.31998 338.57999 l225.53998 339.35999 l227.51999 340.20001 l229.5 341.10001 l231.3 342.17999 l232.86 343.32001 l234.36 344.57999 l235.62 345.95999 l236.64 347.28 l237.48 348.89999 l238.08 350.45999 l238.5 352.01999 l238.62 353.70001 l238.5 355.38 l238.08 356.94 l237.48 358.5 l236.64 360.12 l235.62 361.44 l234.36 362.70001 l232.86 363.95999 l231.3 365.10001 l229.5 366.17999 l227.51999 367.13998 l225.53998 367.97998 l223.31998 368.70001 l221.03998 369.23999 l218.64 369.66 l216.12 369.84 l213.48 369.95999 lW* n0 0 0 scnBT/T3_0 1 Tf1.92 0 0 -11.52 188.15999 368.63998 Tm(\000\000\000\000\000\000\000\000\000\000\000\000\000)TjETEMC /P <>BDC Qq213.48 369.95999 m210.90001 369.84 l208.44 369.66 l205.92 369.23999 l203.64 368.70001 l201.42 367.97998 l199.44 367.13998 l197.45999 366.17999 l195.65999 365.10001 l194.10001 363.95999 l192.65999 362.70001 l191.40001 361.44 l190.31999 360.12 l189.48 358.5 l188.87999 356.94 l188.45999 355.38 l188.34 353.70001 l188.45999 352.01999 l188.87999 350.45999 l189.48 348.89999 l190.31999 347.28 l191.40001 345.95999 l192.65999 344.57999 l194.10001 343.32001 l195.65999 342.17999 l197.45999 341.10001 l199.44 340.20001 l201.42 339.35999 l203.64 338.57999 l205.92 338.10001 l208.44 337.67999 l210.90001 337.44 l213.48 337.38 l216.12 337.44 l218.64 337.67999 l221.03998 338.10001 l223.31998 338.57999 l225.53998 339.35999 l227.51999 340.20001 l229.5 341.10001 l231.3 342.17999 l232.86 343.32001 l234.36 344.57999 l235.62 345.95999 l236.64 347.28 l237.48 348.89999 l238.08 350.45999 l238.5 352.01999 l238.62 353.70001 l238.5 355.38 l238.08 356.94 l237.48 358.5 l236.64 360.12 l235.62 361.44 l234.36 362.70001 l232.86 363.95999 l231.3 365.10001 l229.5 366.17999 l227.51999 367.13998 l225.53998 367.97998 l223.31998 368.70001 l221.03998 369.23999 l218.64 369.66 l216.12 369.84 l213.48 369.95999 lW* n0 0 0 scnBT/T3_0 1 Tf1.92 0 0 -11.52 188.15999 357.12 Tm(\000\000\000\000\000\000\000\000\000\000\000\000\000)TjETEMC /P <>BDC Qq213.48 369.95999 m210.90001 369.84 l208.44 369.66 l205.92 369.23999 l203.64 368.70001 l201.42 367.97998 l199.44 367.13998 l197.45999 366.17999 l195.65999 365.10001 l194.10001 363.95999 l192.65999 362.70001 l191.40001 361.44 l190.31999 360.12 l189.48 358.5 l188.87999 356.94 l188.45999 355.38 l188.34 353.70001 l188.45999 352.01999 l188.87999 350.45999 l189.48 348.89999 l190.31999 347.28 l191.40001 345.95999 l192.65999 344.57999 l194.10001 343.32001 l195.65999 342.17999 l197.45999 341.10001 l199.44 340.20001 l201.42 339.35999 l203.64 338.57999 l205.92 338.10001 l208.44 337.67999 l210.90001 337.44 l213.48 337.38 l216.12 337.44 l218.64 337.67999 l221.03998 338.10001 l223.31998 338.57999 l225.53998 339.35999 l227.51999 340.20001 l229.5 341.10001 l231.3 342.17999 l232.86 343.32001 l234.36 344.57999 l235.62 345.95999 l236.64 347.28 l237.48 348.89999 l238.08 350.45999 l238.5 352.01999 l238.62 353.70001 l238.5 355.38 l238.08 356.94 l237.48 358.5 l236.64 360.12 l235.62 361.44 l234.36 362.70001 l232.86 363.95999 l231.3 365.10001 l229.5 366.17999 l227.51999 367.13998 l225.53998 367.97998 l223.31998 368.70001 l221.03998 369.23999 l218.64 369.66 l216.12 369.84 l213.48 369.95999 lW* n0 0 0 scnBT/T3_0 1 Tf1.92 0 0 -11.52 188.15999 345.60001 Tm(\000\000\000\000\000\000\000\000\000\000\000\000\000)TjETEMC /InlineShape <>BDC Q0.09399 0.827 0 scn188.34 337.13998 50.52 32.82001 ref238.64998 353.63998 m238.64998 353.63998 l238.64998 362.63202 227.37401 369.92999 213.48 369.92999 c199.58601 369.92999 188.31 362.63202 188.31 353.63998 c188.31 344.64801 199.58601 337.35001 213.48 337.35001 c227.37401 337.35001 238.64998 344.64801 238.64998 353.63998 chS328.97998 426.53998 m328.97998 426.53998 l328.97998 435.46603 317.798 442.70999 304.01999 442.70999 c290.242 442.70999 279.06 435.46603 279.06 426.53998 c279.06 417.61401 290.242 410.37 304.01999 410.37 c317.798 410.37 328.97998 417.61401 328.97998 426.53998 cf329.16 426.53998 m329.16 426.53998 l329.16 435.56503 317.897 442.88995 304.01999 442.88995 c290.14301 442.88995 278.88 435.56503 278.88 426.53998 c278.88 417.51501 290.14301 410.19 304.01999 410.19 c317.897 410.19 329.16 417.51501 329.16 426.53998 chS0.933 0.161 0.17999 scn228.06 365.03998 m219.23999 350.03998 l236.81998 350.03998 l228.06 365.03998 lf*0.42 w 228.17998 364.91998 m219.36 349.91998 lS0.3 w 219.36 349.91998 m236.93999 349.91998 lS0.42 w 236.93999 349.91998 m228.17998 364.91998 lS307.67999 365.03998 m298.85999 350.03998 l316.44 350.03998 l307.67999 365.03998 lf*307.73999 364.91998 m298.97998 349.91998 lS0.3 w 298.97998 349.91998 m316.56 349.91998 lS0.42 w 316.56 349.91998 m307.73999 364.91998 lS290.04001 437.94 m281.28 422.88 l298.85999 422.88 l290.04001 437.94 lf*290.16 437.88 m281.34 422.75998 lS0.3 w 281.34 422.75998 m298.97998 422.75998 lS0.42 w 298.97998 422.75998 m290.16 437.88 lS235.37999 437.94 m225.95999 422.88 l244.67998 422.88 l235.37999 437.94 lf*235.5 437.88 m226.08 422.75998 lS0.3 w 226.08 422.75998 m244.8 422.75998 lS0.42 w 244.8 422.75998 m235.5 437.88 lS197.15999 362.28 m189.3 350.57999 l201.12 342.72 l208.98 354.53998 l197.15999 362.28 lf*197.28 362.16 m189.42 350.45999 l201.23999 342.60001 l209.10001 354.41998 l197.28 362.16 lS333.84 364.80002 m321.72 357.35999 l329.22 345.23999 l341.28 352.73999 l333.84 364.80002 lf*333.95999 364.67999 m321.78 357.23999 l329.34 345.12 l341.39999 352.67999 l333.95999 364.67999 lS318.12 440.39999 m306.47998 432.12 l314.75998 420.47998 l326.39999 428.75998 l318.12 440.39999 lf*318.23999 440.28 m306.60001 432 l314.88 420.35999 l326.51999 428.63998 l318.23999 440.28 lS208.86 437.75998 m201 425.94 l212.87999 418.01999 l220.73999 429.89999 l208.86 437.75998 lf*208.98 437.63998 m201.12 425.81998 l212.94 417.95999 l220.8 429.78 l208.98 437.63998 lS1 0.612 0.188 scn156.60001 378.41998 208.14 19.2 refEMC /P <>BDC 0 0 0 scnBT/TT1 1 Tf13.44 0 0 13.44 164.7 383.03958 Tm(Community Forestry Coalition Community Forestry ProjectsCommunity Forestry OrganizationsSupport NGO or Gov. AgencyCommunity-based Organization Figure 2.2: CFC Internal Agents Unlike Mexico, Nepal or Guatemala where the community-based groups run the enterprise, in western Amazonia community timber extraction is still at an experimental stage. The autonomy of the community associations is very limited. This explains why I chose to use the forestry project as my unit of analysis. Each project was composed of a community-based group and a private or public agency that supported the activity technically and financially (Figure 2.2). In a material sense, the community forestry coalition would incorporate the

PAGE 48

36 forestry projects, which were composed of the two types of organizations: community associations (triangles) and partner organizations (diamonds). The objectives of this component of my research were three-fold. First, to describe the basic characteristics of each organization. Second, to assess the internal collaboration characteristics by looking at leadership, participation and management capacity. Third, to compare the projects in terms of partner relationships, degree of appropriation by community, and training focus. I expected to integrate these three aspects to sketch out the common collaboration pattern for the GPF. I define this pattern as the behavioral template or collaboration model that would implicitly guide the formation of the GPF as a second-degree collaboration platform. My perspective is that the traditional CFM view of two distinct sets of actors with insiders (community members) and outsiders (NGO personnel) in community projects is misleading. Instead, when these groups work together, they form a functional community. Just as any other, this community has hierarchies, boundaries, and functions. It exists parallel to and in conjunction with the more traditional communities of place. Researchers (P. H. C. Amaral, 2001; Stone, 2003) have pointed to the fact that not all members of the community are involved in the timber projects. In the same way, not everybody in the NGO or government agency participates in the projects, either. The functional community (or community of practice) of the timber project tends to be small and includes a diversity of individuals with different backgrounds and roles. Understanding this functional community is

PAGE 49

37 critical for evaluating the potential success of the proposed coalition. The figure below represents this emergent community made in the intersection between the land-based community association and the partner organization staff. \000\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000)TjETEMC /P <>BDC Qq318.17999 503.87985 m318.17999 503.87985 l318.17999 517.90601 302.509 529.28998 283.20001 529.28998 c263.89102 529.28998 248.22 517.90601 248.22 503.87985 c248.22 489.85403 263.89102 478.47 283.20001 478.47 c302.509 478.47 318.17999 489.85403 318.17999 503.87985 cW* n0 0 0 scnBT/T3_0 1 Tf0 Tc 0 Tw 1.92 0 0 -11.52 247.67998 518.40002 Tm(\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000)TjETEMC /P <>BDC Qq318.17999 503.87985 m318.17999 503.87985 l318.17999 517.90601 302.509 529.28998 283.20001 529.28998 c263.89102 529.28998 248.22 517.90601 248.22 503.87985 c248.22 489.85403 263.89102 478.47 283.20001 478.47 c302.509 478.47 318.17999 489.85403 318.17999 503.87985 cW* n0 0 0 scnBT/T3_0 1 Tf0 Tc 0 Tw 1.92 0 0 -11.52 247.67998 506.87985 Tm(\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000)TjETEMC /P <>BDC Qq318.17999 503.87985 m318.17999 503.87985 l318.17999 517.90601 302.509 529.28998 283.20001 529.28998 c263.89102 529.28998 248.22 517.90601 248.22 503.87985 c248.22 489.85403 263.89102 478.47 283.20001 478.47 c302.509 478.47 318.17999 489.85403 318.17999 503.87985 cW* n0 0 0 scnBT/T3_0 1 Tf0 Tc 0 Tw 1.92 0 0 -11.52 247.67998 495.35999 Tm(\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000)TjETEMC /P <>BDC Qq318.17999 503.87985 m318.17999 503.87985 l318.17999 517.90601 302.509 529.28998 283.20001 529.28998 c263.89102 529.28998 248.22 517.90601 248.22 503.87985 c248.22 489.85403 263.89102 478.47 283.20001 478.47 c302.509 478.47 318.17999 489.85403 318.17999 503.87985 cW* n0 0 0 scnBT/T3_0 1 Tf0 Tc 0 Tw 1.92 0 0 -11.52 247.67998 483.84 Tm(\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000)TjETEMC /InlineShape <>BDC Q318.26999 503.87985 m318.26999 503.87985 l318.26999 517.97302 302.55902 529.40997 283.20001 529.40997 c263.841 529.40997 248.12999 517.97302 248.12999 503.87985 c248.12999 489.78702 263.841 478.34998 283.20001 478.34998 c302.55902 478.34998 318.26999 489.78702 318.26999 503.87985 chSEMC /P <>BDC 0 0 0 scnBT/TT0 1 Tf-0.0004 Tc 0 Tw 16.98 0 0 16.98 138.72 608.99988 Tm(Partner OrganizationAssociation Community of practice(those involved in thetimber project)Community (land based) Figure 2.3: Timber Project Community of Practice A useful metaphor is provided by the theory of communities of practice developed for business management by organizational specialists (Wenger, 1988). Wenger notes that the idea of communities of practice presents a theory of learning that starts with this assumption: engagement in social practice is the fundamental process by which we learn and so become who we are. The primary unit of analysis is neither the individual nor social institutions but rather the informal communities of practice that people form as they pursue shared enterprises over time... 10 The fundamental idea behind a community of practice is that identity, trust and sense of belonging derive from the activities in 10 Text in the first page presentation (Wenger, 1988).

PAGE 50

38 which an individual engages rather than from the formal organization (Figure 2.4, below). If one looks at community forestry projects as such an activity, one can study the communities that build around them. Identity Practice Learning (Community)(Challenges, conflicts, adaptations,innovations)(Timber extraction, timber business) Figure 2.4 Communities of Practice Model I was interested in several aspects of such communities. The issue of trust was of particular interest, as it is fundamental for any collaborative effort. This is distinct from to the concept of trust in social capital research, where it is identified and assessed for quantifiable purposes (Babbington & Carroll, 2003; Dasgupta, 2002). My interest was in the practice of collaboration, the activity that depends upon and reproduces social capital. This is why my purpose was to delineate the common pattern of collaboration by looking at the different organizations experience. To operationalize this concept I adapted a set of variables from other studies on collaboration (Table 2.1) to the particularities of an emergent

PAGE 51

39 coalition. I provide the definition of these variables and their analysis in the next section. Variables and analysis Variables for the internal factors analysis are clustered at two levels: first, the organizational level that includes participation, leadership and management capacity of both the community associations (CA) and partner organizations (PO) of each project, and secondly, the projects level, which includes analyzing the relations among project partners (CA and PO), the degree to which the CA has appropriated the timber operation, and the training focus used by the PO in each project. To assess participation I used the framework developed by Stone (2003), which emphasizes the complexities of communities involved in timber projects and establishes a gradient of participation according to a larger set of external and internal conditions. My interest in the issue of community participation in the timber projects is much more focused than Stones. To operationalize participation, I considered the number of individuals involved in the community associations, and the assessment by both the CA leaders and the PO staff of the quality of involvement of those participants in association activities. Leadership was considered in both a material and a formal sense, since not all leaders of the groups were heads of the organizations 11 In the formal 11 Heifetz (1997) associates leadership with the idea of adaptive work, which is the learning process required to approach the conflicts between peoples values and the challenging reality. Leadership is about learning new ways of behaving. Another useful idea of this author is the notion of leading with-authority and leading without-authority. This refers to the fact that sometimes the individuals guiding the adaptive work do not hold a formal leadership position, and that at other times, the individuals in such positions avoid this type of work.

PAGE 52

40 aspect, I wanted to look at changes in formal leadership positions. In a community group, the lack of change in leadership might mean a lack of broader participation and a concentration of power. Such conditions might cause the NRM project to reinforce internal inequalities, which in turn would have an effect on the long-term sustainability of the natural resource base (Baland et al., 2002; Baland & Platteau, 2002). In contrast, the effect of frequent changes in the partner organization leadership (NGO, government agency) is often negative because it directly affects the execution of the project and prevents the consolidation of trust among participants. Changes in project leadership usually represent changes in technical orientation of the projects and in the way community participation is understood and incorporated. For management capacity, I concentrated on eliciting a qualitative assessment of the activities conducted by every CA. By management capacity I am referring to the activities and tasks that the different organizations are capable of doing. This issue is more relevant for the CA than for the PO. I wanted to identify the complexity of these activities, the duration, the number of people involved, and the reasons why the activities were abandoned, if that was the case. At the project level, I began with a basic description that included duration, objective, activities and results. I then concentrated on the key aspects of relations, training and appropriation. With regard to project relations, I first focused on analyzing conflicts within groups. Conflicts can represent a defining event that solidifies a groups identity.

PAGE 53

41 They can also create a negative environment that prevents any collaboration from occurring. Conflicts between the partner organization and the community group can also have a dual meaning. They can be the expression of the communitys agency as its members strive for greater control of the NRM project. Conflict can also be the expression of the individual agendas of leaders reluctant to loosen their grip over the project resources. For the purpose of assessing the potential for collaboration, the way in which conflict was managed was more important than the conflict itself. Although it was not expected that projects would have explicit mechanisms to deal with internal conflicts, conflictive experiences should build members skills and set organizational patterns. How those skills and patterns constructed the organizational culture was of primary interest in this analysis. My interest in the projects training focus was linked to the concept of capacity development. This refers to the process of acquiring new organizational skills that allow groups to increase the number and complexity of their activities. A projects approach to training is a good indicator of the role social learning plays. Approaches that are more explicit about social learning tend to foster experimentation and encourage collaboration between the PO and CA (Wondolleck & Yaffee, 2000). Appropriation refers to the degree to which the PO had successfully transferred the control and operation of the project tasks to the community-based groups. Stone discussed this aspect as a dimension of community participation. Since I was interested in delineating the common collaboration pattern, I looked

PAGE 54

42 at this issue in terms of process and as a direct indicator of the collaboration between POs and CAs in each project. In order to give greater emphasis to this, I treated appropriation separately, and not as a part of participation. To analyze these internal characteristics, I used matrices to contrast the opinions and views of individuals from the groups involved. Graphic displays, such as stakeholder and conceptual maps, facilitated the integration of information across all the variables. Once processed, the information from the needs assessment and the collaborative practice analysis had to be integrated with the other components of this framework. This integration needed to be matched to some reference state, which defined successful collaboration and adequate conditions for coalition establishment. Both the strong context dependency of CFC processes and the paucity of literature on the subject limited the composition of such benchmarks for this study. Since part of this analysis refers to intra-organizational dynamics and conditions, assessment tools geared toward institutional evaluations of non-profit organizations can also be used as benchmark references (Smutylo, 2001). The Nature Conservancys Institutional Self-Assessment provided a good example (Devine et al., 2001). External Factors The objective of this component was to identify key external constraints to the coalition and the main cross-scale interactions shaping its emergence. Based on the CFC cases analyzed above in this chapter, coalitions appear to be driven in a top-down manner by community-support groups (government agency, NGOs), not the result of strict bottom-up community initiative. Whether this is

PAGE 55

43 positive or negative would remain to be defined in each case, but undoubtedly external agents, institutions and markets have a significant role in the shape and purpose of CFCs. Policies and legislation delineate the use and appropriation of resources and also set the basic financial, fiscal, and structuring rules for social organization. Regulations represent the hand of the government (federal and state) over community activities. International agents such as foreign governments became part of the External Factor arena by providing funding for projects and markets for forest products. Figure 2.5 represents the array of external agents and the diverse links that tied them to the community forestry activity. This model is based on similar cross-scale, hierarchy-based methodologies proposed to study NRM in Amazonia (Schmink & Wood, 1987; Wood, 2003). In Chapter Four, I will complete this diagram with the actual stakeholders from the GPF case. \000\000\000\000\000\000\000\000\000\000\000\000\000 \000\000\000\000\000\000\000\000\000\000\000\000\000 \000\000\000\000\000\000\000\000\000\000\000\000\000 \000\000\000\000\000\000\000\000\000\000\000\000\000 Community Forestry Coalition donor donor Donor Federal Gov. AgencyState Gov. AgencyCommunity timberprojects NGO NGO State Gov. Agency PRIVATE Co. RegionalMarket International Market. Local Market. Figure 2.5: CFC External Agents

PAGE 56

44 Variables and analysis I selected three categories for studying the external environment for the GPF: policies and laws, markets, and stakeholders. I identified three variables of law and policy that I found relevant for my case: land tenure, forest management and social organization. Land tenure legislation defines the appropriation of the natural resources. This is the key structuring element in the resource use system. Forest management regulations define the resource use process, and are the principal instrument to implement government conservation policies. Finally, social organization laws and policies provide the framework for collective action, setting economic and political limits to what community associations can do. Market analysis in this study was limited both conceptually and operationally. Conceptually, I am referring exclusively to the market for sustainably managed timber coming from the community projects. Operationally, I have limited myself to the same dataset that I used for the rest of the study, consisting mainly of the interviews with parties involved in the GPF initiative and other organizations related to community forestry. Due to time constraints I did not look into the actual financial market information. I constructed a vision of the timber markets based on the interviews and the literature, which did not directly address timber market conditions. Asking the GPF-involved individuals from the PO to tell me who were the key external players in the community forestry activity, I identified external

PAGE 57

45 stakeholders 12 I describe the role of the relevant stakeholders and diagram their role in the GPF initiative using a stakeholder map. Data for this component were collected through interviews. A specific questionnaire was designed for the external stakeholders, and questions about their role and the impact of policies and regulations were included in the questionnaires for the POs and CAs. Information from documents such as laws, policy papers and project proposals were also be examined. For the analysis, each variable will be considered and presented separately (laws and policies, markets, stakeholders). The interactions between scales will be inferred from the interviews as well as from the analysis of the other factors (internal and design) in the analytical framework. The analysis of the external factors focused on how external variables constrained the GPF initiative that was located at the scale below it. This unidirectional effect was assumed to constitute the main link between the external factors (laws, policies, markets, stakeholders) and the internal (timber forest projects). I hypothesized that external factors encouraged the GPF initiative (occurring at the internal level). Not only did I expect to find an influence from the larger scale aspects to the lower scale, I expected this influence to be positive for the formation of the GPF coalition. This expectation was based on my interpretation of the socio-ecological evolution of the state of Acre, based on Hollings (2002) model of the adaptive 12 This could be described as following a snowball-like methodology. I asked my interviewees who else I should interview regarding community forestry and the GPF initiative that were not part of the timber projects. A total of 15 individuals were mentioned, of which I interviewed 10.

PAGE 58

46 cycle. According to my interpretation of this cycle for the case of Acre, the situation in the late nineties and early twenty-first century was one that was very open to innovations coming from the ground up. This was the case with the RESEX that rose from a social movement proposal in the late eighties into a nationally sanctioned and internationally acclaimed institution during the nineties. I anticipated that the conditions that enabled the development of the RESEX would still be present in 2002. rk Vulcanization 1844 Pacific Road? Rubber Tapers Social Movement 80s Resex 1989 Acre War 1903 Asian Rubber 1910 WWII State Dev.Plans 60s 1.Starting point: the Boom2.The Bust lethargic years 3.State Driven Integration4. Contesting Integration Figure 2.6: Historic Trajectory Modeled on the Adaptive Cycle. Figure 2.6 displays the main events, as detailed in the timeline in Chapter One, using the adaptive cycle diagram developed by C.S. Holling to describe the trajectory of complex socio-ecological systems. The figure begins with the invention of vulcanization, which made possible the commercial use of natural rubber in industrial and commercial products. The subsequent industrialization of the late nineteenth century (and the development of automobiles in particular) increased the demand for rubber, bringing international attention to the Amazon.

PAGE 59

47 This attention resulted in the booming of the local economies in Amazonia, through the expansion of the rubber extraction operations. This expansion relates to what Holling describes as the accumulation phase of the system; in the case of Acre (and the rest of Amazonia) it lasted until the 1910s. At the end of this decade, rubber plantations in East Asia took over the market, signaling the beginning of the bust period in the Amazonian regions economy. Here the system flipped downward, after the economic collapse, and remained in a disorganized state until the late sixties when the military government began its strong intervention to develop the Amazon. An important point to make here is that the war between Brazil and Bolivia that resulted in the creation of Acre, significant event as it was, did not represent enough of a perturbation to bring about a change in the socio-ecological systems trajectory. The land use pattern, as well as the economy, remained the same. In the late sixties, under the influence of the governments policies, the system began slowly to reorganize. However, the dramatic impact of these policies on the western Amazonian landscape was not truly felt until the eighties. By this time the political context (global, national and regional) was no longer favorable for the military governments plans. Democracy had been re-instated, international concern for the environment was growing, and local, grassroots social movements were articulated around the protection of the forest and the populations that used it (the rubber tappers). The intense cross-scale interactions of this period were expressed in the form of social conflict for land. State and international mediation to resolve this conflict resulted in the

PAGE 60

48 emergence of innovative land tenure institutions such as the RESEX. The innovations in land tenure opened an opportunity for a different land use pattern and a new economy to emerge, bringing the system back to begin a new r-phase in its cycle. This new land use type would have standing forest as the source of a diversified economy of sustainable products and services. However, this opportunity had yet to deliver substantial results. The undertow of the cattle-centered development promoted by the national government since the sixties was very strong. Infrastructural improvements necessary for any kind of regional development seemed to reinforce the cattle economy more positively than the forest one. The state government found itself in a very difficult position: on the one hand, creating new infrastructure, while on the other building a new institutional capacity capable of enforcing forest economy policies. Whether the government succeeds and interest in cattle production decreases, remains to be seen. It is at this point that the figure ends with the question about how the impacts of the road building plans will affect the future trajectory of the next accumulation phase. This is the context in which the community timber activities were taking place. As with the internal component, the characteristics constituting an enabling environment that would foster a successful coalition process have not yet been defined. However, these characteristics can be extrapolated from successful examples such as the Mexican Plan Piloto Forestal a state government strategy that took advantage of the federally sanctioned Ejido common property system to

PAGE 61

49 successfully promote community forestry enterprises. The state level policy worked alongside other initiatives, like a long-term technical aid program. Design Factors This component of the research was based on the assumption stated by Krishna that the design of the institutions delivering public goods can influence the level of social capital (Krishna, 2003). In my case the institution was an organization (a coalition) and I believed its design would influence the amount and the quality of collaboration (i.e. social capital) among the involved parties. Here, I was interested in how the way the coalition was designed and implemented would affect collaboration among parties. Powell (1990), in his study of network forms of organization, pointed out that while hierarchical organizations were the result of human design and markets were the result of human action (not of design); network forms of organization (like coalitions) were the result of both action and design (Powell, 1990). So far in this framework I have presented the actions that could potentially shape the GPF coalition. This includes actions by both the internal and external stakeholders, in a variety of categories: participation, leadership, relationships, management, training, policies, laws, enforcement, and funding among others. This third, and last, component of my analytical framework studied the coalitions design, as the context in which these actions took place. The design would affect the coalition, but not the laws policies, etc, which were external.

PAGE 62

50 Variables and analysis Design, according to Powell (1990), refers to the way in which a network is planned and made. Based on this definition I considered two dimensions of coalition design: concept and process. Concept design refers to the idea, or definition of the coalition, and considers mostly structural and strategic aspects such as objective, goals, members and structure. Process design is the strategy used to promote the coalition idea; this includes looking at the how the idea was presented to the interested parties and how it evolved as it was appropriated (or not) by those parties. To identify the concept design elements in the GPF case, I used interview responses and the few documents that had, at the time of my fieldwork, been written about the GPF. Process design elements were collected in a similar way. To analyze the information regarding concept, I developed an ideal model of the GPF structure and contrasted this with another model based on its actual functioning. While the ideal model was extracted from the documents and answers given by the GPF promoters, the real model was built from my participant observation of the functioning of the GPF and the critical assessment provided by some of the interviewees, particularly those external stakeholders familiar with the GPF proposal but not directly involved in the process. To present the evolution of the GPF process used a timetable matrix. I used an innovation system perspective to analyze the process design of the GPF coalition. An innovation system perspective is a view of innovation processes, which sees innovation as being produced by networks of actors that

PAGE 63

51 co-evolve with the technologies (in this case social technologies) they generate. Co-evolution occurs as a result of iterative experiential learning between the actors involved. Successful innovation results from strong interactions and knowledge flows within the network (Douthwaite et al., 2003). In the context of innovation systems, the idea of a coalition encompasses several things. First, it is an inventionan idea or set of ideas that sparks the innovation process. Then it becomes an innovation process as the idea starts to be socialized. An innovation process is driven by how the proposal serves immediate problems confronted by parties. The second element is its legitimacy, which is dependent upon the qualities of the proponents and the way they promote the idea. The important aspect to remember is that the initial proposal design and the success of its diffusion (or up-scaling) go hand in hand. This requires that design be viewed both in its structural and procedural dimensions. The guiding principle is that in order to have up-scale success you need to understand process (Ashby, 2003). New ideas about NRM organizations such as adaptive management (Gunderson, 2003; Lee, 1993; Salafsky et al., 2001) and social learning (Buck et al., 2002) provide a strong characterization of what an ideal coalition process and structure should look like. The emphasis of these frameworks is on how the organizations elements foster learning and deal with uncertainty and change in their natural and social environment. Although learning and adapting abilities are typical characteristics of successful coalition cases (Coughlin et al., 1999; Fisher

PAGE 64

52 et al., 2001; Monge, 2001), purposefully designed cases for these characteristics are rare. Nevertheless, different efforts on a variety of scales are being made. As an example, the Biodiversity Support Program (BSP), a consortium of conservation organizations, prepared a detailed guide to design adaptive and learning projects based on field research (Margoluis & Salafsky, 1998). Similarly, the Center for International Forestry Research (CIFOR) conducted research on adaptive co-management of community forestry projects to profile such characteristics (Cunha, 2002). One of the sites for the latter project served as a case for my research, as well. I planned to adapt the CIFOR project-scale guides to evaluate the GPF coalition ex-ante. A more detailed set of assessment criteria was provided by Douthwaite (2002) based on his guidelines for a learning-selection innovation model. These criteria look at the relationship between innovation promoters and their target adopters (Table 2.3). Table 2.3:Elements of the Innovation Process Start with a plausible promise Find a product champion Keep it simple, stupid (KISS) Work with innovative and motivated partners Work with pilot site or sites where the need for innovation is great Set up open and unbiased selection mechanisms Dont release the innovation too widely too soon Dont patent anything unless it is to stop attempts trying to privatize technology Realize that culture makes a difference Know when to let go (Douthwaite, 2002)

PAGE 65

53 According to the author, a flexible initial design will facilitate the engagement of the interested parties. As these parties go about changing and transforming the initial idea, they begin to appropriate it. The level of appropriation will signal the degree of success of the idea. Douthwaite also points to the fact that the proposed innovation must demonstrate its usefulness upfront. It should solve an immediate problem or provide a direct benefit to the adopters. Integrating the Analysis In order to study the factors that conditioned the successful emergence of a coalition I laid out a three-layered framework based on a hierarchical view of the coalition phenomenon (see Table 2.2). On a higher level, I clustered the institutions and policies that constrained the community timber management project. The focus here was on the specific top-down links (laws, policies, etc) conditioning collaboration among projects stakeholders. In the middle of this system, I placed the study of the groups and functional communities that were interested in becoming part of the proposed coalition. I drew specific attention to characterizing the collaborative practices occurring within and between these groups. Below, I placed the analysis of the coalition itself; looking at the conceptual proposal and actual process, I considered the structural and functional design of the GPF and its early evolution stages. Bringing the information together to form an adequate picture of the situation of a coalition required considering the cross-relations between the factors. For this purpose, Table 2.4 presents a grid that brings together the organized variables by category, along with a set of ideal conditions or benchmarks to identify the quality of each category of variables. A final column

PAGE 66

54 should include the findings from the particular case analyzed. For this chapter I have left this last column empty, to be filled in the conclusion chapter with the data from the fieldwork. The matrix tool, once completed with the information from the studied case (or cases), was meant to facilitate the analysis of relationships between the different components. The variety of these relationships depends on the type and number of cases available, as well as the quality of the information available. Table 2.4: Integrating Matrix Categories Factors (Variables) Ideal Condition or Benchmark Case Studied At the Group level: o Participation o Leadership o Management Capacity Need for a coalition should be explicit among group members Time should be available to participate in coalition building Trust should be strong among members and partners. Internal: At the Project Level: o Partner relationships o Capacity development o Appropriation by community Projects should foster social learning Projects should encourage appropriation by the community Conflict management capacity should be a developed skill External: State and federal government agencies, policies and laws on conservation, forestry development and social organization NGO and Aid agencies Forest products markets Laws and policies should secure land and resource use rights for the communities Credit and technical support should be available for the community enterprises Stable long term relations should exist between communities and NGOs and government agencies Coalition Design: o Objectives o Structure o Leadership o Resources Structural design should be flexible and open to change Initial proposal should provide room for adaptation by the potential members Design: Innovation Process o History o Perceptions o Adaptations Process design should provide room for adaptation by the potential members Process should have an explicit learning and reflection strategy with built-in mechanisms for monitoring and evaluation Although some of the relationships will not be clear until the matrix is completed, it is important to point to some foreseeable links, starting with the comparison between the ideal condition and the information from the case study.

PAGE 67

55 To clarify, ideal conditions (column three, Table 2.4) are presented in a synthetic form, using key qualifying words by which they should be observed. The comparison between the ideal and the real conditions required attention to the context and history of each coalition. Although some general stages can be identified in the evolution of a coalition (Edwards et al., 2001), not all cases follow the same order; therefore they cannot be associated with the same set of characteristics. Once this is considered, a more limited comparison can still provide valuable information. Another important relation to keep in mind is between aspects of coalitions that bring together variables from different categories and at different scales. An example could be the role inequality (or equality) could play in shaping the proposed coalition. In order to assess this question, different scales of action must be analyzed: intra-community; intra-organization; between community and organizations; between the project champions and those more passive or skeptical; along gender lines. Although inequality might be fostered through structural aspects such as laws, policies and finances, at a coalition level it is an issue less to do with structure and more to do with process. There is no coalition structure that would guarantee full participation or equity. These are issues that need to be overseen constantly, through monitoring and evaluation. The goal is to aim constantly for more equity and participation. Just how process and structure should respond to existing inequalities is contingent upon the particular case.

PAGE 68

56 One of the goals of the analysis was to provide feedback on an on-going process with the analysis results. I was examining the conditions for the emergence of a coalition in a range of time when that proposal had already been discussed and some actions have been taken to build it. At that moment existing proposed structures of the group needed to be tested, along with the programmatic aspects such as objectives and activities. The information about external and internal factors should help to establish how the proposed operative design could be adapted to deliver the purpose of the coalition. The adequacy of the current design can be assessed based on two elements: the efficiency of the design (and the coalition idea in general) to satisfy concrete needs of the participant groups, and the adaptive capacity (resilience to unexpected change) of the current process-structure design. The coalitions purpose has to respond to the concrete reality of the participants, solving relevant problems and providing direct tangible benefits. While demonstrating that it is efficient in reaching its purpose, its structure has to be flexible, adaptive, and open to change. One way to achieve such flexibility is by purposefully leaving things undefined. These aspects will be filled out in the process of putting the original idea into action. This could include things like leaving open the type of activities the coalition could assume or testing different decision-making structures. The valuable thing is the awareness with which those things are left undefined. This facilitates the social learning process and

PAGE 69

57 the accounting of these loopholes as strategic decisions in an evaluation process. In an organic process, unrecognized loopholes will require adaptive responses as well, but the process of generating those adaptations could be very costly. The cost might come in the form of internal conflicts among participants or as unbudgeted issues that distort available funding. Personal and group emotional conditions harnessing uncertainty and surprise play a key role in how adaptations are generated. Under a rigid coalition structure an unexpected event may cause conflict and undermine collaboration. Conflict dynamics normally include different degrees of blaming and scape-goating that, depending upon the collective emotional resilience, could destroy the coalition or severely hamper reciprocal trust. Under a negative emotional environment structural, causes of problems are overshadowed by personalized accusations or explanations. An issue such as the rigidity of decision-making structures is hard to bring up in the middle of a polarized dispute. Even in the aftermath of conflicts, the opportunity to learn is missed because the group might not have the adequate social learning platforms(Buck et al., 2002) to process it. In a coalition, learning platforms and emotional awareness are essential, given that the added value of participating in a coalition should be its efficiency to foster dialog, joint problem solving and interactions. An open design also allows appropriation to take place. As with technological innovations, part of the success comes from participation. The

PAGE 70

58 greater chance interested parties have to define what the coalition becomes, the more successful the collaborative proposal will be. Analytical Framework Limitations Thus far, I have identified three major limitations of this analytical framework. First, biophysical aspects were mentioned but not developed into a full variable. Second, market analysis was considered in very narrow terms. And third, technical forestry and timber managements impact on the organizational behavior were only indirectly contemplated. Despite the need to incorporate the study of biophysical aspects in the assessment of external conditions to the emergence of the coalition, I was unable to develop this variable in the analysis. I was not able to identify which biophysical aspects and at what scales needed to be incorporated. Literature analyzing collaborative efforts from an environmental conflict management perspective presented the same void (Brick et al., 2001; Wondolleck & Yaffee, 2000). Even among literature on community conservation and conservation project development the role of biophysical aspects in organizational development has not been well documented (Borrini-Feyerabend et al., 2001; Clay, 2001). I reproduced this handicap without shedding light onto what could be the potential bridges towards this connection between ecological environment and scale of social organization. Indirectly it is clear that there is a relationship, evident in the current trend to shift from a site-based NRM to an ecosystemor landscape-scale management. Establishing this connection extends beyond the capacity of this study.

PAGE 71

59 The effect of market dynamics on the potential success of a CFC is very clear. Community forestry activities are mostly oriented towards selling forest products to local, regional or international markets. Nevertheless the methodology of this framework does not explore the correspondence between the characteristics of the market and those of the organizations. Empirically, it is known that part of the motivation to collaborate among community forestry enterprises comes from the need to take advantage of the market benefits of belonging to a larger group. The methodologys emphasis on process issues and interpersonal relations overlooks the role of material conditions such as markets. The problem of incorporating market factors adequately is methodological: what information to retrieve? How to relate to the emergence of the coalition? How does its relation to the market affect the coalition? The degree to which the analytical framework is tailored towards community forestry was given by the literature review. Literature on community forestry organizations establishes some explicit links between issues such as property rights and type of product extracted from the forest. Nevertheless, the most important aspect defining structure of organization is function, and function is defined by the characteristics of the activity undertaken. The way an NTFP enterprise is conducted is substantially different from a timber projectthe structure of organization should be as well. I would assume that those differences in operation and organization would affect the way a CFC would be organized if it were composed of NTFP or of timber projects. However the framework does not account explicitly for these differences. I expected that the

PAGE 72

60 management activitys effect on the organization would come from the data, but that may or may not be the casethe framework does not target this issue directly. Synthesis This chapter began by characterizing the concept of community forestry coalitions. It also presents the analytical framework built to analyze the case of an emerging coalition: the GPF in Acre and Rondnia. The analytical framework was presented in terms of the factors within three components: internal, external and design. The variables of each component were then described, as were the analytical tools used. A final section presented the integration of the research framework. It is through the integration of the research components that the research hypothesis on the hampering condition of the internal factors will be tested. It will provide a complete vision of the constraints and opportunities for the GPF coalition success. The next three chapters present the results from each of the components. The first to be addressed are the internal factors affecting the GPF initiative.

PAGE 73

CHAPTER 3 INTERNAL FACTORS As a first step to map the conditions for inter-project collaboration among the community timber projects in Acre and Rondnia, I started out by identifying the characteristics of intra-project collaboration. I was interested in identifying the different groups collaborative behaviors, as these were critical to identify the common pattern of collaboration that would inform the GPF initiative. I had stated in my hypothesis that internal factors would be the source of the major constraints for the development of the GPF, while the external (chapter 4) and design factors (chapter 5) would be positive for the emergence of the coalition. I used the theory of communities of practice as a metaphor to guide this assessment. I considered each community-NGO (or government agency) as an actual community-of-practice of collaboration, given that this was the type of activity I was interested in studying. Through this theoretical framework, I chose to see each project as a particular system, defined by its own experience of collaboration. I evaluated the effectiveness of each case, and also the similarities among them, to assess the role that the collaboration experience would have in the success of the GPF. To do this, I interviewed individuals from all organizations involved in the timber projects: sixteen from the four community groups and fourteen from the partner organizations. Based on my analytical framework I focused the interviews on six variables: leadership, participation, management 61

PAGE 74

62 capacity, project relationships, project training focus and project appropriation by the community. While the first three considered each organization as a whole and included its involvement in the timber projects, the second three focused exclusively on their experience with the timber projects. The results from this chapter confirmed my hypothesis about the weak management capacity of the community associations. The complexity of the timber operations seemed to overwhelm the community participants, and their learning curve seemed to be longer than was expected in the projects original plans. The external pressure to show results (particularly profits) led the partner organizations to maintain a stronger control over the operation, creating a negative paternalistic-bias in the projects collaboration pattern. However a conscious social learning process in some of the projects was leading them to experiment with broader community participation in the period from 2001 to 2002 when this research was conducted. This chapter is organized in three larger sections. The first describes the stakeholders involved in the GPF. The second presents a comparison of the four projects organizations on the set of key variables (leadership, participation, management capacity, project relationships, project training focus and project appropriation). The last section discusses the results and synthesizes the main findings of the chapter. The GPF Coalition Members The timber project community interested in participating in the GPF initiative consisted of four projects: Porto Dias, Peixoto and Cachoeira from the state of Acre, and Cautrio from the state of Rondnia. The Porto Dias project involved

PAGE 75

63 The Center for Amazonian Workers (CTA, Centro dos Trabalhadores da Amazonia) and the Association of Rubber Tappers of Porto Dias (ASPD: Associao de Seringueiros do Porto Dias). The Peixoto project was managed by EMBRAPA, Brazilian Agricultural Research Corporation, along with the association of timber managers of Pedro Peixoto (APRUMA). In Cachoeira the local residents association (AMPPAE-CM or AMC) had been supported by two state government agencies: SEFE (State Secretary of Forestry and Extractivism) and SEATER (State Secretary of Rural Extension). In this case the local group also had a close relationship with two timber processing companies, the Polo Moveleiro and AVER. Cautrio, the timber project in Rondnia, was a joint effort of ECOPOR, a local NGO, and the AGUAPE rubber tappers association. In this case the Rubber Tappers Organization of Rondnia (OSR) also played an important role. I will describe these projects and organizations in more detail in the following sections. A synthesis of the potential GPF members is presented below in the form of a stakeholder map (Figure 3.1). It important to appreciate that the GPF was an initiative to build a platform of interaction among four different community timber projects, but these projects were in themselves collaborative initiatives that each involved at least two very different organizations. Every case included a grassroots community association and also a set of partner organizations providing technical assistance. As will be discussed later, these support organizations had a large stake in these projects and exercised a high level of control over them. Although projects were presented as community timber

PAGE 76

64 projects, and the GPF was presented as a community coalition, the partner organizations actually had much initiative and decision making power. To represent this situation, figure 3.1 presents each project with both their local and technical partner, and the GPF coalition as an initiative that would bring these different partnerships into a broader umbrella organization. \000\000\000\000\000\000\000\000\000\000\000\000\000\000)TjETEMC /P <>BDC Qq175.86 554.70001 m173.03998 554.64001 l170.45999 554.40002 l167.75998 553.91998 l165.23999 553.38 l162.84 552.53998 l160.73999 551.64001 l158.58 550.62 l156.66 549.41998 l154.92 548.15997 l153.36 546.77997 l151.98 545.27997 l150.90001 543.83997 l150 542.10004 l149.28 540.35999 l148.86 538.62 l148.73999 536.82001 l148.86 534.95996 l149.28 533.21997 l150 531.53998 l150.90001 529.79999 l151.98 528.29999 l153.36 526.79999 l154.92 525.53998 l156.66 524.27997 l158.58 523.14001 l160.73999 522.12 l162.84 521.15997 l165.23999 520.38 l167.75998 519.77997 l170.45999 519.35999 l173.03998 519.12 l175.86 519 l178.67999 519.12 l181.44 519.35999 l184.01999 519.77997 l186.48 520.38 l188.87999 521.15997 l190.92 522.12 l193.08 523.14001 l195 524.27997 l196.67998 525.53998 l198.3 526.79999 l199.62 528.29999 l200.75998 529.79999 l201.65999 531.53998 l202.31999 533.21997 l202.8 534.95996 l202.92 536.82001 l202.8 538.62 l202.31999 540.35999 l201.65999 542.10004 l200.75998 543.83997 l199.62 545.27997 l198.3 546.77997 l196.67998 548.15997 l195 549.41998 l193.08 550.62 l190.92 551.64001 l188.87999 552.53998 l186.48 553.38 l184.01999 553.91998 l181.44 554.40002 l178.67999 554.64001 l175.86 554.70001 lW* n0 0 0 scnBT/T3_0 1 Tf0 Tw 1.92 0 0 -11.52 147.84 552.95996 Tm(\000\000\000\000\000\000\000\000\000\000\000\000\000\000)TjETEMC /P <>BDC Qq175.86 554.70001 m173.03998 554.64001 l170.45999 554.40002 l167.75998 553.91998 l165.23999 553.38 l162.84 552.53998 l160.73999 551.64001 l158.58 550.62 l156.66 549.41998 l154.92 548.15997 l153.36 546.77997 l151.98 545.27997 l150.90001 543.83997 l150 542.10004 l149.28 540.35999 l148.86 538.62 l148.73999 536.82001 l148.86 534.95996 l149.28 533.21997 l150 531.53998 l150.90001 529.79999 l151.98 528.29999 l153.36 526.79999 l154.92 525.53998 l156.66 524.27997 l158.58 523.14001 l160.73999 522.12 l162.84 521.15997 l165.23999 520.38 l167.75998 519.77997 l170.45999 519.35999 l173.03998 519.12 l175.86 519 l178.67999 519.12 l181.44 519.35999 l184.01999 519.77997 l186.48 520.38 l188.87999 521.15997 l190.92 522.12 l193.08 523.14001 l195 524.27997 l196.67998 525.53998 l198.3 526.79999 l199.62 528.29999 l200.75998 529.79999 l201.65999 531.53998 l202.31999 533.21997 l202.8 534.95996 l202.92 536.82001 l202.8 538.62 l202.31999 540.35999 l201.65999 542.10004 l200.75998 543.83997 l199.62 545.27997 l198.3 546.77997 l196.67998 548.15997 l195 549.41998 l193.08 550.62 l190.92 551.64001 l188.87999 552.53998 l186.48 553.38 l184.01999 553.91998 l181.44 554.40002 l178.67999 554.64001 l175.86 554.70001 lW* n0 0 0 scnBT/T3_0 1 Tf0 Tw 1.92 0 0 -11.52 147.84 541.44 Tm(\000\000\000\000\000\000\000\000\000\000\000\000\000\000)TjETEMC /P <>BDC Qq175.86 554.70001 m173.03998 554.64001 l170.45999 554.40002 l167.75998 553.91998 l165.23999 553.38 l162.84 552.53998 l160.73999 551.64001 l158.58 550.62 l156.66 549.41998 l154.92 548.15997 l153.36 546.77997 l151.98 545.27997 l150.90001 543.83997 l150 542.10004 l149.28 540.35999 l148.86 538.62 l148.73999 536.82001 l148.86 534.95996 l149.28 533.21997 l150 531.53998 l150.90001 529.79999 l151.98 528.29999 l153.36 526.79999 l154.92 525.53998 l156.66 524.27997 l158.58 523.14001 l160.73999 522.12 l162.84 521.15997 l165.23999 520.38 l167.75998 519.77997 l170.45999 519.35999 l173.03998 519.12 l175.86 519 l178.67999 519.12 l181.44 519.35999 l184.01999 519.77997 l186.48 520.38 l188.87999 521.15997 l190.92 522.12 l193.08 523.14001 l195 524.27997 l196.67998 525.53998 l198.3 526.79999 l199.62 528.29999 l200.75998 529.79999 l201.65999 531.53998 l202.31999 533.21997 l202.8 534.95996 l202.92 536.82001 l202.8 538.62 l202.31999 540.35999 l201.65999 542.10004 l200.75998 543.83997 l199.62 545.27997 l198.3 546.77997 l196.67998 548.15997 l195 549.41998 l193.08 550.62 l190.92 551.64001 l188.87999 552.53998 l186.48 553.38 l184.01999 553.91998 l181.44 554.40002 l178.67999 554.64001 l175.86 554.70001 lW* n0 0 0 scnBT/T3_0 1 Tf0 Tw 1.92 0 0 -11.52 147.84 529.91998 Tm(\000\000\000\000\000\000\000\000\000\000\000\000\000\000)TjETEMC /InlineShape <>BDC Q0.937 1 0.659 scn148.73999 518.76001 54.42 35.94 ref/CS0 CS 0 0 0 SCN0.36 w 10 M 1 j 1 J []0 d 202.92 536.82001 m202.92 536.82001 l202.92 546.65704 190.79701 554.64001 175.86 554.64001 c160.923 554.64001 148.8 546.65704 148.8 536.82001 c148.8 526.98303 160.923 519 175.86 519 c190.79701 519 202.92 526.98303 202.92 536.82001 chSEMC /InlineShape <>BDC q108 362.22 324.89999 219.78 reW n1 0.612 0.188 scn392.70001 432.17999 m392.70001 432.17999 l392.70001 470.84802 339.31601 502.22998 273.54001 502.22998 c207.76401 502.22998 154.37999 470.84802 154.37999 432.17999 c154.37999 393.51202 207.76401 362.13 273.54001 362.13 c339.31601 362.13 392.70001 393.51202 392.70001 432.17999 cfEMC /InlineShape <>BDC 392.78998 432.17999 m392.78998 432.17999 l392.78998 470.94702 339.33902 502.41 273.47998 502.41 c207.62102 502.41 154.17 470.94702 154.17 432.17999 c154.17 393.41302 207.62102 361.94998 273.47998 361.94998 c339.33902 361.94998 392.78998 393.41302 392.78998 432.17999 chSQ360.09 390.84 m360.09 390.84 l360.09 400.59402 348.03403 408.50998 333.17999 408.50998 c318.32602 408.50998 306.26999 400.59402 306.26999 390.84 c306.26999 381.08603 318.32602 373.16998 333.17999 373.16998 c348.03403 373.16998 360.09 381.08603 360.09 390.84 cf360.20999 390.78 m360.20999 390.78 l360.20999 400.617 348.10101 408.60001 333.17999 408.60001 c318.259 408.60001 306.14999 400.617 306.14999 390.78 c306.14999 380.94302 318.259 372.95999 333.17999 372.95999 c348.10101 372.95999 360.20999 380.94302 360.20999 390.78 chS257.04001 478.97998 m257.04001 478.97998 l257.04001 488.75 244.97101 496.67999 230.10001 496.67999 c215.229 496.67999 203.15999 488.75 203.15999 478.97998 c203.15999 469.20999 215.229 461.28 230.10001 461.28 c244.97101 461.28 257.04001 469.20999 257.04001 478.97998 cf257.22 478.97998 m257.22 478.97998 l257.22 488.84998 245.06998 496.85999 230.10001 496.85999 c215.12999 496.85999 202.98 488.84998 202.98 478.97998 c202.98 469.10999 215.12999 461.10001 230.10001 461.10001 c245.06998 461.10001 257.22 469.10999 257.22 478.97998 chS192.06 367.20001 54.53999 36 ref1 1 1 scn219.17998 403.20001 m216.36 403.07999 l213.78 402.84 l211.08 402.35999 l208.56 401.94 l206.22 401.10001 l204.06 400.20001 l201.90001 399.17999 l199.98 397.97998 l198.3 396.72 l196.67998 395.34 l195.36 393.89999 l194.22 392.39999 l193.31999 390.66 l192.60001 388.91998 l192.17999 387.17999 l192.06 385.38 l192.17999 383.51999 l192.60001 381.78 l193.31999 380.10001 l194.22 378.35999 l195.36 376.85999 l196.67998 375.35999 l198.3 373.97998 l199.98 372.72 l201.90001 371.57999 l204.06 370.56 l206.22 369.60001 l208.56 368.81998 l211.08 368.28 l213.78 367.80002 l216.36 367.56 l219.17998 367.44 l222.06 367.56 l224.75998 367.80002 l227.34 368.28 l229.86 368.81998 l232.2 369.60001 l234.36 370.56 l236.51999 371.57999 l238.43999 372.72 l240.12 373.97998 l241.73999 375.35999 l243.06 376.85999 l244.2 378.35999 l245.10001 380.10001 l245.75998 381.78 l246.23999 383.51999 l246.36 385.38 l246.23999 387.17999 l245.75998 388.91998 l245.10001 390.66 l244.2 392.39999 l243.06 393.89999 l241.73999 395.34 l240.12 396.72 l238.43999 397.97998 l236.51999 399.17999 l234.36 400.20001 l232.2 401.10001 l229.86 401.94 l227.34 402.35999 l224.75998 402.84 l222.06 403.07999 l219.17998 403.20001 lf*EMC /P <>BDC q219.17998 403.20001 m216.36 403.07999 l213.78 402.84 l211.08 402.35999 l208.56 401.94 l206.22 401.10001 l204.06 400.20001 l201.90001 399.17999 l199.98 397.97998 l198.3 396.72 l196.67998 395.34 l195.36 393.89999 l194.22 392.39999 l193.31999 390.66 l192.60001 388.91998 l192.17999 387.17999 l192.06 385.38 l192.17999 383.51999 l192.60001 381.78 l193.31999 380.10001 l194.22 378.35999 l195.36 376.85999 l196.67998 375.35999 l198.3 373.97998 l199.98 372.72 l201.90001 371.57999 l204.06 370.56 l206.22 369.60001 l208.56 368.81998 l211.08 368.28 l213.78 367.80002 l216.36 367.56 l219.17998 367.44 l222.06 367.56 l224.75998 367.80002 l227.34 368.28 l229.86 368.81998 l232.2 369.60001 l234.36 370.56 l236.51999 371.57999 l238.43999 372.72 l240.12 373.97998 l241.73999 375.35999 l243.06 376.85999 l244.2 378.35999 l245.10001 380.10001 l245.75998 381.78 l246.23999 383.51999 l246.36 385.38 l246.23999 387.17999 l245.75998 388.91998 l245.10001 390.66 l244.2 392.39999 l243.06 393.89999 l241.73999 395.34 l240.12 396.72 l238.43999 397.97998 l236.51999 399.17999 l234.36 400.20001 l232.2 401.10001 l229.86 401.94 l227.34 402.35999 l224.75998 402.84 l222.06 403.07999 l219.17998 403.20001 lW* n0 0 0 scnBT/T3_0 1 Tf0 Tw 1.92 0 0 -11.52 192 403.20001 Tm(\000\000\000\000\000\000\000\000\000\000\000\000\000\000)TjETEMC /P <>BDC Qq219.17998 403.20001 m216.36 403.07999 l213.78 402.84 l211.08 402.35999 l208.56 401.94 l206.22 401.10001 l204.06 400.20001 l201.90001 399.17999 l199.98 397.97998 l198.3 396.72 l196.67998 395.34 l195.36 393.89999 l194.22 392.39999 l193.31999 390.66 l192.60001 388.91998 l192.17999 387.17999 l192.06 385.38 l192.17999 383.51999 l192.60001 381.78 l193.31999 380.10001 l194.22 378.35999 l195.36 376.85999 l196.67998 375.35999 l198.3 373.97998 l199.98 372.72 l201.90001 371.57999 l204.06 370.56 l206.22 369.60001 l208.56 368.81998 l211.08 368.28 l213.78 367.80002 l216.36 367.56 l219.17998 367.44 l222.06 367.56 l224.75998 367.80002 l227.34 368.28 l229.86 368.81998 l232.2 369.60001 l234.36 370.56 l236.51999 371.57999 l238.43999 372.72 l240.12 373.97998 l241.73999 375.35999 l243.06 376.85999 l244.2 378.35999 l245.10001 380.10001 l245.75998 381.78 l246.23999 383.51999 l246.36 385.38 l246.23999 387.17999 l245.75998 388.91998 l245.10001 390.66 l244.2 392.39999 l243.06 393.89999 l241.73999 395.34 l240.12 396.72 l238.43999 397.97998 l236.51999 399.17999 l234.36 400.20001 l232.2 401.10001 l229.86 401.94 l227.34 402.35999 l224.75998 402.84 l222.06 403.07999 l219.17998 403.20001 lW* n0 0 0 scnBT/T3_0 1 Tf0 Tw 1.92 0 0 -11.52 192 391.67999 Tm(\000\000\000\000\000\000\000\000\000\000\000\000\000\000)TjETEMC /P <>BDC Qq219.17998 403.20001 m216.36 403.07999 l213.78 402.84 l211.08 402.35999 l208.56 401.94 l206.22 401.10001 l204.06 400.20001 l201.90001 399.17999 l199.98 397.97998 l198.3 396.72 l196.67998 395.34 l195.36 393.89999 l194.22 392.39999 l193.31999 390.66 l192.60001 388.91998 l192.17999 387.17999 l192.06 385.38 l192.17999 383.51999 l192.60001 381.78 l193.31999 380.10001 l194.22 378.35999 l195.36 376.85999 l196.67998 375.35999 l198.3 373.97998 l199.98 372.72 l201.90001 371.57999 l204.06 370.56 l206.22 369.60001 l208.56 368.81998 l211.08 368.28 l213.78 367.80002 l216.36 367.56 l219.17998 367.44 l222.06 367.56 l224.75998 367.80002 l227.34 368.28 l229.86 368.81998 l232.2 369.60001 l234.36 370.56 l236.51999 371.57999 l238.43999 372.72 l240.12 373.97998 l241.73999 375.35999 l243.06 376.85999 l244.2 378.35999 l245.10001 380.10001 l245.75998 381.78 l246.23999 383.51999 l246.36 385.38 l246.23999 387.17999 l245.75998 388.91998 l245.10001 390.66 l244.2 392.39999 l243.06 393.89999 l241.73999 395.34 l240.12 396.72 l238.43999 397.97998 l236.51999 399.17999 l234.36 400.20001 l232.2 401.10001 l229.86 401.94 l227.34 402.35999 l224.75998 402.84 l222.06 403.07999 l219.17998 403.20001 lW* n0 0 0 scnBT/T3_0 1 Tf0 Tw 1.92 0 0 -11.52 192 380.16 Tm(\000\000\000\000\000\000\000\000\000\000\000\000\000\000)TjETEMC /P <>BDC Qq219.17998 403.20001 m216.36 403.07999 l213.78 402.84 l211.08 402.35999 l208.56 401.94 l206.22 401.10001 l204.06 400.20001 l201.90001 399.17999 l199.98 397.97998 l198.3 396.72 l196.67998 395.34 l195.36 393.89999 l194.22 392.39999 l193.31999 390.66 l192.60001 388.91998 l192.17999 387.17999 l192.06 385.38 l192.17999 383.51999 l192.60001 381.78 l193.31999 380.10001 l194.22 378.35999 l195.36 376.85999 l196.67998 375.35999 l198.3 373.97998 l199.98 372.72 l201.90001 371.57999 l204.06 370.56 l206.22 369.60001 l208.56 368.81998 l211.08 368.28 l213.78 367.80002 l216.36 367.56 l219.17998 367.44 l222.06 367.56 l224.75998 367.80002 l227.34 368.28 l229.86 368.81998 l232.2 369.60001 l234.36 370.56 l236.51999 371.57999 l238.43999 372.72 l240.12 373.97998 l241.73999 375.35999 l243.06 376.85999 l244.2 378.35999 l245.10001 380.10001 l245.75998 381.78 l246.23999 383.51999 l246.36 385.38 l246.23999 387.17999 l245.75998 388.91998 l245.10001 390.66 l244.2 392.39999 l243.06 393.89999 l241.73999 395.34 l240.12 396.72 l238.43999 397.97998 l236.51999 399.17999 l234.36 400.20001 l232.2 401.10001 l229.86 401.94 l227.34 402.35999 l224.75998 402.84 l222.06 403.07999 l219.17998 403.20001 lW* n0 0 0 scnBT/T3_0 1 Tf0 Tw 1.92 0 0 -11.52 192 368.63998 Tm(\000\000\000\000\000\000\000\000\000\000\000\000\000\000)TjETEMC /InlineShape <>BDC Q0.937 1 0.659 scn192.06 367.20001 54.53999 36 ref246.36 385.31998 m246.36 385.31998 l246.36 395.173 234.20999 403.16998 219.23999 403.16998 c204.26999 403.16998 192.12 395.173 192.12 385.31998 c192.12 375.46701 204.26999 367.47 219.23999 367.47 c234.20999 367.47 246.36 375.46701 246.36 385.31998 chS343.91998 478.97998 m343.91998 478.97998 l343.91998 488.75 331.85101 496.67999 316.97998 496.67999 c302.10901 496.67999 290.04001 488.75 290.04001 478.97998 c290.04001 469.20999 302.10901 461.28 316.97998 461.28 c331.85101 461.28 343.91998 469.20999 343.91998 478.97998 cf344.10001 478.97998 m344.10001 478.97998 l344.10001 488.84998 331.94998 496.85999 316.97998 496.85999 c302.00998 496.85999 289.85999 488.84998 289.85999 478.97998 c289.85999 469.10999 302.00998 461.10001 316.97998 461.10001 c331.94998 461.10001 344.10001 469.10999 344.10001 478.97998 chS0.933 0.161 0.17999 scn223.14 400.67999 m213.65999 384.12 l232.67998 384.12 l223.14 400.67999 lf*0.48 w 223.25998 400.56 m213.78 384 lS0.36 w 213.78 384 m232.8 384 lS0.48 w 232.8 384 m223.25998 400.56 lS315.47998 408.78 m306 392.39999 l324.95999 392.39999 l315.47998 408.78 lf*315.60001 408.72 m306.06 392.28 lS0.36 w 306.06 392.28 m325.07999 392.28 lS0.48 w 325.07999 392.28 m315.60001 408.72 lS301.91998 488.75998 m292.38 472.25998 l311.39999 472.25998 l301.91998 488.75998 lf*302.03998 488.63989 m292.5 472.13992 lS0.36 w 292.5 472.13992 m311.51999 472.13992 lS0.48 w 311.51999 472.13992 m302.03998 488.63989 lS245.58 480.47998 m235.37999 463.97998 l255.65999 463.97998 l245.58 480.47998 lf*245.7 480.41998 m235.5 463.85999 lS0.36 w 235.5 463.85999 m255.72 463.85999 lS0.48 w 255.72 463.85999 m245.7 480.41998 lS0.765 1 0.078 scn195.23999 408.60001 m186.84 395.70001 l199.5 387.12 l208.01999 400.07999 l195.23999 408.60001 lf*195.36 408.47998 m186.95999 395.57999 l199.62 387 l208.14 399.95999 l195.36 408.47998 lS338.34 414.12 m325.20001 405.84 l333.35999 392.57999 l346.5 400.85999 l338.34 414.12 lf*338.45999 414 m325.32001 405.72 l333.47998 392.51999 l346.56 400.73999 l338.45999 414 lS332.22 483.12 m319.67999 474.06 l328.62 461.34 l341.16 470.39993 l332.22 483.12 lf*332.34 483.06 m319.80002 473.94 l328.73999 461.22 l341.28 470.28 l332.34 483.06 lS208.8 483.06 m200.34 470.03998 l213.12 461.45999 l221.58 474.41998 l208.8 483.06 lf*208.92 482.94 m200.40001 469.91998 l213.17998 461.34 l221.7 474.30002 l208.92 482.94 lS1 0.612 0.188 scn156.90001 425.22 224.75998 16.56 refEMC /P <>BDC 0 0 0 scnBT/TT1 1 Tf-0.0119 Tc 0 Tw 11.03999 0 0 11.03999 233.34 429.36829 Tm(GPF Coalition Community Forestry ProjectsCommunity Forestry OrganizationsSupport NGO or Gov. AgencyCommunity-based Organization PeixotoPto. DiasCautrioCachoeiraCTAEMBRAPAECOPORESEATER-SEFEA.Pt D.APRUMAAGUAPEAMPPAE-CM OSR Figure 3.1: Stakeholder Map Following in this section are the profiles from each of these potential member-projects. Each profile presents the timber project and the main partner organizations that teamed up for each project. The profile describes briefly the history of the project, its basic information (type, years, finance, administration) and the significant traits that each project (and each organization) had in terms of leadership, collaboration and management experience. The following section aggregates the information from the four cases to establish the baseline collaboration model on which the GPF would potentially be developed.

PAGE 77

65 Porto Dias Project Port Dias was the first community forestry project to be established in the region. It was started in 1996 with the financial support of PPG7 and ITTO 13 The PAE Porto Dias was an INCRA agro-extractivist project created in 1989 with an area of 22,145 Ha; approximately ninety families live in the settlement. The timber project aimed at developing timber extraction as an economic alternative for rubber-tapper families in the PAE Porto Dias. It was established as a groundbreaking project that could later be replicated in other PAEs and RESEX 14 The idea for this project came from CTA and had already been rejected in other rubber tapper areas because of the negative perception the social movement had about timber extraction. The rubber tappers movement had fought to save the forest and cutting trees seemed an aberrant contradiction. The success in getting the Porto Dias Association to agree to the project came from various internal factors. First, Porto Dias had not been an area of strong conflict over land during the eighties. Therefore it lacked the strong social movement background presented in other areas such as the Alto Acre Region. This fact made people more open to trying timber harvesting as an alternative activity. At the time CTA approached Porto Dias with the idea, economic hardships had caused some families to abandon the PAE, and also more 13 PPG7 is a fund for the conservation of forest in the Amazon basin established by the countries of the G7 group (the 7 most industrialized countries in the world). The Fund is administered by The World Bank and IBAMA (Instituto Brasilero de Meio Ambiente). ITTO is the International Tropical Timber Organization. 14 A second community project supported by CTA in So Lus do Remanso was certified for timber and non-timber products by the FSC in 2004

PAGE 78

66 agriculturally oriented families had come in. This last fact plus the recurrent illegal logging that went on had brought back the discussion of whether this PAE should be transformed into a regular agrarian settlement. Given the institutional abandonment in which INCRA had left the PAE after its creation, the timber project not only represented an economic alternative but, in the short term, it represented an apparently strong external institution supporting rubber tapper families and their way of life. This strengthened the Rubber Tapper Association and their interest in maintaining the extractivist category for this settlement. The timber project had its basis in the reduced impact logging developed by IMAZON and the Tropical Forest Foundation (FFT) in Eastern Amazonia, with a cutting cycle of 30 years and very low volume extracted per hectare (1 to 2 trees)(Capossoli, 1999; Kornexl, 1997). This technical approach was accompanied by a strong capacity building component that aimed at eventually transferring the whole operation to the community. Each community participant, known as a manejador, had to sign, and commit to, a letter of principles that included: keeping their kids in school, participating in training and applying what was learned, willingness to work in other peoples areas, and agreeing that part of the profits should go the association. To guarantee this approach, as well as to take advantage of potential market benefits, the project was submitted for sustainable forestry certification (FSC) and received its seal in 2002.

PAGE 79

67 As early as 1997 a project financial evaluation (Kornexl, 1997) done by the CNPT 15 found that the financial and economic viability of the project was low. The high initial investments as well as the pre-investments (government infrastructure) required made the project a bad investment when compared to the establishment of a Brazil nut processing plant in the site. The evaluator nevertheless supported the project given its innovative characteristics and the potential model that it represented for the rest of the Amazon region and its rural communities. The evaluation assumed a potentially fully implemented project, and did not consider the cost of bringing things from their status quo to a state of full project implementation. It did remark that the costs to move from the initial conditions were significantly high, but it did not quantify these costs. Community organization skills as well as business skills were underlined as areas that needed significant attention in order to have the project running. This evaluation, as well as the projects proposal, relied mostly on training (conducting workshops) to develop these managerial capacities. Centro de Trabalhadores da Amaznia Founded in 1981 as a grassroots, adult education organization to provide support to the rubber tappers social movement on health and education issues, the CTA turned its attention to timber management in 1996 with the first community timber project in Amazonia (Porto Dias). This change came as a result of the need to develop economic alternatives for the rubber tapper families living in the recently created RESEXs and PAEs. The multiple forest use project 15 Conselho Nacional de Povos Tradicionais, the subdivision within IBAMA in charge of the RESEX

PAGE 80

68 was designed with a strong forestry focus concerned primarily with assessing the forest timber potential and preparing the legally required documents such as planos de manejo (management plans to be approved by IBAMA). At the time of the design and initial implementation of this multiple forest use project, CTA was still a small organization led by forest engineers who decided to get the project running as soon as possible. Although the project proposal called for strong community participation throughout the whole process, the complexities associated with the timber activity, and the limited capacity of the community to get involved, left much of the control and execution in the hands of CTA. This established a noticeably vertical relationship between CTA and the community project participants. A few interviewees mentioned that the strong leadership style of the initial project coordinator resembled the behavior of the old patron (seringalista) toward whom the manejadores (project participants) reacted very passively. During this time the project confronted several external challenges such as the lack of IBAMAs approval, which took over a year to obtain leaving the project on hold. IBAMA had not differentiated between small and large timber projects and it was not until 1998 that it established the small-scale timber project guidelines. The negative perception of timber management within the rubber tapper movement also affected the project. As a result of changes in coordination, and significant internal conflicts, the initial technical and paternalistic approach later changed to be more community development oriented.

PAGE 81

69 At the time of this study, CTAs technical team comprised a cohesive group of forest engineers who shared a similar view of the project. They were close friends beyond their work activities. They had gone to college together and came to Acre at almost the same time, around the year 2000. Based on a commitment to improve the rubber tappers living conditions, to the pioneering work developing economic alternatives such as innovative NTFPs and community timber management, CTA was a well-respected organization. Beyond the Porto Dias project work, they were involved in a variety of fora, including the Amazonian Community Forestry Working Group, the Grupo de Trabalho Amazonico(GTA) a consortium of NGOs and other grassroots groups following Amazonian development policies. Associao de Seringueiros de Porto Dias The Associao de Seringueiros de Porto Dias (association of rubber tappers of Porto Dias) was established in 1987 with the purpose of consolidating the Porto Dias agro-extractive settlement (PAE). The association bylaws established a series of objectives for the group among them: the promotion of collective action (associativismo), commercialization of the members products, and conservation of natural resources in the PAE. Membership was restricted to rubber tappers that lived inside PAE Porto Dias. The association had a general assembly that met annually, and a board composed of a president, a secretary, a treasurer and three overseers (fiscales), who were all elected for a period of two years. Among the groups key activities was the management of a community store that received local products, such as rubber and Brazil nuts, from the

PAGE 82

70 members in exchange for house goods and food. The treasurer of the association managed the store. This store reported small gains for the association during the first half of 2002. The group was also in charge of several oxen and mules used to transport rubber, Brazil nuts and timber. It is important to note that only ten members of the association directly participated in the timber project. This group included most of the members from the associations board. As discussed by Stone (2003) the range of local participation in the timber project varied widely. Of the ninety families reported to be living in the settlement (Cunha, 2002), the association had only 24 members 16 (typically male heads of households, that supposedly represented their families). Of this number, only ten association members participated in the timber project. In the six years of the project, a few families left the project. However, not until late 2002 were there discussions regarding bringing new families into the project. Peixoto Project The Peixoto timber project is located in the southeastern corner of the state of Acre between the municipalities of Acrelndia and Senador Guiomard. Pedro Peixoto was one of the largest agrarian development projects in Brazil, with an extension of 378,395 ha and a total of 3000 families. The initial area of the timber project covered 440 ha, but with the addition of new families, this area expanded to 1000 ha. The participants of the projects were mostly farmers; some were originally rubber tappers, but most migrated from the south of the country. 16 The number of members varied slightly between 21 and 24 members. The association treasurer and the secretary mentioned twenty-four during my summer 2002 interview.

PAGE 83

71 The project was designed and managed by EMBRAPA, the Brazilian Agricultural Research Corporation, to work within the forested area of each familys lot. According to the Brazilian Forest Code each landowner in the Amazon must leave an area of 80% under forest cover. This limitation represented a problem for the small settler because it significantly reduced the land they could use for agriculture. The timber projects objective was to develop sustainable logging as an economic alternative for these forest reserve areas. The strategy was to identify appropriate low impact practices, and technologies that could be adopted by the farmers without requiring large initial investments. The project did not aim at transforming the farmer into a timber-manager, since the potential timber to be exploited from the small 40 ha family lots was very small. The idea was to complement the families income with an activity that could be conducted with available human resources. This project was also designed with a forestry research focus, with a control area and a well-established system for monitoring activities and ecological responses in the treatment plots (initially ten). Financial studies were conducted, which estimated the average income increase for the participant families at 700 Reais per year (about $ 250). As part of the project, EMBRAPA provided training for the participants in tree inventories, cutting and transporting timber, and equipment maintenance. An association of participant families was created in order to transfer some of the equipment (chainsaws, truck, etc) and to facilitate the creation of a platform for marketing products.

PAGE 84

72 After seven years the project had increased the number of participant families to 24 and had trained several teams to conduct the full logging operation from inventory to transportation. Some families had incorporated timber production into their activity calendar and relied on this activity for a significant part of their income. The association (APRUMA) had achieved a degree of autonomy and management capacity. However, the project was also challenged by some problems. On the economic side, the small scale of the project threatened its sustainability. The project provided all the necessary monetary pre-investment for the farmers, including the equipment and the legal authorization, yet as EMBRAPA prepared to withdraw, it was uncertain who would cover these costs. Timber sales also represented a problematic dimension of this project. To begin with, the quality of the timber available was very low due, among other reasons, to pre-project illegal logging exploitation. Timber industries were interested in larger volumes that compensated for the high transportation costs; therefore individual families offering 40m 3 of timber per harvest had little chance of getting a significant price. Although the APRUMA represented a platform to sell the harvest jointly, the cost of organizing and maintaining an association was very high in terms of time and organizational skills. The project required a professional forester to oversee the operation. EMBRAPA had provided these services so far. How would the costs of a forester be covered? Certifying the timber production through the FSC was sought as a means to guarantee a higher

PAGE 85

73 price; however, who could pay for the costs of maintaining this certification was not clear. Empresa Brasiliera de Pesquisa Agrcola The Brazilian National Agricultural Research Agency was created in 1973 to provide scientific support for extensive agriculture and cattle ranching. Forest management became part of EMPRAPA Acres agenda only a few years before this study. The Acre research station was one of thirty-two distributed around the country. This station focused on providing economic development for the ecosystem and the agro-industrial complex in the state. Sustainable forest management and silviculture were specific areas of action. Under this mandate, the Peixoto project was a prominent initiative to promote low impact, sustainable forest management in family agricultural areas. As a typical EMBRAPA research project, its financial support came both from the agency and from international organizations (ITTO 17 ). The project had a principal investigator who had led the research since the very beginning. Project leadership appeared to be concentrated in this person as he made most administrative and strategic decisions for the project. He was responsible for the results of the project as well as for the administration of the funds and the personnel. He was also the contact person with the community, the person who initially proposed the idea and under whose leadership the farmers had agreed to participate. 17 International Tropical Timber Association

PAGE 86

74 As with all projects dealing with forest management, it had both a technical-ecological dimension that looked at environmental impacts and technological approaches used, as well as an economic dimension that related to the viability of the project. Research results had to integrate these two aspects. However, the successes of this project were more in the technical and ecological than in the economic aspects. Beyond the simple economic viability of the project, staff recognized the need for further training of the participants in issues dealing with the business aspects of timber. Also, the imminent withdrawal of EMPRAPA from the project would require APRUMA, the community association, to be able to take over the leadership of the project, raising its organizational capacity. So far this association had passively submitted to the vertical management style of the research project, showing very little self-management capacity. This was recognized by the research team at EMPRAPA as a serious limitation for the future of the project. Association of Rural Producers of Peixoto The association of rural producers of the Peixoto agrarian development project (APRUMA) was created in 1998 to provide a platform to group the participants in the timber project. EMBRAPA was able to transfer some of the project equipment to APRUMA, especially the larger timber processing plant and the truck, which were meant to help the farmers add value to the products they were offering. This association would also help organize the collective work that the timber operation required, since it had to move from one participants plot to the next. The association started with eight founders and, as noted above, had twenty-four active members by mid-2002.

PAGE 87

75 The association required that 2% of the income generated by each participant from the sale of timber go to the group. The group met every two months according to the project schedule set by EMBRAPA. A truck from EMBRAPA picked everybody up and later brought them back after the meeting. The group had a board of directors comprising 14 members, with a president, a secretary, a treasurer, two vice-presidents and 8 overseers (fiscales). The group had all its legal documents updated and kept a well-organized set of books regarding activities and financial matters. The group had had two presidents since their creation. Recently a crisis had challenged the organization as one member was expelled after failing to contribute the 2% from the timber sale, that corresponded to the association. Despite their very positive relationship with EMBRAPA, the transfer of the projects equipment was seen as a difficult process. The maintenance of this machinery would demand more collective responsibility than the group had developed so far. Cachoeira Project The Cachoeira community forestry project was located in the municipality of Epitaciolndia in the upper Acre river valley, on the border with Bolivia. It was based in the Chico Mendes agro-extractivist settlement also known as PAE Cachoeira. This PAE covered an area of 24,898ha. It had 68 families that lived mainly on subsistence agriculture and the sale of Brazil nut and rubber. The closest city was Xapuri, the center of the rubber tapper social movement during the eighties. The area had a long history of political engagement and collective action. Many of the families that lived in Cachoeira had participated in the

PAGE 88

76 empates 18 to stop the forest from being cut by the ranchers. As a result of that struggle, the PAE was created by INCRA through the acquistion of the area and it was reassigned as a place for rubber tapper families. The timber project in Cachoeira originated as part of the initiative to establish a wood products industrial park in Xapuri. The Polo Moveleiro, as it was known, was a joint endeavor of the Catholic Church and the Municipality of Xapuri with strong support from the areas state congressman. 19 The idea was to establish a furniture-making school and open a manufacturing site in the city for wood-related companies. Italian groups linked to the Catholic Church directly supported the furniture-making school. The timber project in Cachoeira was to provide the raw materials for these industries. The mayor of Xapuri and the local state congressman had strong political and personal connections to the community of Cachoeira. These key players facilitated the introduction of the idea, although its approval by the community was a gradual process. CTA was initially brought in to start the project by establishing the initial forest inventories, but they were later replaced by staff from the recently created SEFE (State Forestry and Extractivism Secretary) 20 The timber project was directly managed by the Associao de Produtores e Moradores do PAE Chico Mendes (The Association of Producers and 18 During the eighties, empates were peaceful demonstrations organized by the rubber tappers who physically put themselves between the forest, and the forest clearing machinery and workers brought in by cattle ranchers to clear the forest for pastures. 19 Ronaldo Polanco, from the Partido dos Trabalhiadores (PT). 20 In 1999 the PT won the state government election. Their platform, as discussed later in this chapter, was to center Acres economy in the valuing of forest products such as rubber, Brazil nut and timber. As part of this effort SEFE was created to conduct that policy.

PAGE 89

77 Neighbors of the Chico Mendes PAE). A total of nine families from the association participated. The project took place in each familys colocao (traditional family subdivision within the PAE). The project was based on the idea of Jardinagem Florestal developed by Virgilio Viana, a professor from the University of So Paulo (and later Secretary for the Environment for the state of Amazonas) who initially acted as the scientific supervisor for this timber project. Jardinagem Florestal is a low impact logging strategy that used a simple language to facilitate its adoption by community members. For example, forest reproduction was presented by using the metaphor of mae, filha e neta: mother plant, daughter plant and granddaughter plant. The project used animal traction to transport the logs within the forest. A truck provided by the municipality of Xapuri later picked these up. Most of the wood was sold to the Polo Moveleiro in particular to the AVER company, a designer furniture company from So Paulo. In contrast to the other timber projects in Acre, Cachoeira benefited from having a secure buyer for their product. The income from the wood went directly to the family from whose land the timber was harvested, although 10% of the income went to the association. This 10% was divided in two: half for education and health care, and half for the project itself. In 2002 SEFE left Cachoeira, and the State Secretary of Extension (SEATER) took over support of the projects forester. Throughout its development the project had counted on several sources of funding, routed through the State Government, among them ITTO, IDB and the PPG7 fund.

PAGE 90

78 The timber initiative in Cachoeira confronted several challenges in the four years between 2000 and 2004. One of these challenges was the enforcement of the management rules established jointly by SEFE and the community participants. Internal control systems were weak and monitoring was not followed up. Another challenge was the demand from other community members to participate in the timber endeavor. The process of including new families required thinking about introducing new buyers to take up the increase in production. This timber project saw its technical staff change every two years as a result of conflicts among the staff, the community and the scientific coordinator. Changing the staff created a lag in the operation, which affected production. Formally, hiring new staff or bringing in new institutions solved these conflicts. Associao de Moradores e Produtores do PAE Chico Mendes The association of residents and producers of PAE Chico Mendes was founded in 1995 with the support of the local state congressman to help attract projects to the PAE. Fifty families were associated with the group. The association represented the community at the municipal and state level and was in charge of several other projects besides the timber initiative. According to Stone (2003) the group had strong support from the community. The group had what are known as delegados sindicais (union delegates) that represented the association in the furthest corners of the PAE. They acted as liaison officers of the association and at the same time solved local conflicts and reported on residents needs.

PAGE 91

79 The group had a long history of relationships with external organizations, given the active role the community had during the 1980s social movements. At the time of the research the association had strong ties to the Xapuri municipal government and the State government of Acre both from the PT workers party. The Association was also a part of the Extractivist Cooperative, CAEX, based in Xapuri and the National Rubber Tapper Council (CNS). As noted earlier, nine members of the association participated in the timber project. This group had a lot more interaction than the rest of the members. They participated in the projects decision making as well as in training workshops and community forestry meetings at the state level. After SEFE left in 2002 the Association took control over the complete timber operation from acquiring the permits to selling the wood. Progressively they used the income to cover the cost of equipment maintenance and other associated costs like transportation and technical assistance. Divisions had started to appear within the community, especially between those living close to the center of the community and those far away. This division also corresponded to kinship differences. In 2002 a group of association members, together with other non-members living in the far south corner of the PAE, began forming another association. One of their proposals was to start a new timber project. Cachoeira Projects External Partners A number of organizations had played the role of external partners in Cachoeira. Among them were: CTA, the NGO involved in the Porto Dias Project; SEFE, the State Secretary of Forestry and Extractivism and SEATER,

PAGE 92

80 the State Secretary of Rural Extension. The role of the latter two was limited when compared to the way EMBRAPA or CTA had participated in their respective projects. The strength of the local association, and the particular origin of the projects link to the Polo Moveleiro project, were identified as the explanation for the lesser role of the external partners (Stone, 2003). Certainly the history of collective action and political activism placed Cachoeira on a different level than other rubber-tapper communities. The social and political capital were strong enough to suggest that their relationship with external organizations in a development project would be more balanced. Their capacity to organize themselves and state their demands required a more horizontal approach from the external agents. This was recognized by SEFE staff who were involved in the project between 2000 and 2002. However, the origin of the timber project in this case was just as foreign to the community as in the other cases studied. An external agent came in and proposed the idea, which was accepted by the rubber-tappers after a period of doubt and debate. Just as in the other cases, the project was limited to a small number of association members, most of whom sat on the board of directors and shared other connections among themselves (e.g. kinship and friendship). In Cachoeira the project was proposed by two political figures from outside the community: the mayor of Xapuri and the local state congressman. A third proponent of the project was Virgilio Viana, the university professor and later Environment Secretary of the state of Amazonas who served as scientific supervisor of the project. Although these three people did not constitute an

PAGE 93

81 organization in the way EMBRAPA or CTA did, they played a similar role to these organizations but in a personal capacity. These figures were sought after whenever the project had any type of crisis. The scientific supervisor frequently visited the project and played a role in the few project crises regarding technical assistance. Cachoeiras timber project involved a very small group of people from the community. This minority not only benefited from this project, they were also the same group that implemented most of the other community initiatives. There were friendship and kinship ties among them, too. The association had had only two presidents since its creation and the rest of the members of the board had rotated more than they had actually changed. Even within this small group of people, the actual leaders of the group were a few individuals. These individuals controlled access to information and also represented the community in relation to external agents. This was a similar situation to that of the other community timber projects. The difference in Cachoeira seemed to lie in the capacity and vision of this very small group of leaders, particularly to maintain decision-making control over what went on in the community. Formally, this group of individuals had progressively gained more control over the timber project, making the political role and power of the external founders of the project (the state congressman and particularly the scientific supervisor) very difficult to assess. Yet the conflicts discussed above showed evidence of the strong role they still played. It was hard to determine the intensity or extent of that influence, because the nature of the relationship

PAGE 94

82 between the external founders and the community leaders was more personal than professional. Most of the exchange that went on was hard to follow. It was uncertain whether the community leaders had the same capacity to challenge these external players as they showed with SEFE or CTA. Cautrio Project The Cautrio community timber project was a joint venture of the NGO Ecopor; the Organization of Rubber-Tappers from Rondnia, OSR; and the Association of Rubber Tappers of the Valley of Guapore, AGUAPE. The project was located in the Rio Cautrio Extractive Reserv. This was a state sanctioned extractive reserve as opposed to the federal reserves like the ones in the state of Acre. The Cautrio RESEX was located in the municipality of Costa Marquez on the border with Bolivia on both sides of the Cautrio river. The RESEX had a total area of 146,400ha and a population of 210 individuals. It was created in 1995 by the state government of Rondnia as a result of the PLANAFLORO project. This was a statewide initiative supported by the World Bank to re-orient Rondnias development to a more sustainable path. PLANAFLORO was implemented after the PLANOROESTE initiative (also from the World Bank), whose environmental impact originated an international movement against World Bank-supported development initiatives. The timber project started in 1995, harvesting a total area of 964 ha, with a 30-year cutting cycle. It was established as a selected logging project using mechanical transportation within the forest and building a sawmill facility that was run by the AGUAPE association. The project was supported originally by the World Wildlife Fund (WWF) and later by the PDA-PPG7 fund, administered by

PAGE 95

83 the Brazilian Government. The projects main objective was to develop timber management as an economic alternative. The project also represented a strategy to defend the RESEX from the continuous invasions by illegal loggers and land speculators. Since 1995, AGUAPE had sued over 10 invasions to the extractive reserve. These invasions were made more complicated by the lack of action of the state government to defend the areas. In some cases the state government went as far as cutting deals with the squatters, allowing their presence in the RESEX under the condition that they stop clearing land. The project provided some limited capacity to patrol the areas borders and react early to potential incursions. After seven years the project had achieved a strong level of community participation in all field operations; however, community control over the sawmill had confronted various problems. The most significant were the management and marketing difficulties, given the lack of business skills of the participants. Other issues had to do with the long time required for legal permits, and the inadequacies of the machinery scale (too small for the type of job). Most of the timber offered in Rondnia came from legal deforestation (to clear for agriculture or pasture) or illegal logging; this kept prices low, affecting the viability of the project. For this reason, the project had focused on completing a business plan and had hired new personnel for the marketing of the products. A new road into the reserve would greatly benefit the project, but contrary to the situation in Acre, the government in Rondnia had little interest in sustainable development or community forestry. Through the support of WWF, ECOPORE

PAGE 96

84 was pursuing FSC certification of the enterprise, hoping this would bring a plus to the timber selling prices. Ao Ecolgica Guapor ECOPORE was a non-governmental organization with thirteen years of experience working for the sustainable development of the extractivist communities in Rondnia. Along with the Organization of Rubber-tappers from Rondnia (OSR), they participated in the early movement to stop the PLANONOROESTE development plan, and pressured the government for the creation of State RESEXes like Cautrio. In this partnership OSR did most of the political work while ECOPORE provided the scientific information and the technical assistance. The relationship continued in the formulation of the forest management project for the Cautrio RESEX. The strong articulation among the groups generated high levels of trust among the different organizations leaders and representatives. Originally WWF-Brazil was one of these partners, but progressively it withdrew to a more indirect role. ECOPORE was a small technical organization led by an agronomist, who was the team coordinator. Along with this coordinator three other professionals worked in the Cautrio project: a forester, an administrator and a marketing specialist. Associao dos Seringueiros do Vale do Guapor The AGUAPE association was founded in 1992 as the rubber tappers from the Guapor river valley, on the west side of the state of Rondnia, formally organized the group that would later cover three different extractive reserves:

PAGE 97

85 Cautrio, Courrralinho and Pedras Negras. The main objective of the group was to defend the socioeconomic interest of its members through mutual assistance and the development of activities to strengthen the quality of life of the communities. The association had three bodies: a general assembly, a coordination team and a directory board. The Board was the governing body that was elected by the general assembly every two years. The coordination team was composed of representatives of the three extractive reserves and oversaw the equitable development of activities. AGUAPE had over 365 members, and partnerships with a series of NGOs and government agencies in different projects, from schools and adult education centers to health and ecotourism facilities. One of these projects was the Cautrio timber project, in which the association participated directly in all field operations and also ran the sawmill. Through this project they were able to open twenty-nine kilometers of roads within the Cautrio reserve; however, the project required much more road investment to secure its financial sustainability. AGUAPE was continually challenged by the technical complexity of the timber project. Rubber-tappers were not familiar with the activity, and there was a serious cultural gap between their traditional lifestyle and the running of the project. Low levels of formal education among the rubber-tappers were also a contributing element. Nevertheless, steady progress had been shown as community control had increased throughout the existence of the enterprise. The community endured despite the intense involvement required by the project, even though very little profit had been generated. However, the rules

PAGE 98

86 collectively established to manage profits defined that 62.5% would stay in the hands of the association for re-investment (37.5%) and association activities (25%). The other 25% gross profit was distributed among the participants working in the project, and 12.5% corresponded to the OSR. This arrangement was very different from that of the other community timber projects in this study. In the other three, most of the profits went to the rubber-tappers in whose area the logging was taking place, and a small proportion went to the association. In this case, not only did most of the profits go to the association, but some of it was targeted to re-investment in the project. Comparative Analysis The purpose of this chapter is to identify the internal factors affecting the collaboration experience both at the organizational and the project level, as I expected this organizational experience to inform the effectiveness and significance of the GPF as a collaborative platform. So far I have introduced and described the four projects and their respective organizations involved in the GPF initiative. From this pool of stakeholders I made a deeper analysis of the variables more closely related to collaboration (Table 3.1) I took these variables from the literature on conservation and community forestry coalitions and used them to elicit the common collaboration pattern. I defined this pattern as the implicit template that modeled the style and effectiveness of collaboration in the GPF. In each section I start by describing the results from my data collection. I aimed at assessing the condition of each variable across the projects, to identify the common elements. I also mention

PAGE 99

87 some of the relevant differences among the projects, emphasizing the ones that might pose a future threat (or an opportunity) for the GPF process. Table 3.1: Comparative Analysis Variables Level Variable Participation Leadership Organizational: Community Assoc (CA) Partner Organizations (PO) Management Capacity Partners Relationship Projects training Focus Projects: (Usually represent a collaboration between a CA and a PO) Appropriation of the Project by the Community Participation The four community timber projects analyzed in this study were designed originally as community participatory projects. In practice these labels were hard to assess, as the projects were neither completely open to every member of the community nor did they explicitly exclude intra-community groups. The projects did take place in a community setting, and their direct beneficiaries were poor rural households living in communities. All of them were designed as pilot projects with the aim of developing a production model that could later be extended within the community and to similar communities. In the four to six years since these projects began, this scaling-up had not yet occurred on a significant scale. In fact, a few participants from each project had abandoned it. The projects ranged from 10 to 25 participants 21 Three of the four projects used the individual areas of the participants as the area for the timber project, establishing a direct relationship between what was cut from an area and what 21 Project staff referred to the number of families involved, but the direct participants in the projects (referred to as manejadores) were mostly the male heads of household.

PAGE 100

88 the owner of that specific area received. The experimental nature of these timber projects explained the small number of participants. The expectation remained that once consolidated they would include the rest of their communities. Participation in a project context also refers to the involvement of the beneficiaries in the decision-making process. In the cases analyzed this kind of participation was an explicit goal. The timber projects aimed at building capacity in the community to run the timber operation and manage the resulting enterprise. The goal of transferring the project to the community required extended participation by the locals, and progressive decision-making autonomy. I discuss this issue in a later section on project appropriation. Association members were typically the male heads of household, with little participation from women, elders or youth. Associations were created as a result of an external institutional intervention: either the creation of the settlement or the timber projects. In periods that varied from four to eight years, the boards of directors had experienced little change in membership. Board members changed positions with each election (every two years) but very few people joined as new members. Typically only two (three in one case) individuals had become association presidents in each association. Presidents tended to return after one period or less than that if the group experienced a crisis. Kinship and friendship tended to play a role in who got elected to the boards and, incidentally, participation in development projects (like the timber management) tended to include those board members first and then other affiliates.

PAGE 101

89 In general the participation of the community residents in the local association was scarce. Complaints from the local leaders about the lack of member participation in the associations were constant. They found it hard to keep the group motivated and engaged in the activities. In their vision, people only came to meetings when there was an immediate benefit they would get out of it. The concentrated attention given by the leaders to the timber project seemed to have limited the time they dedicated to other activities that involved the whole community (or the whole association). A wider participation in the associations seemed to be dependent upon the potential success and expansion of the timber activity. The paradoxical situation was that the limited participation in the association activities and duties (such as board positions) constrained the potential expansion of the timber projects. The associations needed to take control over the projects and for this they needed to raise their management capacity, which was limited by the scant participation. PO staff interviewed trusted that by providing training opportunities in the business dimension of the timber activity, other community members would be motivated to participate. However, these opportunities required a level of education higher than the average in the communities. In summary, participation was stagnant, and strategies to increase it depended upon the success of the timber projects, which were at the same time limited by the scarce participation. Leadership Through direct observation and interviews with both community participants and PO staff, I tried to identify the principal leadership traits among the projects

PAGE 102

90 studied. I also tried to track the influence that certain PO project leaders had on the implementation and results of the projects. Leadership in all four community timber projects came from outside the community. The individuals with the vision and the charisma to advance the idea of timber extraction were almost all NGO staff, either forest engineers or agronomists. In two cases the initiators were also regional leaders linked to the rubber tapper social movement, while in the other two they were forest engineers working for the partner organizations (one NGO, one government agency). Only one of these individuals lost complete contact with the project (he was substituted by someone who played the same role); the rest remained as bearers of the vision that sparked the projects. In three of the cases these individuals had administrative control over the project, and they had also become the reference name and face of their projects. As mentioned in the first chapter, timber extraction was highly contested by political and social leaders in Acre, especially among rubber tappers. The proponents of timber extraction projects confronted a difficult environment that only began to change by 2002, when this study was conducted. This fact was mentioned as an explanation of the strong leadership style that the initial project coordinators showed. According to PO staff that came later to the projects, as well as to some of the external stakeholders interviewed 22 this leadership style reinforced an existing paternalism in the relationship between the communities 22 A total of ten interviews were conducted among NGO and government (state and federal) personnel. These were individuals familiarized with the timber projects in Acre and Rondnia, but that were not were directly involved in such projects.

PAGE 103

91 and external agents like the NGOs and government agencies involved in timber projects. Paternalism was described as a negative trait in which communities (rural peoples in general) behaved passively and with a sense of obedience towards the powerful external individuals that interacted with them, by virtue of an economic relationship or a development project. According to the interviews, paternalism was a trait present in other areas of Brazilian culture, especially with regard to the relationship between the government, its agencies and citizens. The interviewees mentioned the establishment of strong personal relations between the PO project coordinators and community leaders, as examples of paternalism. Through these personal ties, the community members granted an extended amount of power to the project coordinators. This power went beyond the project issues, turning the coordinators into political advisors to the associations and even authoritative dispute mediators. The development needs of the communities involved in the timber projects extended past the projects reach. Therefore, having an advisor and a respected mediator were direct added (and appreciated) benefits obtained by the community. The communities accepted paternalistic leadership as it provided this particular services and a general sense of security and protection. Later efforts by new PO staff to practice a different type of leadership were not always received positively. Efforts to leave more decisions in CAs hands and to withdraw the project coordinators from their roles as mediators or advisors resulted in confusion, uncertainty and sometimes conflicts. Although leadership

PAGE 104

92 could be generally described as paternalistic, an awareness of the problems associated with this trend was growing among the later project coordinators and the community leaders. Management Capacity Another aspect to understand the characteristics of the interaction between CAs and POs in the timber projects was provided by their previous management experience. Prior to (or beyond) the timber project, I wanted to get a sense of what these organizations (CAs and POs) were able to accomplish. By raising this issue I would be able to corroborate trends in other variables. For example, the limited participation in the timber projects and the paternalistic leadership style presented in them was correlated with the meager management experience from the CAs. It was clear that the complexity of the timber activity made it completely foreign to these groups. As mentioned earlier, the projects had the intention of preparing the community associations to autonomously manage the timber projects. The intention was to run the projects as business enterprises that could secure a stable income for their families. Although the projects varied in scale, and each scale demanded different degrees of collective action, all initiatives were challenged by the restricted management capacity of the associations. The experience of the rubber tapper social movement was not evenly distributed among the associations. Only two of the associations had leaders that were involved in the struggle to create the extractive reserves and settlements. One of the groups was made up of migrant farmers from outside the region. Even among those with collective action experience, administering an association and

PAGE 105

93 its activities proved to be very different from the political mobilization that they had experienced in the eighties and early nineties. Association work started with the basic bookkeeping that helped to keep their legal status current. This work included keeping records of their meetings, periodically electing their board officials and presenting this to the state registrar to update their CNPJ their associations identification document. All the associations claimed to have had their legal documentation updated, and to be conducting all the legally required activities such as keeping minutes of meetings and keeping accounts of the project activities they were involved in. This documentation was a pre-requisite to receiving government funding such as the PDA-PPG7 funds that had supported all the timber projects. The associations studied all had timber management as their main activity. Some of them had other activities or community projects, but they were of a smaller scale than the timber endeavor. Among the activities run by associations were projects on ecotourism, wildlife reproduction, handcrafts and health. Most of them were joint efforts with other NGO or government agencies. Only one of these associations managed a local store that sold locally produced rubber and, in exchange, brought goods from the city for local consumption. Two other groups had experimented with similar initiatives to commercialize rubber and Brazil nut production but they abandoned them. Price fluctuations, along with management problems, were cited as the main reasons. Problems dealt with post-harvest storage of the products and lack of maintaining an adequate operational capital to last through the production cycle.

PAGE 106

94 The timber project proposals and early implementation had a higher expectation of the CAs management capacity than they were able to deliver. This gap was cited by the partner organizations staff as the most difficult problem faced by the timber projects. This multifaceted issue began with the difference between the typical daily (and weekly) labor cycle of the rubber tappers and the intense industrial labor requirement of the timber activity. Although all project designs tried to adapt the timber production to the cultural practices of the rubber-tappers, this was gradually abandoned. The objective of achieving economic viability was, in practice, hardly compatible with the cultural sensitivity expressed in the project proposals. Only the small-scale projects attained a better balance, although their viability remained in question, particularly if projects were to move to sell production together through the GPF coalition. Quality standards and production deadlines would change the small projects production cycles. Another problem confronted by the associations in their timber undertakings was with regard to training and labor specialization. Three of the four projects were designed with the goal of incorporating timber management into the production cycle of the rubber-tappers (and farmers). As a result, project participants were trained in all aspects of the production, hoping that they would be able to do most of it by themselves. This everybody does everything strategy was progressively abandoned by the projects (including the small-scale) in favor of internal specialization. Certain individuals in each project started to take care of specific activities: inventory, felling, transporting. Specialization had

PAGE 107

95 its own complications, as was experienced in one of the projects that managed a sawmill: they trained two individuals to operate the main saw and shortly thereafter both left. One went back to farm his plot and collect Brazil nuts and the other, with the new skill, found a well-paid job in a timber company. This situation stopped production halfway into the season. To solve the problem they hired outside operators. The most serious management problem for the partner organizations was the high degree of personnel turnover. The support organizations had teams made up of one to three forest engineers with 5 to 10 years of experience. Changes in staff, particularly project coordination, significantly disrupted project development because decision-making and information were concentrated in these coordinators. In some cases changes had a positive effect. Community participants considered the current staff teams to be more open to their concerns than the teams had been at the beginning of the project. The management capacity of the CAs did not match what was expected of them in the timber projects, nor was their learning as expedited as the PO expected. The change in Porto Dias to outsource certain portions of the operation was an example of that. Outsourcing solved the operative problem, by bringing in somebody with expertise. This strategy did not address the underlying issue: CA limited management capacity. With the current capacity, there were no signs that the CAs could achieve fair and efficient agreements with the outsourcers in the absence of the POs.

PAGE 108

96 Project Relationships To understand any type of natural resource collaboration effort, the relationship between the main parties involved needs to be considered. In my framework I have understood the study of the relationships as particularly referring to the inter-personal interaction between CA members and PO staff with regards to the timber projects. As I explored matters of efficiency and power in other variables (such as leadership and management capacity) I devote this section to perceptions from both sides (CAs and POs) of the quality of their mutual relationship: whether they found it was adequate and how it had changed through time. In my interviews I directly asked about conflicts but obtained very little information. Although other questions as well as my participant observation confirmed the positive interactions between most POs and CAs, it seemed to me that asking directly about conflicts was not a good approach to elicit information on the subject. The perception of the projects by the communities was positive. Direct tangible benefits such as the opening of roads or protection of the area from invasions and illegal loggers were cited as examples of the projects successes. These benefits reached beyond the direct project participants and created a constructive environment for the partner organizations involved in the timber projects, as well as general members of the communities participating in the projects. In general the presence of the partner organizations in the communities appeared to have filled the vacuum created by the lack of government attention and support to these agrarian settlements and extractive reserves.

PAGE 109

97 The relationships established through the projects between the staff and the community participants were also encouraging. However, the unforeseen operational costs that had eroded profits from the initial timber sales had, in some of the projects, shaken this positive perception. The preoccupation was more with the long-range success of the projects than with the decisions made or the financial management, because the level of trust between community associations and partner organizations was very good. The projects had generated a strong web of relationships between community participants and staff. Personal friendships, as well as a common set of values had grown from the community of practice that the timber projects had generated. Only one project had experienced a direct confrontation between staff and community leaders (as detailed earlier in this chapter). An important element of the relation between the community and the support organization was that the relationship was established between a very small number of individuals. Two or three staff engineers interacted mostly with two or three community leaders. This was very similar in all the projects. Interviewees from both the partner and the community groups mentioned other problems regarding intra-community relationships. First was the social and financial cost of having local leadership positions. Association leaders complained that they had to sacrifice a lot of their own personal work time and personal life for the group. The group members in general did not pursue leadership positions. This lack of interest on the part of others obliged the leaders to maintain their positions, despite the personal cost. The second issue

PAGE 110

98 mentioned was the gradual division in the community between those involved in the projects and those who were not. Because little to no profit had yet been generated through timber harvesting, this division did not cause conflict. The inclusion of new participants had become an issue in a couple of projects, and in another one a group of disenfranchised association members was about to start a new association. Beyond the associations meetings there was a lack of formal conflict management platforms. Project staff dealt with disputes among community participants on an informal basis sometimes in the context of these association meetings, other times simply in the development of their project work. Playing this mediator role was an expression of the authoritative position that the project staff had within the community. This had begun to be recognized as something that needed to change. The development of local conflict management institutions was seen as an opportunity to strengthen the communitys control over the project. Although in general relationships between parties in the projects were positive, this only meant that they were not conflictive. They were, however, characterized by a strong paternalism, with the PO staff holding the powerful position and the CAs playing a passive role to secure immediate benefits. Benefits were sometimes restricted to the very small group of project participants, in some cases generating intra-community divisions and even conflicts. A non-conflictive relationship sometime resulted in a transfer of

PAGE 111

99 conflicts. Intra-community conflicts could increase between those more benefited by the PO, and those less benefited. Projects Training Focus Given the significant gap between the communities traditional activities and logging, training was one of the basic instruments to build capacity in the CAs. Approaches to training that were more explicit about social learning tended (by definition) to foster experimentation and encourage effective collaboration between PO and CA. The critical use of transferred skills was clearer from these conscious social learning approaches than from the traditional ones. I included in my interviews to both CA and PO members a set of questions on the training provided by the projects. I also had the chance to interview two external trainers brought in by two of the projects. The two main challenges confronted by these four community timber management initiatives were: first, the high initial investment needed to start up the projects (basically for machinery); and second, the broad gap between the traditional rubber-tapper culture and logging. Project proposals emphasized these aspects by budgeting resources for equipment (chainsaws, tractors, trucks) and planning substantial time for workshops, training courses and cross-site visits. According to the interviewees, capacity development was a very important component because their social sustainability depended on its success. It had to start with training on basic timber operation and only through a cautious learning process could it advance to include long-term logging planning, for example. In the meantime, those decisions would remain in the POs hands.

PAGE 112

100 All projects, regardless of their scale, included training activities for the main aspects of the forestry field operation based on their technical model. All the projects technical models were based on Reduced Impact Logging (RIL). Cachoeira had Vianas Jardinagem Florestal, while Porto Dias and Cautrio both adapted the model from The Tropical Forest Foundation (Fundao Floresta Tropical). This group from Eastern Amazonia had developed a sustainable tropical forestry model. Peixoto, given its very small scale, had its own version of RIL, supported by the research of its project coordinator. Training ranged from forest inventories to felling, and to tractor and animal transportation of logs in the forest. Workshops and training activities for these tasks were always accompanied by practical sessions in the field, and learning was put to use soon after, as the dry season was the time both for training and for harvesting. Training was mostly conducted in the field where the activities were being carried out. For this reason it could be described as practical training. All projects relied on preparing local teams as the working units for carrying out the logging. These teams were coordinated by the projects staff, who in many cases acted as tutors or on-site trainers. In cases where certain local participants showed more interest, training was individualized, to the point where the project staff and his or her local apprentice became a working team. This was particularly true for the Cachoeira case. Among these sets of activities, equipment maintenance was mentioned as the most difficult aspect one for which external assistance was permanently

PAGE 113

101 needed. However, the overall success of capacity building of these field aspects of the timber project could be noted in the later development of cross-site visits, where locals were sharing their knowledge and experience on these subjects. The success in these areas could be contrasted with the mediocre level of capacity development in other areas necessary for the timber management, such as business skills and organizational development. The first element to contrast is that, while the forestry aspects of these initiatives were well covered by the project staff, all other aspects lacked adequate personnel. Up until the summer of 2002, almost all available field personnel for the projects were foresters or forestry technicians. Among the group of foresters involved in the projects, several events helped to raise the interest in collaborative learning between the projects. These events started in 1999 and later evolved into the GPF proposal, that recognized the valuable role of each projects experience and the importance of fostering learning not only across the PO staff but also among the CA participants. I highlight some of the main events leading to social learning in the following section. The change towards a social learning approach Three different initiatives surrounding the timber projects (in Acre) started to promote a change towards a more interdisciplinary vision of the timber projects beyond their initial forestry bias. The first one was a working group on community forestry in the Amazon supported by WWF, IMAZON and IEB (as well as other organizations). This working group brought together practitioners

PAGE 114

102 from the different community timber projects in the Amazon to discuss their issues and share their experiences. They organized annual meetings starting in 1999, where practitioners began to reflect on the non-forestry challenges of the timber projects (Capossoli, 1999; Macedo, 2000). Things like community organization and marketing skills were brought up as new demands after three or four years of project development. Another initiative, with an orientation that complemented the forestry work of the timber projects, was the Adaptive Collaborative Management (ACM) research project, conducted in Acre by CIFOR, UF, PESACRE and CTA. The ACM project focused its work more in Porto Dias, but nevertheless developed a model adaptable to the other projects. The ACM looked at participation within the project as well as on the social learning process that went on through several timber harvest seasons. The project generated reflection for both staff and local participants, helping them understand the group dynamics and non-linear nature of the learning process. Related to this project was an exchange visit organized between Acre and the Mexican ejidos of Quintana Roo. This exchange started with a trip to the ejidos by a group of Acreanos that included both local timber project participants and a number of municipal and state political authorities, including the State Secretary of Forest and Extractivism. This trip and the visit to Acre by a similar Mexican group, was cited throughout the interviews as the single most significant learning experience for the manejadores. It provided a benchmark for looking at the future of the community projects in Acre. It helped clarify the business

PAGE 115

103 dimension of the timber projects, and the key role that government policies could play to support community timber initiatives. People cited the twenty years of support received by the government in Mexico, before the local enterprises could start to stand on their own. Another very important thing cited about this trip was the idea of collaboration among timber projects and the notion of a centralized organization that provided technical as well as political support to the communities. The example of the Sociedad de Ejidos served later to articulate the idea of the GPF coalition. Capacity building also took place beyond the organized workshops and training courses. Sometimes workshops were too isolated to have a meaningful effect, especially if not linked to an immediate set of practices (like in the case of field tasks). In general, evaluation of the impact of these workshops was not documented beyond recording the number of participants or the number of similar workshops. The ACM project represented a good model of what needed to be done in the rest of the projects to make a more effective use of capacity development activities, particularly if the topic of training was far from the familiarity of the trainees, like the case of business skills such as marketing or negotiation. Training for the latter aspects, as well as for organizational development, seemed to happen through direct participation, more than through formal instruction. Direct involvement in discussions and confronting challenges provided the ultimate learning ground for these issues. An interesting aspect of this experiential-learning was the role played by project leaders (both staff and

PAGE 116

104 locals) as models of behavior for the rest of the group. Interviewees from the community groups as well as lower staff of the projects emphasized the importance of the coordinators personality in setting the environment for the project. When describing what they had learned from these people in leading roles, participants focused more on attitudes in addressing certain situations, than on actual skills or techniques. Values and sense of priorities were also described as acquired through proximity with these leaders. Interviewees also cited the chance to understand or grasp difficult things, such as the difficulties of fundraising or the complications of measuring the sustainability of timber management. Although the local participants did not master these issues, the close relationship with the project coordinators gave them the chance to begin to understand them and to comprehend their importance for other actors. The next subsection provides an example of the learning process from one of the projects (Porto Dias). This brief description focuses on the change in a technical feature of the project; nevertheless it helps to show a more general trend also presented in the other projects. Experimentation in Porto Dias: The Porto Dias project had the individual family colocaes 23 as management units. An extraction plan was developed for each of these areas. Not all the colocaes were exploited in the same year. Project participants had to agree to work on other members areas without receiving anything but the days pay. When the project was designed, the fact 23 Colocaes or colocao (singular) are the traditional rubber-tapping land tenure unit. They are subdivided into rubber-tapping trails (estradas de seringa). In the current RESEX and PAE system, families dont own their colocaes but have exclusive rights to use its resources (Stone, 2003)

PAGE 117

105 that it was adapting itself to this existing internal land tenure division was emphasized in the proposal as a technical adaptation to the cultural context. However, CTA technical staff from the project later perceived this adaptation as one of the projects mistakes. The dependency on CTAs technical support to run the timber operation made it impossible for the manejadores to work on their respective colocaes as individual management units. They progressively aggregated areas and conducted the timber activities in groups of colocaes, as detailed by Stone (2004). The use of the colocaes had reduced the viability of the project. In practice, it made more sense to plan for the PAE as a whole and only subdivide the profits among the families. This reduced costs, time and bureaucratic procedures. Another lesson in which the initial intentions of the project were wrong was the issue of transportation of logs. Originally the project had relied on a mix of animal and mechanical means to get the logs out of the forest, but by the second harvest, animal transportation had been ruled out. Transportation was contracted from outside loggers in order to meet the volumes and the dates agreed with by buyers. Technical issues, such as the projects transportation logistics, were easily identified and changed by the project staff. Most of them were openly explained and discussed with local participants. Managerial aspects such as the layoff or hiring of new personnel were more troublesome to deal with and communicate. This limitation was also a reason for the tercerizao (hiring a third party for certain activities, mainly transportation) of some project activities.

PAGE 118

106 The problem was that the training for these skills, although timed to fit in with the production cycles, did not allow sufficient time for the community members to fully learn the processes. As a result, the community members were not ready to run the operation, which therefore continued to be heavily controlled by CTA. In addition, CTAs staff was composed mainly of foresters with little knowledge of the timber market and few marketing skills. The staff learned about selling wood over the first two harvests. The problem was that the project design did not anticipate this slow learning curve. This blind spot in the projects design had major costs for the projects implementation. In terms of budget and staff, the project was not prepared for the running of the timber activity as an enterprise or for enabling the community to do it. Six years into the project the operation was still under the control of the support staff. More than support, they were the bosses the ones with the vision and interest to run it. They were even extending themselves beyond their own set of skills (technical forestry aspects) into social organization and business skills, for which they lacked proper expertise. Some authors describe the experiential learning of values and working principles through the modeling behavior of the leaders as second loop learning (D'Arcy, 2001; Margoluis & Salafsky, 1998; Smutylo, 2001). This refers to the learning process that, although parallel to the process of learning technical skills (such as timber operations) (first loop), involves changes in values and world-view. Projects for the most part remained unaware of the second loop learning, although this had probably been occurring all along. Training up to 2002 had

PAGE 119

107 focused on the basic skills of logging, which prepared the CA participants to work in the projects but not to manage them. This limited reach of most of the training strategies becomes clear in the next section where I discuss the appropriation of the projects by the communities. Projects Appropriation by the Community Associations The degree to which each project was run by the community associations was of utmost importance to the POs. For them, the transfer of project tasks to the community was a significant indicator of project success. The appropriation of the projects by its respective community meant two things; first, a sign of the incorporation of timber into the household economy and second, that communities running timber enterprises would be capable of coming together to form an independent GPF coalition. Given this importance, I included several questions on this issue in my interviews. I also incorporated a series of cards detailing each step in the timber operation and asked each interviewee (both from POs and CAs) to describe whether the CA or the PO did each activity. The list of tasks was the result of conversations with the technical coordinators from the Porto Dias and the Pexioto projects. Answers were not clear-cut; most of them involved degrees of participation. For this reason, in analyzing the results I describe each activity as mainly executed by the CA or the PO. Each project, in its own right, pursued its appropriation by the communities from the very beginning. Strategies varied significantly between projects, just as did technology and scale. After four to six years of project development, the degree to which each project had achieved community appropriation was hard

PAGE 120

108 to assess and even harder to compare among the groups. The scale and conceptual approach of each project set a different standard for each project with regard to project appropriation. For this reason comparing appropriation among projects would be comparing very different things. This caveat aside, I consider the comparison to be a useful way to establish a baseline of knowledge capital for the GPF coalition 24 It would allow, among other things, the establishment of training priorities. To analyze the results, I summed all the results from the 30 interviews (16 CA, 14 PO) and classified them according to the main actor involved in that activity. Results were very consistent between CAs and POs, which allowed me to treat them as equal at the time of the analysis. Table 3.2 presents the results, by project, according to the task distribution at the time this research was conducted in July 2002. Figure 3.2 presents the same results but in the form of a bar chart, distributed by project and by the main group involved. Table 3.2: Task Distribution by Projects Activity Pt. Dias Cachoeira Peixoto Cautrio Management Plan PO PO PO PO Licensing PO CA-PO CA-PO PO Inventory CA-PO CA-PO CA CA-PO Tree Selection PO PO-CA CA-PO PO-CA Felling CA-PO CA-PO CA CA Transportation in the forest CA-PO, T(PO) CA CA CA-PO Licensing & Tax for Transport PO CA CA-PO PO Transport (sawmill, Costumer) T (PO-CA) T (CA) T, CA-PO CA-PO, T(PO) Sawmill processing T (PO-CA) T(CA) CA, T (CA) CA-PO Marketing PO CA CA PO-CA Negotiating & Contracting PO-CA CA CA PO-CA Administrating (income) PO, CA CA CA CA-PO (CA= Community Association, PO= Partner Organization, T= Contracted Party in parenthesis is who contracted; PO-CA= mainly partner, CA-PO mainly community) 24 This is similar to the way Stone (2003) analyzed the participation of the communities of Porto Dias and Cachoeira in the projects.

PAGE 121

109 2345678910 tasks performed m Partner tasks performed m Community tasks performed m Party Figure 3.2 Task Distribution by Project Table 3.2 and Figure 3.2 show that Cachoeira and Peixoto concentrated the largest number of activities mainly conducted by the community. Porto Dias and Cautrio had more PO involvement in the process. In these last two projects CAs were involved mainly in the field chores. The low level of formal education among manejadores (project participants) would explain the reduced participation in administrative chores in all projects 25 Also, the fact that Cautrio and Porto Dias were significantly larger than Cachoeira and Peixoto would suggest that the scale of the projects played a role in the speed at which communities could take over certain tasks. Although it might seem as if Cachoeira and Peixoto had stronger community organizations than Porto Dias and Cautrio, actually Cachoeira and Peixoto were organized to function at a household level, with production and timber sales done by 25 Among those interviewed (16) very few had completed primary education, and some could only sign their names.

PAGE 122

110 individuals. The higher number of activities in control of the community is more related to that individual focus of those two projects. Among all projects there was an expressed sense of accomplishment regarding tasks such as the inventorying of trees or felling and transporting logs through the forest. Mainly the community groups conducted this activity in all cases (Table 3.2). Nevertheless, the learning process required several harvest seasons (two or three). The difficulties in matching the learning process with the need to generate income from the timber became a significant challenge. Table 3.2 also illustrates how the challenge of completing the harvest on time, or of dealing with the unexpected failure of machinery, made it necessary to hire contractors to take care of some of the tasks. This was not foreseen in the original proposals of any of the four cases studied. The goal in the original designs had been to bypass the middleman, having the community doing it all. Typically transportation and processing were the two activities contracted to a third party. This normally had to do with the scale and conditions of the projects. Figure 3.3 aggregates the information from Table 3.2 into the general categories of the parties involved in the timber projects. After four to six years of participation in the timber projects, the community had become responsible for over half of the operational tasks, as well as the timber sales. They were also capable of contracting third parties to do some of the tasks that involved more heavy machinery. Table 3.3 summarizes the 12 tasks used in the questionnaire into four larger categories (project development, operation, marketing and management).

PAGE 123

111 The operation component is also divided into four main subcomponents (licensing, field operation, transportation and processing). The results for Table 3.3 were obtained by crossing information from the interviews from different questions, and also including information from the series of ten external stakeholders interviewed. Partner 33% Community 52% third Party 15% Figure 3.3: Task Distribution Overall Three aspects of project appropriation communities by can be extracted as the most significant from the above summary of Table 3.3. First, although most of the training and transfer of activities had occurred in field operations, this step was insufficient for an adequate appropriation by communities. For a real appropriation to occur, the planning and management tasks needed to be transferred. Given the communities organizational limitations, this would require more time and investment. Secondly, the fact that some communities had more control over the projects did not mean that they had a stronger organization, but rather that those projects were oriented towards individual operation. Thirdly, it is

PAGE 124

112 good to keep in mind that these four projects were meant to be pilots, from which to draw lessons for up-scaling community timber management Table 3.3: Project Appropriation by the CAs Component Project Appropriation by the CAs Project Development LOW: Neither the initiative, the design nor the fundraising for the initial projects came from the community. There was little understanding of the process and implications of raising money from donors. Operation: Licensing MEDIUM: The technical character of the licensing process required a forester to complete and sign the management plan for IBAMA. The legal procedure and the complications of dealing with the different bureaucracies (state & federal) required significant knowledge and time allocation, which the associations did not have. Field Operation HIGH: this was the area where project transfer had been must successful: after several harvest seasons community control over the process was substantial. It did vary in relation to the scale of the project, being higher in the smaller scale ones. This was where the partner organizations had the most expertise, and was also the area that concentrated most of the projects budgets. Transportation MEDIUM: Although most of the projects had planned to build transportation capacity (buying tractors and trucks), they were obliged to rely on third party contractors to be able to meet deadlines and avoid operation lags. Small-scale projects that used animal traction to move the logs within the forest had had more success in developing the capacity of the participants to manage log transportation. Capacity building had shifted from getting the community to do the transportation to training them in how to effectively hire a third party to do it. Processing LOW: Just like transportation, the construction of processing facilities had aimed at getting the middleman out, but transferring the capacity to do this had taken longer than planned. Three of the four projects planned to build sawmills to be run by the community, but only in one case was the facility working. Even in this case operation was not profitable. Marketing & Selling LOW: Except for the project with a set arrangement with a buyer (Cachoeira) the rest had experienced numerous difficulties to sell their products. Problems ranged from product quality to sales to a buyer that disappeared without paying. Certification had been sought as the alternative to guarantee more secure and profitable markets for the projects timber. By 2003 three of the four projects were certified by FSC. Administration & Management LOW: The differences in origin and project-design established significant differences among projects with regard to the elements of the enterprises that were managed at the community level. However, all projects recognized the lack of business management skills among participants and the limited collective action capacity of the groups. Although most of the time it went undocumented (particularly when it failed), experimentation was constantly taking place in these projects 26 As a 26 The CIFOR-UF ACM project documents the experimentation and social learning process over three harvest seasons in Porto Dias (Cunha, 2002).

PAGE 125

113 result of that experimentation, Cachoeira and Peixoto had progressively developed from an individual focus into more collective projects, where the association played a more central role. Synthesis Four community timber projects participated in the process: three from Acre (Cachoeira, Peixoto and Porto Dias), and one from Rondnia (Cautrio) Peixoto and Cachoeira were smaller projects that used family land tenure units as their production base and relied on animal transportation for the logs in the forest. Porto Dias and Cautrio were somewhat larger projects with mechanical transportation and incipient processing facilities. All projects were collaborative efforts between community associations and a partner organization (NGO and government). The GPF coalition had as its main driving factor the commercial limitations of the four community timber projects. Up until 2002, the two larger projects (Porto Dias and Cautrio) had not been able to generate profits from the harvested wood. The two smaller projects, Cachoeira and Peixoto, had very limited success selling their product. It became clear that one of the things that projects required to achieve economic sustainability was an innovative marketing strategy. To articulate this new strategy, project staff drew on the example of the Mexican Sociedad de Ejidos and proposed the creation of a coalition of the community projects to pursue the joint marketing of their products. This initiative was backed by the WWF as it also represented a way to streamline the forest certification of the community projects. WWF had supported forest certification of the community projects, but the small production of these projects dissuaded

PAGE 126

114 certified wood buyers. Joint commercialization would allow the communities to jointly offer substantial volumes of certified wood. As the purpose of this research was to assess the potential success of the GPF, I looked at a series of aspects regarding the relation between the community group and their project partner. These included: participation, leadership, management capacity, projects relationships, project training focus and project appropriation by communities. The projects appeared in general to be mostly driven by the partner organizations, with community groups playing a passive role. I list below the main findings from each variable: Participation: participation was stagnant and strategies to increase it depended upon the success of the timber projects, which were at the same time limited by the scarce participation. Leadership: Although leadership could generally described as paternalistic, an awareness of the problems associated with this trend was growing among the later project coordinators and the community leaders. Management Capacity The management capacity of the CAs did not match what was expected of them in the timber projects, nor was their learning as fast as the PO expected. The change of strategy to outsource certain portions of the operation was an adaptive solution, that allowed reducing stress in areas where the available capacity was very limited. Project relationships: Although in general relationships between parties in the projects were positive, this only meant that they were not conflictive. They were characterized by strong paternalism, with the POs staff holding the powerful position and the CAs playing a passive role to secure immediate benefits. Projects training focus: Training up to 2002 had a focus on the basic skills of logging, which prepared the CA participants to work in the projects but not necessarily to manage them. Conscious collaborative experimentation between POs and CAs had only started to happen by the time this study was conducted but it promised to provide a wider learning perspective Project appropriation by community: the training and transfer of activities had occurred in field operations but this step was insufficient for an adequate appropriation by communities. For a real appropriation to occur the planning and management tasks needed to be transferred

PAGE 127

115 The partner organizations mostly did planning and design activities, while operational tasks were shared, showing an increasing participation from the community participants. Projects also showed difficulties addressing the business dimension of the timber activity, concentrating more on the technical aspects. This problem was evident in the way projects were staffed and what their training priorities were, which emphasized the forestry aspect. In some projects, local participants had more control over the timber operation. However, this was the result of projects allowing more individualistic work; the collective action capacity of all community groups seemed to be at a fairly similar low level. My leading interests with regard to the internal factors component, was identifying the collaboration model (or collaboration pattern) that would work as a template for the GPF coalition process. It was my assumption that this collaboration model would be identified in the group-dynamics of both the associations and the partner organizations, as well as in the interactions between both groups. The main ground for both of these interactions (intra and inter group) was the timber projects, which paired each community association with at least one partner organization. An integral analysis of the five comparative variables made it clear that collaboration did not constitute a strategic aspect for these projects. Stakeholders mentioned collaboration as an important value but not as a skill. The technical orientation of the projects design, and the intense and complex work demand, concentrated the attention of the projects staff away from gaining an awareness of the collaboration model. The result was a paternalistic

PAGE 128

116 collaboration pattern, evident through all the variables analyzed, but particularly strong with regards to participation, leadership and relationships. Another aspect of the projects model of collaboration was the lack of attention given to social learning. 27 Social learning in the projects took place organically and intuitively. It was not incorporated in the evaluations or monitoring systems. Although tacit, the innovative condition of the projects made social learning pretty common and intense among the projects. The manejadores and the tcnicos constituted a community of practice, but it simply went unnoticed. These functional communities were made tighter by the high degree of uncertainty of the projects. Issues that represented that uncertainty were: the absence of any parameter for sustainability; the lack of marketing and commercialization know-how on the side of the tcnicos; and the conflicts where the tcnicos authority was challenged. In some cases conflicts represented clear indicators of learning and community empowerment; however, they tended to result in people (tcnicos or manejadores) pulling out of the projects. This mobility represented a loss of social and human capital from the projects. 27 Only one project had an initiative focused on documenting the collaborative learning process: Porto Dias ACM research project by UF-CIFOR-PESACRE-CTA

PAGE 129

CHAPTER 4 EXTERNAL FACTORS The objective of this chapter is to identify the key external constraints shaping the GPF coalition. According to my research framework, external factors such as policies, laws and markets would mold the structure and process of a CFC initiative, such as the GPF coalition. I hypothesized that the effect of these factors on the GPF emergence would be positive, as national and international conditions encouraged local natural resource innovations. This would contrast with the internal factors that I expected to hamper the development of the GPF. I interviewed government and NGO officials as well as natural resource researchers working in Acre and Rondnia (ten total). Also, I reviewed laws, policy documents, project proposals and project reports to characterize the role of key external stakeholders in community forestry in Acre and Rondnia. I developed a stakeholder map to match external and internal actors main interactions. The socio-political environment surrounding the emergence of the GPF could be described as positive yet conflictive. There is no doubt that the last fifteen years represent a transformation of the traditional Amazonian (and Brazilian) development policy style. Local populations and forests normally disregarded, or seen as obstacles to development, became an important part of the new paradigm of sustainability. Nevertheless the cultural, political and economic heritage of the old system persisted. Traits like institutional 117

PAGE 130

118 paternalism, electoral cycle expectations and weak grassroots organizational capabilities diluted the potential of the new sought out alternatives. The process of consolidating these alternatives was characterized by intense policy experimentation (Kainer et al., 2003). Institutional and organizational innovations over natural resource management were promoted simultaneously at various scales. These innovations were happening at the same time and influenced one another creating a very complex environment where issues, concepts, group interests and political agendas of various kinds interacted, reinforcing or neutralizing one another. The fate of these innovations was all tied up together. Certainly, their success would complete the transition into a new economy. However, unintended consequences of some of these innovations might result in tilting the direction of the reforms. The persistent uncertainty over the success of the policy experiments represented a political, as well as a social risk that characterized the environment of the GPF coalition initiative. In this chapter I appraise the socio-political and economic environment surrounding community timber extraction projects and their networking efforts. I start by reviewing public policies regarding land tenure, forest management, and social organization, as these areas have a direct relation to the GPF coalition effort. Next, I consider some of the timber market characteristics that challenge the communities marketing of their products. I map out the institutional stakeholders that accompany community timber extraction projects in Western Amazonia. The analysis section is devoted to the cross scale interaction

PAGE 131

119 affecting the GPF initiative. The following section presents the historical development of these interactions to the moment of the emergence of the GPF proposal in 2002. The final section summarizes the key findings and conclusions for the chapter Law and Policies There were three main areas of policy that affected the emergence of the GPF coalition: land tenure, forest management and social organization. Land tenure encompassed the creation and development of extractive reserves, a keystone innovation that emerged as a compromise between the social and environmental demands that confronted the newly democratic government of Brazil at the end of the nineteen eighties. The creation of extractive reserves represented the breaking point between the decaying economic and land tenure system of the old rubber era and a new order still in the making. The cornerstone of this new system was the extractive reserve model. The other aspects of policy analyzed in this section (forest management and social organization) were directly related to the creation of this land tenure institution. Land Tenure The cornerstone of the GPF coalitions policy environment was the extractive reserve model. Extractive reserves (RESEX) and agro-extractivist settlements (PAE) were established in the late 1980s by the Brazilian federal government. These two forms of legally sanctioned common property regimes secured the land tenure of the rubber tapper communities. The security of tenure empowered them to look beyond the traditional uses of the forest into other alternatives that would render more income and well being for their families.

PAGE 132

120 RESEX and PAEs allowed a redefinition of the relationship between extractivist communities and the forest where they had lived for years as mere occupants. Tenure security has been singled out as the most important element for a forest-based community development to succeed. However, it does not work by itself and must be backed up by a broader set of public policies (Agrawal, 2001; White & Martin, 2001). Also, the common tenure forced the rubber tappers to engage in new organizational arrangements not only to manage the land but also to relate economically to external agents. Legally sanctioned associations became the authority and the formal representation for PAEs and RESEX communities. Development projects as well as state subsidies were to be channeled through the associations for the extractivist groups. It is important to underline that prior to the RESEX and PAE regulations, such organizations did not exist in the communities. Although associations would be made up entirely by community members and represent local leadership and power relationships, as a form of organization they were foreign to the extractive communities. Laws that required the formation of associations made no reference to the potential cultural gap between existing forms of rubber tapper social organization and the association model. According to Cardoso (2002) the model for the extractive reserves was the indigenous reserves, already established in Brazilian Law. In the case of the indigenous reserve, the legal recognition of the territory was accompanied by the recognition of the existing social organization. In the case of

PAGE 133

121 rubber tappers, while aspects such as internal divisions of the land (colocaes) were incorporated, social organization was not; the law imposed associations 28 Starting in the early nineties and taking advantage of the new land security, development for the rubber tappers was sought by the same network of social movements and supporting organizations that had been key in establishing the RESEX model (Brown & Rosendo, 2000; Cardoso, 2002; Kainer et al., 2003). This took the form of community development projects centered on bringing forest products to new markets. Projects had the combined objectives of meeting community development needs while securing forest conservation. Some of these initiatives concentrated on the more traditional products such as rubber and Brazil nut; others focused on agricultural production or on new non-timber forest products (NTFPs) such as seeds and oils (copaiba). Timber extraction became part of the alternatives explored through these development projects. Summarizing: the RESEX model represented an important point of departure for the economic development of extractivist communities, as it secured their land tenure. Consolidating the legally required community associations remained a challenge, which would be closely linked to the community timber projects and the GPF initiative. 28 This issue is much more complex. Associations were part of the rubber-tapper proposal for the management of the Extractive Reserves. They were viewed as an adequate mechanism of guaranteeing community control over the future reserves. The lack of organizational capability at this level was recognized from the very beginning and support was sought to build such capacity (Cardoso, 2002). For this reason associations could not be considered as entirely imposed.

PAGE 134

122 Forest Management Forest management was an activity under IBAMAs (Brazilian Institute of the Environment) jurisdiction. IBAMA was the federal governments institution in charge of enacting and enforcing forest use legislation (Toni, 2003). Technical forest management was not required until 1995 when sustainable forest management plan guidelines were approved (P. Amaral & Amaral, 2000). Guidelines for small-scale timber management were not passed until 1998. As of 2004, guidelines for non-timber forest product harvesting had not been developed, although all forest uses should be under a management plan according to the law. The lack of technical instruments for management, and of an adequate staff to enforce the legislation, were a constant complaint from the community-based timber projects about IBAMAs role (P. H. C. Amaral, 2001). Nevertheless, financial support for most of the community-based timber projects came from the federal government. The G7 countries had established a fund controlled by the Federal Government for the conservation of the Amazon, known as the PPG7. This fund had a special portfolio for alternative development projects (PDA) that supported several of the initial timber projects in eastern and western Amazonia. This included two of the four projects analyzed in this study. State legislation and policies complemented the federal laws and programs. However, the western Amazonian states of Acre and Rondnia represented two opposite approaches to forest management at the state level. While traditional right wing political groups dominated Rondnias government, since 1998 Acre had had a progressive government based on the social movements of the rubber

PAGE 135

123 tappers. As a result, Rondnias development strategies were centered on large-scale agriculture and cattle ranching, while Acre was following a forest-centered strategy. This forest-centered strategy had generated a set of legislations and policies to advance sustainable use of the forest. Rondnias rubber tappers, on the other hand, had moved to timber management as a strategy to secure their extractive reserves against the governments explicit and implicit attempts to weaken the common property system (OSR, 2002). Acres situation was very encouraging, as the self proclaimed Governo da Foresta had enacted a whole package of policies to provide a framework for productive conservation practices(Kainer et al., 2003). As extractivist activities represented a more traditional form of forest use, the government started with a subsidy to rubber production (Chico Mendes law). This subsidy was strategically aimed at maintaining forest occupation, to halt rural migration to the cities while other economic alternatives were being developed. A state secretary for forestry and extractivism was created (SEFE) and placed in charge of developing markets for traditional and new NTFPs as well as timber-based projects. After its initial focus on NTFPs, the state government moved to emphasize sustainable timber extraction both at the large and small scale. In 2001 a state forestry law set the rules for the industrial concessions in State Forests ("Lei n o 1426", 2001). The idea was to attract foreign timber companies with the latest technology and technical expertise. The governments idea was that all production forests in the state would be producing certified timber.

PAGE 136

124 The lack of differentiated licensing processes for small-scale timber production was a serious obstacle for the community timber projects in Acre and Rondnia. A couple of the projects were suspended for over a year while regulations were passed and documentation was prepared and approved. The other significant problem for the projects was the lack of enforcement of the laws regarding illegal logging. Illegal timber kept local timber prices too low for the community sustainable production to be profitable. Social Organization The Governo da Forestas strategy for the development of rubber tapper communities in extractive reserves started by creating a subsidy for rubber that would motivate families to stay in the forest and avoid migrating to the cities. Although this plan had been somewhat successful, the overall perception was that rubber tapping would eventually disappear (Kainer et al., 2003; Stone, 2003). As explained by Carlos Vicente, at the time State Secretary of Forestry and Extractivism, the idea was to gain time while new alternatives were consolidated. Among these alternatives were copaiba oil, seeds, Brazil nuts and timber. The chances of these alternatives to take off depended, according to Vicente, on the rubber tappers organizational and managerial capacities, so building this capacity required action at different levels and through different mechanisms. The Chico Mendes law established that the subsidy was available only to rubber tapper associations. This requirement promoted collective action and organizational development.

PAGE 137

125 Two other state policies directly promoted social organization: one was law 1428 from January 2002 (the statute on civil society organizations) and the other was the Acre cooperative project, which started in 2001. Law 1428 defined the purpose and requirements to establish an association. Sustainable development and economic promotion were among the valid objectives. Nevertheless, one of the problems of this law was that it limited the economic activity of the associations. As non-profit organizations, they could not declare taxes or extend receipts for products such as timber. Legally, the only option was for rubber tappers to form cooperatives whose main purpose was economic. In practice, what associations did was to sell their products under the name of one of its members (normally the president) 29 The Cooperacre project was a large-scale integration of all NTFP community production. The government established Cooperacre, a cooperative made up of a number of smaller cooperatives and associations with rubber, Brazil nut and other NTFP production. Cooperacre was seen as a marketing platform, but also as an instrument to integrate the smaller groups and provide management and organizational skill training. This project started with an assessment of the organizational capacity of the rubber tapper groups. The study concluded that while capacity was low overall, there was also a concentration of most of the available human and social capital in some areas, creating a problematic regional disparity. 29 This issue was of great concern among the community timber projects given that relying on the president or a member of the supporting organization to sell the product further complicated an already complicated issue: the sale and the distribution of income. Solving this issue was one of the motivations behind the creation of the GPF coalition.

PAGE 138

126 Community timber activity was not considered to be part of Cooperacre. One reason for this was the very small number of communities involved in timber. Another was the controversy around timber extraction among the rubber tapper movement an aspect detailed in Kainer et al (2003); and also in Stone (2003). In 2002, during the fieldwork, I attended a meeting between the Acre cooperative project consultant and the GPF coalition promoters. At the time the idea was to explore the incorporation of timber as a product of Cooperacre. It became clear that in doing so the timber project POs would have to give up a significant part of their control over the projects. Cooperacre and SEFE staff would replace them. This did not please the GPF promoters, as they were all from this group of supporting organizations. While a next steps plan was sketched-out and the next meetings were set, the GPF promoters discreetly abandoned this alternative 30 The GPF coalition proposal was hard to fit into any of the available legal organizational frames (association, federation, cooperative, corporation). Confronted with the same problem, CFCs elsewhere opted to remain as informal groups, leaving the issue of legal formalization to a later stage in their development. Laws and public policies such as RESEX, the PDA pilot projects, or Acres Governo da Foresta provided a fostering environment for the development of the 30 Beyond the risk of losing control over the project to a stakeholder whom some groups did not trust (SEFE), the conversation also made clear that things at Coopeacre were not going as smoothly as planned. Leadership and management problems had obligated SEFE to step in and take control of the operations several times in the last year. The potential benefits of joining Coopeacre seemed very low as this organization was struggling with the same problems it meant to solve.

PAGE 139

127 community timber projects. These timber projects later gave ground to the idea of a coalition of timber communities, which is the object of this study. This sequential development is represented below in Figure 4.1, showing how national policies (RESEX, PDA) created the opportunity for other state policies (Acres forestry law, Chico Mendes law). These policies provided the enabling environment for the community timber projects and later the GPF coalition. Establishment of Extractive Reserves Community Timber Projects StateGovernment of the Forest:Forestry law, Social Org Law, Chico Mendes LawGPF Coalition Federal Government: forest-based Pilot ProjectsPPG7-PDA Figure 4.1: Policies Supporting Timber Extraction and the GPF Coalition These policy conditions were what I expected from my hypothesis. The only minor obstacle was the lack of an adequate formal structure for the GPF. This in itself should not be a problem, as the coalition would need to work on other aspects of development (building trust and operational capacity) before legal representation became an issue. Markets and Community Timber Projects The goal of the GPF coalition, according to its proponents, was to become the platform for the projects to interact with timber markets. Up to the date when this research fieldwork was conducted, timber sales had been sporadic, isolated and experimental. Only one of the projects had an established permanent commercial relationship with a client (Cachoeira). This relationship was prior to

PAGE 140

128 the project itself, as both the client woodcraft workshop and the community timber production were established as parts of the same initiative. Timber price in this case was set by a broad set of stakeholders that included the government, the donor agencies, the companies involved, and the community group leaders. For the rest of the timber projects, marketing, and in general the business dimension of the timber activities, were just beginning. Identification of the type of information needed and the sources was being developed. Two signals from the market were important to develop the GPF initiative. First, there was recognition that at the local and regional markets, illegal timber brought lumber prices below what would be sustainable for community projects. A single-year harvesting operation with the sustainable management plans used in this community project could not be afforded with the prices offered for total production. Efforts to add value to the product would require more investment and ultimately an increase in operation costs. Project staff had tried leaving the marketing in the hands of the individual producers, but neither this nor whole-project deals had experienced profits. The opportunity for profitability (and economic sustainability) seemed to depend on product certification. A community of certified-timber buyers was identified and organized by another Brazilian organization (Friends of the Earth Brazil). The certification label intended to provide price recognition for the ecological and social characteristics of these projects. In order to sell certified timber, certain conditions must be achieved, particularly in terms of volume and quality. Individual projects would not meet the

PAGE 141

129 necessary volume. This established a strong connection between the GPF proposal and the certification of the projects production. In order for certification to make sense (and the projects to have a chance to make profits) the projects had to come together and be able to offer enough lumber volume for the buyers to become interested. An important thing to consider was that certification was also a system in the making. As a market tool for end-consumers to select environmentally sustainable products, it was still being developed. In the case of timber, the total volume of certified wood commercialized was still very small to be considered a generalized commercial practice (Scherr et al., 2002). The chance for certification to provide the promised financial sustainability to the projects was still to be seen. It depended upon the whole certification systems consolidation as a market mechanism. The promotion of certification among the west Amazonia community timber projects had not come from the certified products markets, but from the organizations promoting the consolidation of the certification system. This included external support organizations like WWF and the state government forest secretary, SEFE. Risks and uncertainties associated with infant stages of this market mechanisms were high; however projects didnt seem to be working on other market alternatives. External Stakeholders In the previous section on the internal conditions for the GPF, I laid out the basic stakeholder map presenting the parties involved in the formation of the coalition. I complement this initial layer with the set of external stakeholders

PAGE 142

130 involved in the community timber activity. These actors were mainly organizations both public and private and played either a controlling or a supporting role to the groups involved in the timber projects. I have divided them into three groups: federal agencies, state agencies and donors. Among them the state agencies, and the state government as a whole, were the most involved stakeholders. Reasons for this were explained above in the policy section. One donor in particular, the World Wildlife Fund (WWF), had a particular interest in forest certification and had played a major role in getting the timber projects products certified. It also had had a significant role in the development of the GPF coalition, because commercially viable certification required larger amounts of timber than those produced by the individual projects. Joint marketing of timber through the GPF would attract more certified timber buyers. Federal Government The federal government had a number of agencies involved in resource management and conservation in the Amazonian region. The most important with regard to timber activity was IBAMA, as it was in charge of deforestation and timber harvesting control. Approval of the sustainable forest management plan was under IBAMAs control; however, the lack of adequate resources limited the quality of this control. Part of the problem was the experimental condition of most of the small-scale timber activity. Management parameters were at best based on industrial scale projects. IBAMAs control agents recognized the lack of science-based criteria for the standards applied to the community-based timber projects.

PAGE 143

131 Within IBAMA a special office was set up to oversee extractive reserves. This was the National Council for Traditional Populations (CNPT). CNPT oversaw the creation and maintenance of RESEX. It also established the main management policies and the training of the community groups within the RESEX. Although CNPT had had little to do with the current timber projects, none of which took place in federal RESEX, it was expected that CNPT would become a strong promoter of timber management within RESEX. The new federal government administration that took office in 2002 had taken the example of Acres Governo da Foresta (along with some of its staff) and wanted to promote a timber-based development in extractive reserves. In the past, Federal support to community timber projects had been limited to the financial aspect, by means of the Pilot Projects Program (PDA) component of the PPG7 fund. The allocation of these funds was not entirely in the hands of the federal government. It had to negotiate this with the donor countries (from the G7 group). Federal government policies had been adequate for the development of community timber projects so far. It was likely that this condition would remain. The persistent problem was the lack of effective control over illegal activities, which created an economic disincentive for sustainable timber management. Acre State Government Of the four community timber projects analyzed in this study, three were located in the state of Acre and one in the state of Rondnia. As I explained above, while the Rondnia government was very passive and was described by some of the interviewees as anti-community development, the government of Acre was not only active but had placed forest at the center of its development

PAGE 144

132 platform. The Government of Acre was directly involved in community timber activity and followed very closely the development of the GPF. For this reason the government of Acre was considered in detail in this study, but not the government of Rondnia. The state government of Acre was a key player in the development of community timber activities. Since 1999 when the Viana administration took office (re-elected in 2002), not only had laws and policies been drafted, but a number of institutions had also been created or re-launched to enforce the new sustainable development policies. The Acre Institute for the Environment (IMAC) directed the ecological and economic zoning plan. This was a statewide effort to identify land use and land capability to establish a baseline for development. This comprehensive instrument integrated a variety of information sources into a set of documents that were later made available to the different institutions and stakeholders. The Secretary of Agricultural Extension (SEATER) was in charge of rural development extension, some of which was carried out by NGOs under contract with the state government. With a new emphasis on agro-forestry, SEATER became a key player in one of the community timber projects as it provided technical support after a dispute led to the withdrawal of the SEFE team. SEFE was the Secretary of Forest and Extractivism; the Viana government created it to promote forest-centered development, particularly for the rubber tapper communities. SEFE was responsible for the application of the neo

PAGE 145

133 extractivist 31 policies such as the Chico Mendes Law and the Coopeacre initiative in the above section. SEFE also developed a strategy to bring large timber companies to Acre under a concession system to harvest timber from the State Forest, a new type of direct use protected area. According to SEFEs forest policy, an external body would certify both community and industrial timber production. This secretary was also interested in expanding timber activities to new common property units such as the Chico Mendes Extractive Reserve. SEFEs interest in certification and its initial role in the Cachoeira timber project had kept them close to the GPF coalition initiative. They had hosted some of the small committee meetings and also tried to incorporate the GPF under the umbrella of Coopeacre, the state supported extractive products cooperative. This idea was abandoned early in the process, as I explained earlier in this chapter. Kainer et al. (2003) mentioned the drain of human resources from the university and the NGOs to the government under the Viana administration. According to those authors, this left the organizations short-staffed at a moment when the government required more of their collaboration. Based on my interviews, there are two other aspects that I would like to point out regarding the NGO-Government relationship. First, was the fact that transferring the staff from the NGOs to the government also brought in a few well-established interpersonal 31 This term was coined by Jose Fernandes do Rgo, Acre State Secretary of Production during Vianas first administration. The concept refers to the transformation of the traditional extractivist economy based on rubber and Brazil nuts into an a more diverse economy based on a variety of forest products and supported by an adequate level of social organization and government support(Rgo, 1999). Kainer et al. (2003) explain how the governo da foresta used this concept to articulate its forest-centered policies.

PAGE 146

134 conflicts that existed among NGO leaders. The second was the fact that with the Viana government, debate and discussions over policy alternatives were clearly reduced. Conflicts among social-movement and NGO leaders had always been present. However the intensity of the struggle that led to the creation of extractive reserves and later led to the electoral success of the progressive parties coalition in Acre halted these disputes. Some of these conflicts persisted becoming in some cases conflicts between the government agencies and the NGOs or social organization. Although most of them were not explicit, they were a strong factor influencing the implementation of government policies regarding community timber activities. Prior to the Viana administration, most of the voices arguing against government policies came from within the more progressive NGO and social organizations. Many key staff members transferred from these organizations to the new government. The role of the traditional political right wing groups also diminished, especially after the 2002 elections, when they lost by a landslide. Critical voices arguing against the administrations policies became scarce, especially from within the more progressive NGO and social organizations. With the reelection of the Jorge Vianas Governo da Floresta, in 2002, plans to expand community timber extraction were swiftly developed. SEFE was divided into two secretaries: Timber Extraction (SEF) and Community Based Production (SEPROF). The secretary for forestry moved to work with 20

PAGE 147

135 communities on timber extraction. Eleven of these communities supported by the state had since become members of the GPF 32 The state governments relationship with the community timber projects was very positive. The same was true regarding the GPF proposal. However, potential tensions would arise from the differences in time-scales between the state government plans and projects. The government on its four-year electoral cycle would need results at a faster pace than the community project had achieved thus far. World Wildlife Fund WWF had a much more direct role in the use and the impact of the resources they provided for specific aspects of the projects (certification). WWF had provided the seed money for the start up of the GPF initiative. One of the Projects (Cautrio) had had a long-standing relationship with WWF, who had provided funding for the project since the beginning and was key in its original design. WWF perceived itself not a donor but as a partner of the NGO that it supported. This paosition, although more committed than a hands-off donor, tended to be more conflictive. This problem was highlighted both by WWF representatives and their partner NGOs. The degree of interpersonal trust and correspondence in project vision are not always there. This situation generated different degrees of conflicts that, although not always explicit, ended up affecting the projects implementation. This active-donor role had demanded more time 32 Marcelo Argelles de Souza, Secretaria de Florestas personal communication.

PAGE 148

136 and energy than WWF had expected, and had affected the projects budgets and schedules. Furniture Company AVER AVER furniture company had very little connection to the GPF initiative. AVER was the partner and principal buyer for the Cachoeira project and had its manufacturing plant in the city of Xapuri. This company shared space with the church and state supported Polo Moveleiro project. This project provided infrastructure for companies to come to Xapuri to process timber products. This initiative was parallel to the Cachoeira project, but was independent from it. Currently their relationship was mainly as buyer for the community projects timber. External Stakeholder Map Figure 4.2 is a graphic representation of the GPF initiative stakeholders and their relationships. The core section of this diagram was presented in chapter three to describe the community of groups directly involved in the GPF. As a progression of that version, these new ones include the related public agencies, NGOs and private companies. These new organizations do not participate directly in any of the timber projects. On the left side of the diagram are the state government agencies: IMAC, SEFE and SEATER. SEATER had a relationship with the Cachoeira project as it provided the forest engineer for the project. SEFE that was earlier involved in Cachoeira had moved to provide support for the GPF initiative. SEFE hosted meetings between the interested parties and followed closely the GPF process. On the top, IBAMA, CNPT and PDA represented the federal government agencies. PDA provided financial support to

PAGE 149

137 Porto Dias and Peixoto projects. ITTO also provided funding to Peixoto and Porto Dias as WWF did to Cautrio. WWF, with its interest in promoting the certification of forest had also supported the development of the GPF coalition. The AVER company was also linked to the Cachoeira project, from whom they acquired most of their raw material. ITTO PDA WWF IBAMA AVER RegionalMarket International Market. Local Market. CNPT IMAC \000\000\000\000\000\000\000\000\000\000\000\000\000\000\000)TjETEMC /P <>BDC Qq226.44 438.23999 m223.44 438.12 l220.67998 437.88 l217.8 437.39999 l215.10001 436.80002 l212.58 435.95999 l210.3 435 l208.01999 433.91998 l205.98 432.72 l204.17998 431.34 l202.5 429.89999 l201 428.34 l199.8 426.78 l198.84 424.97998 l198.12 423.17999 l197.64 421.31998 l197.51999 419.39999 l197.64 417.47998 l198.12 415.67999 l198.84 413.81998 l199.8 412.01999 l201 410.45999 l202.5 408.89999 l204.17998 407.45999 l205.98 406.13998 l208.01999 404.88 l210.3 403.80002 l212.58 402.84 l215.10001 402 l217.8 401.39999 l220.67998 400.91998 l223.44 400.67999 l226.44 400.56 l229.44 400.67999 l232.37999 400.91998 l235.14 401.39999 l237.78 402 l240.3 402.84 l242.58 403.80002 l244.86 404.88 l246.95998 406.13998 l248.75998 407.45999 l250.43999 408.89999 l251.87999 410.45999 l253.08 412.01999 l254.03998 413.81998 l254.75998 415.67999 l255.23999 417.47998 l255.36 419.39999 l255.23999 421.31998 l254.75998 423.17999 l254.03998 424.97998 l253.08 426.78 l251.87999 428.34 l250.43999 429.89999 l248.75998 431.34 l246.95998 432.72 l244.86 433.91998 l242.58 435 l240.3 435.95999 l237.78 436.80002 l235.14 437.39999 l232.37999 437.88 l229.44 438.12 l226.44 438.23999 lW* n0 0 0 scnBT/T3_0 1 Tf1.92 0 0 -11.52 195.84 437.75998 Tm(\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000)TjETEMC /P <>BDC Qq226.44 438.23999 m223.44 438.12 l220.67998 437.88 l217.8 437.39999 l215.10001 436.80002 l212.58 435.95999 l210.3 435 l208.01999 433.91998 l205.98 432.72 l204.17998 431.34 l202.5 429.89999 l201 428.34 l199.8 426.78 l198.84 424.97998 l198.12 423.17999 l197.64 421.31998 l197.51999 419.39999 l197.64 417.47998 l198.12 415.67999 l198.84 413.81998 l199.8 412.01999 l201 410.45999 l202.5 408.89999 l204.17998 407.45999 l205.98 406.13998 l208.01999 404.88 l210.3 403.80002 l212.58 402.84 l215.10001 402 l217.8 401.39999 l220.67998 400.91998 l223.44 400.67999 l226.44 400.56 l229.44 400.67999 l232.37999 400.91998 l235.14 401.39999 l237.78 402 l240.3 402.84 l242.58 403.80002 l244.86 404.88 l246.95998 406.13998 l248.75998 407.45999 l250.43999 408.89999 l251.87999 410.45999 l253.08 412.01999 l254.03998 413.81998 l254.75998 415.67999 l255.23999 417.47998 l255.36 419.39999 l255.23999 421.31998 l254.75998 423.17999 l254.03998 424.97998 l253.08 426.78 l251.87999 428.34 l250.43999 429.89999 l248.75998 431.34 l246.95998 432.72 l244.86 433.91998 l242.58 435 l240.3 435.95999 l237.78 436.80002 l235.14 437.39999 l232.37999 437.88 l229.44 438.12 l226.44 438.23999 lW* n0 0 0 scnBT/T3_0 1 Tf1.92 0 0 -11.52 195.84 426.23999 Tm(\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000)TjETEMC /P <>BDC Qq226.44 438.23999 m223.44 438.12 l220.67998 437.88 l217.8 437.39999 l215.10001 436.80002 l212.58 435.95999 l210.3 435 l208.01999 433.91998 l205.98 432.72 l204.17998 431.34 l202.5 429.89999 l201 428.34 l199.8 426.78 l198.84 424.97998 l198.12 423.17999 l197.64 421.31998 l197.51999 419.39999 l197.64 417.47998 l198.12 415.67999 l198.84 413.81998 l199.8 412.01999 l201 410.45999 l202.5 408.89999 l204.17998 407.45999 l205.98 406.13998 l208.01999 404.88 l210.3 403.80002 l212.58 402.84 l215.10001 402 l217.8 401.39999 l220.67998 400.91998 l223.44 400.67999 l226.44 400.56 l229.44 400.67999 l232.37999 400.91998 l235.14 401.39999 l237.78 402 l240.3 402.84 l242.58 403.80002 l244.86 404.88 l246.95998 406.13998 l248.75998 407.45999 l250.43999 408.89999 l251.87999 410.45999 l253.08 412.01999 l254.03998 413.81998 l254.75998 415.67999 l255.23999 417.47998 l255.36 419.39999 l255.23999 421.31998 l254.75998 423.17999 l254.03998 424.97998 l253.08 426.78 l251.87999 428.34 l250.43999 429.89999 l248.75998 431.34 l246.95998 432.72 l244.86 433.91998 l242.58 435 l240.3 435.95999 l237.78 436.80002 l235.14 437.39999 l232.37999 437.88 l229.44 438.12 l226.44 438.23999 lW* n0 0 0 scnBT/T3_0 1 Tf1.92 0 0 -11.52 195.84 414.72 Tm(\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000)TjETEMC /P <>BDC Qq226.44 438.23999 m223.44 438.12 l220.67998 437.88 l217.8 437.39999 l215.10001 436.80002 l212.58 435.95999 l210.3 435 l208.01999 433.91998 l205.98 432.72 l204.17998 431.34 l202.5 429.89999 l201 428.34 l199.8 426.78 l198.84 424.97998 l198.12 423.17999 l197.64 421.31998 l197.51999 419.39999 l197.64 417.47998 l198.12 415.67999 l198.84 413.81998 l199.8 412.01999 l201 410.45999 l202.5 408.89999 l204.17998 407.45999 l205.98 406.13998 l208.01999 404.88 l210.3 403.80002 l212.58 402.84 l215.10001 402 l217.8 401.39999 l220.67998 400.91998 l223.44 400.67999 l226.44 400.56 l229.44 400.67999 l232.37999 400.91998 l235.14 401.39999 l237.78 402 l240.3 402.84 l242.58 403.80002 l244.86 404.88 l246.95998 406.13998 l248.75998 407.45999 l250.43999 408.89999 l251.87999 410.45999 l253.08 412.01999 l254.03998 413.81998 l254.75998 415.67999 l255.23999 417.47998 l255.36 419.39999 l255.23999 421.31998 l254.75998 423.17999 l254.03998 424.97998 l253.08 426.78 l251.87999 428.34 l250.43999 429.89999 l248.75998 431.34 l246.95998 432.72 l244.86 433.91998 l242.58 435 l240.3 435.95999 l237.78 436.80002 l235.14 437.39999 l232.37999 437.88 l229.44 438.12 l226.44 438.23999 lW* n0 0 0 scnBT/T3_0 1 Tf1.92 0 0 -11.52 195.84 403.20001 Tm(\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000)TjETEMC /InlineShape <>BDC Q0.937 1 0.659 scn197.51999 400.31998 58.08 37.92 ref255.33 419.39999 m255.33 419.39999 l255.33 429.78302 242.38701 438.20999 226.44 438.20999 c210.49301 438.20999 197.55 429.78302 197.55 419.39999 c197.55 409.01703 210.49301 400.59 226.44 400.59 c242.38701 400.59 255.33 409.01703 255.33 419.39999 chS350.60999 512.10004 m350.60999 512.10004 l350.60999 522.367 337.74802 530.70001 321.89999 530.70001 c306.052 530.70001 293.19 522.367 293.19 512.10004 c293.19 501.83304 306.052 493.5 321.89999 493.5 c337.74802 493.5 350.60999 501.83304 350.60999 512.10004 cf350.78998 512.10004 m350.78998 512.10004 l350.78998 522.46704 337.84702 530.88 321.89999 530.88 c305.953 530.88 293.00998 522.46704 293.00998 512.10004 c293.00998 501.73303 305.953 493.31998 321.89999 493.31998 c337.84702 493.31998 350.78998 501.73303 350.78998 512.10004 chS0.933 0.161 0.17999 scn243.65999 438.35999 m229.2 418.07999 l258.12 418.07999 l243.65999 438.35999 lf*0.48 w 243.78 438.23999 m229.31998 417.95999 lS0.36 w 229.31998 417.95999 m258.23999 417.95999 lS0.48 w 258.23999 417.95999 m243.78 438.23999 lS320.34 441.30002 m310.20001 423.89999 l330.41998 423.89999 l320.34 441.30002 lf*320.45999 441.17999 m310.32001 423.78 lS0.36 w 310.32001 423.78 m330.53998 423.78 lS0.48 w 330.53998 423.78 m320.45999 441.17999 lS305.88 528.12 m295.73999 510.84 l315.95999 510.84 l305.88 528.12 lf*306 528 m295.85999 510.72 lS0.36 w 295.85999 510.72 m316.07999 510.72 lS0.48 w 316.07999 510.72 m306 528 lS260.22 519.53998 m246.60001 502.13986 l275.51999 502.13986 l260.22 519.53998 lf*260.34 519.41998 m246.72 502.01999 lS0.36 w 246.72 502.01999 m275.63998 502.01999 lS0.48 w 275.63998 502.01999 m260.34 519.41998 lS353.34 449.70001 m339.35999 441.06 l348.06 427.01999 l362.03998 435.72 l353.34 449.70001 lf*353.45999 449.57999 m339.47998 440.88 l348.17999 426.89999 l362.16 435.60001 l353.45999 449.57999 lS341.03998 522.35999 m327.66 512.82001 l337.20001 499.37985 l350.57999 508.91995 l341.03998 522.35999 lf*341.16 522.23999 m327.78 512.64001 l337.32001 499.25998 l350.70001 508.80002 l341.16 522.23999 lS212.45999 522.23999 m203.45999 508.56 l217.08 499.5 l226.08 513.17999 l212.45999 522.23999 lf*212.58 522.12 m203.58 508.44 l217.2 499.37985 l226.2 513.06 l212.58 522.12 lS1 0.612 0.188 scn238.01999 464.34 89.64 17.39999 refEMC /P <>BDC 0 0 0 scnBT/TT2 1 Tf-0.0099 Tc 11.58 0 0 11.58 244.62 468.72528 Tm(GPF Coalition PeixotoPto. DiasCautrioCachoeiraCTAEMBRAPAECOPOREA.Pt D.A. PA.CA.C SEFE SEATER Figure 4.2: Community Timber External Stakeholders Cross-scale Interactions Interactions among external and internal stakeholders were more diverse and dynamic than could be displayed in a diagram such as Figure 4.2. They also had a direct impact on the use of natural resources and the consolidation or weakening of social institutions and organizations. In this section, my objective is to take the variables analyzed so far in this chapter (policies, laws, market, stakeholders) and represent the type of dynamics that result from their interaction, to help understand the complexities in the evolution of a social

PAGE 150

138 innovation such as the GPF coalition. As a complex phenomenon, the GPF process was not the result of a set of discreet and isolated influences, but part of a larger, evolving web of elements (policies, institutions, organizations). Here I look first at the interactions that led to the establishment of RESEX as a novel land tenure institution in the late eighties, and then at the situation more than a decade later when the GPF was being proposed. Extractive Reserves Starting in the eighties the international campaigns set up by different social movements against multilateral development banks, such as the Inter-American Development Bank (IDB) and the World Bank (WB) brought attention to projects like the Polonoroeste. This project was part of the cross-Amazon development initiative of the Brazilian military government. Polonoroeste became a showcase of the harmful environmental impacts of government-led, WB-supported projects. It was in the context of this international controversy that international attention was drawn to the seringuiero (rubber tapper) movement in Brazil. Led by Chico Mendes, seringuieros suddenly found support beyond Brazils borders, and this eventually led to their major successes (like the establishment of the RESEX). International attention towards these issues resulted in the oversimplification of an otherwise complex movement. The people of the forest label overshadowed the development goals of the movement, such as securing land tenure and increasing income. This label was later reinforced in academic circles or government reports, such as the Zoneamento Ecolgico-Econmico.

PAGE 151

139 Under this romantic vision, policies and support initiatives were geared toward rubber tappers at the risk of falling into the ecologically-noble savage paradox. Incentives were created to continue with activities that otherwise would disappear. This could possibly undermine the capacity of these groups to adapt to new conditions and generate new knowledge. An example was the Chico Mendes Law aimed at subsidizing rubber prices to stimulate rubber production. The need to abandon the romantic vision of RESEX also requires that we understand them as much more than simply a representation of local interests. RESEX represented a compromise among a wide range of stakeholders from international environmental groups to the Brazilian government to individual rubber tappers. For these reasons, and because RESEX were complex institutions made partly, and not solely, of local knowledge, the future of RESEX required multi-scale considerations of this diversity of stakeholders. Political Dilemma I pointed out before how the GPF coalition sat embedded in a multi-scale system. Stakeholders, policies, institutions and markets interplayed, constraining or fostering the GPF initiative. It was generally assumed that the larger scales were less dynamic and that elements of those scales tended to be more stable (Holling, 2002). However, the situation surrounding the GPF seemed to differ from this expected pattern, in that relevant elements from the higher scales tended to be in somewhat of an unstable condition. Figure 4.3 below, presents a series of concepts, policies and practices organized according to their scale of origin.

PAGE 152

140 These elements were all related (directly or indirectly) to the GPF initiative. They all represented resource management innovations recently conceptualized and in the process of being operationalized. According to Holling and Gunderson (Holling & Gunderson, 2002), this was the stage in their adaptive cycle in which they would be more susceptible to changes coming from the smaller scales. The weak or unfinished condition of these large-scale elements created a hierarchical web of innovations in which elements were mutually affecting each other. International: National: State:Coalition Proposal PAE/RESEX:Sustainable DevelopmentSustainable ForestryForest CertificationExtractive ReservesGovernment of the ForestNeo-ExtractivismCommunity timberprojectsForest Management GPF coalition Figure 4.3: Hierarchical Web of Innovations This situation was exemplified with the relationship between certification, community timber management and the GPF. Although forest certification responded to a larger-scale initiative, it was presented as a market-support tool for the community timber projects. For certification to succeed, it needed a critical number of both certified projects and certified buyers. At the time, it was just building that critical base, and the community timber projects were key in that

PAGE 153

141 process. It would be hard to identify whether the GPF projects were dependent on certification or if it was the other way around. Another example of this regards the RESEX model. Although RESEX already had a legal definition, as a social-economic institution it was still blurry. These were now (with the legal definition and the number of examples) shells (Alcorn & Toledo, 1998) that framed the process of defining RESEX as a social-economic institution. The RESEX model was meant to preserve a forest culture. But its mere creation could not achieve this. It had only served as a shell that housed the efforts (such as the timber projects) that could preserve that forest culture. However, the efforts inevitably would transform that forest culture into something unexpected and probably not intended, in the creation of the RESEX. At the end the RESEX became an instrument not to preserve, but to innovate. Synthesis The GPF was conceived within a fairly positive political and institutional environment resulting from the large-scale social and political reorganization process experienced in the 80s. This reorganization was shaped by changes at several levels: global changes such as the emergence of the sustainable development concept and the relevance of issues like biodiversity and climate change; national changes, particularly the return to democracy and the end of the aggressive development policies in Amazonia like the Polonoroeste; and regional changes such as the creation of extractive reserves and agro-extractivist projects. The RESEX represented an important point of departure for the extractivist communities economic development, as it secured their land tenure. Along with

PAGE 154

142 the RESEX the other key law for community timber was IBAMAs 1998 establishment of differentiated requirements for small-scale projects. In terms of its social structure, the GPF coalition proposal was hard to fit into any of the available legal organizational frames (association, federation, cooperative, corporation). However, this was more likely to be an issue for the later development of the coalition, as it had been for other similar coalitions. Low local prices for timber caused by illegal timber extraction raised the projects interest in forest certification. Certification had not been proven as a tool for securing alternative markets. Instead it was another emerging innovation, strongly promoted by WWF. WWF was interested in consolidating forest certification as a market mechanism used both by producers and consumers, and so supported the certification of the projects involved in the GPF. WWF saw the GPF as a springboard to attract certified wood buyers. These external conditions were what I expected from my hypothesis. However, I had assumed the constraints would be unidirectional, going from the higher (external) scale to the lower (internal) scale. The research results made clear that there was interdependency between lower scale (GPF coalition) and higher scale (certification, RESEX) innovations. This mutual conditioning between the GPF proposal (and the community timber projects) and the panarchy of innovations promoted at higher scales (certification, neo-extractivism, sustainable forestry, RESEX) helped explain the direct participation of some of the external stakeholders in the GPF process. As will become clearer in the next chapter on the GPF design, both WWF and Acre state government

PAGE 155

143 played a significant role in the emergence of the GPF. Although the external stakeholders participation could correspond in part to their support for positive local initiative, they also seemed to have had a direct interest in this coalition as a means to achieve their institutional interests.

PAGE 156

CHAPTER 5 DESIGN FACTORS As suggested by Krishna (Krishna, 2003) and Powell (Powell, 1990) the actual design of a network (coalition in our case) plays a key role in its acceptance and its capacity to promote effective collaboration. I believed a coalitions design would influence the amount and the quality of collaboration among the involved parties. Certainly, a degree of previous collaboration was required in order to form a coalition. I analyzed this type of collaboration in the internal component of the framework (Chapter 3). In this chapter I explore the effect of the coalitions design and implementation on its own development. I also examine how the process of setting up the coalition fed back into the original idea of the GPF. In this case, it entailed finding out how the involved parties adapted the original idea to meet the social political context of Acre and Rondnia. As the results of this chapter show, the GPF proposal went through a significant degree of adaptation from its original conceptualization. However, this adaptation seemed to compromise the original objective of the coalition. While the initial proposal aimed at a CA driven coalition, the real GPF resembled a typical NGO project in which community participants were more passive beneficiaries than dynamic partners. This sections results represent the last component of the analytical framework established in Chapter Two. As stated in my hypothesis I expected 144

PAGE 157

145 the results from the design factors to prove positive for the development of the GPF. These would stand together with the external factors (also positive) and contrast with the internal factors that I expected to be negative for the GPF formation. This chapter is organized into four sections. The first presents the concept design and ideal structure of the GPF, as it was originally presented by CTA. The second section presents the perceptions that stakeholders had of this original idea. A matrix synthesizes and compares the perspectives of the three main stakeholder groups (CA, POs and external stakeholders). The third section presents the history of the GPF innovation process using a timetable. This section also describes the functioning structure of the GPF in the period from the introduction of the proposal to the moment the research was conducted in 2002, and adds more recent information on the period from 2002 to 2004. A final section synthesizes the major findings and conclusions of the chapter. Concept and Design The Grupo de Produtores Florestais, GPF, was thought of as an organization made up of community associations from Acre and Rondnia, all involved in sustainable timber management. The associations developed their timber activities through projects supported mostly by NGOs or state government agencies. Projects took place in clearly established individual or collective areas such as agrarian development settlements, agro-extractivist projects and extractive reserves. Initially the purpose of the GPF coalition was to act as a marketing center, with a warehouse located in a city (most likely Rio Branco),

PAGE 158

146 where the different community projects would bring their products for buyers to purchase. As described in the section History of The GPF in Chapter One, the main reason for the development of the GPF was to address the selling problems that the different timber projects were experiencing. It was originally proposed that the GPF would become a cooperative made up of community associations, or options like an association or a private company were also considered. Regardless of the final legal structure, CTA initially envisioned the GPF as having a horizontal and independent structure (Figure 5.1). GPF Coalition A.Pto. DiasCTAEMBRAPAECOPORESEATERAPRUMAAGUAPEAMPPAE-CM Committee Coordinator Timber buyers New Markets External Support Figure 5.1: GPF Coalition Structure The associations would form a cooperative with a central committee comprised of association representatives. The committee would be the executive branch of the group, with a coordinator in charge of the timber sales. This coordinator was initially envisioned as being a member of one of the associations

PAGE 159

147 and would also be responsible for opening new markets and other external relations. As a cooperative made up of community associations, members were meant to play the leading role in the decision making of the new group. As shown in Figure 5.1, the partner organizations, such as CTA or EMBRAPA, would play an external role providing technical support upon request. As the cooperative grew, it would hire its own technical and managerial staff that would respond to the cooperative executive board. To build the GPF coalition into an effective cooperative, the proponents listed available resources. The CAs would contribute equipment (trucks, tractors) from the different projects as well as the expertise developed through the four to six years of timber experience. Another significant resource for the GPF was motivation, which the visit to the ejidos in Mexico had generated among project participants. The success of the Asociacin de Ejidos Coalition from Quintana Roo was seen as an example to follow in Acre and Rondnia regarding community timber development. External partners with an interest in supporting the coalition also represented a source of resources. WWF wanted to consolidate the timber certification system and therefore had a direct interest in promoting the GPF idea. For WWF, the coalition would be able to offer larger volumes of certified wood, making it attractive for buyers from southern Brazil who wanted to acquire community certified wood. The state of Acres Governo da Foresta was also interested in expanding community timber activity to other communities and

PAGE 160

148 viewed the GPF as a platform for this purpose. The state government hosted many of the initial GPF meetings and explored the possibility of using an existing state created cooperative (Cooperacre) to incubate the GPF In summary, all the stakeholders received the initial proposal of the GPF coalition acting as a marketing center very well. The GPF was envisioned as a horizontal and independent organization of CAs, with support and resources coming from the POs and the external stakeholders such as WWF and the state government of Acre. Perceptions about the GPF Perceptions about the GPF proposal were collected during interviews held in the summer of 2002 with representatives of all participating associations, as well as from organizations involved in supporting community timber projects, in Acre and Rondnia I present three aspects of each groups perspectives: their definition and purpose for the GPF, their expectations for the success of the initiative (what would the GPF achieve?) and finally, the obstacles to, and challenges for, the consolidation of the GPF proposal. Partner Organizations The perceptions of the GPFs purpose by the partner organizations was that it would fill an existing gap in most of the projects regarding marketing of products. As projects reached the end of the five year funding cycle of most funding agencies, partners feared that it would be very hard for the isolated projects to raise more money for their marketing needs. Donors might not be willing to put more money into the projects, since they were not yet showing economic viability.

PAGE 161

149 Although not explicit in the partner organizations opinions, the GPF represented, in itself, a mechanism for attracting more donor support. Donors would potentially be more interested in the coalition as an effective way to address the business aspect of community timber. This strategy came from the notion that other countries such as Mexico and Guatemala had had success with similar structures. Funds raised towards the development of the GPF would go to the existing partner organizations, as the associations lacked the management capacity to deal with those funds. A successful GPF would allow most of the projects staff to concentrate on the forestry of the projects, which fell more into their expertise and interest. The organizational and business aspects of the community projects were issues beyond most of the coordinators experience. The general perception was that forestry aspects and ecological aspects of these projects were not receiving enough attention. For example, the necessary monitoring of re-growth to validate the cutting cycle period was absent in most of the projects. The exploration of new species that could be brought to market was seen as another aspect not receiving sufficient attention. For the partner organizations, the main obstacle to the GPF working as a marketing platform was the difficulty in standardizing the products from the different projects. Products from each project had a number of differences, due to the differences in the technical approaches of the project directors and staff, the difference in the forests at each community, and the difference in the equipment available to finalize the products at each project. The perception was that this diversity would hamper the integration process and that an adequate

PAGE 162

150 degree of standardization was needed. It was presumed that such standardization would require a homogenization of the project management. Paradoxically, this homogenization was feared, as it might represent an external intervention (evaluation) into fairly independent projects. Another challenge for the GPF was the limited resources available. Up until that point, no funding source had been made available to adequately finance the coalition initiative. WWF provided funding for a coordinator, but lack of funds for the recurring costs of meetings and transportation slowed the process. It was felt that this could eventually affect peoples interests in the coalition. Community Associations The level of knowledge regarding the GPF varied very much among associations. While some of them had only vaguely heard of it, others felt the idea was their own. In general, even in those cases where I had to explain the idea to the interviewee, the reaction was very positive. When asked to explain their positive view of this idea most of them cited their certainty about the timber activitys overall success: a madeira vai dar certo The timber will be a success! This perception was based on the high value of the resource and the attention timber was getting from the state government and the NGOs. Community association leaders trusted that the GPF would solve the marketing problems that the projects had experienced. Most were clear about the potential benefits, but not the eventual costs (time or money) the coalition represented for them. When confronted with the fact that the GPF would place a demand on their time and resources, they were more reserved about its usefulness. They hoped to see the GPF as a very effective marketing arm of the

PAGE 163

151 community timber enterprises. Although they all wanted to remain directly involved in decision-making, there was no clear idea as to how much time or other resources it would demand from them and their associations. External Stakeholders The set of stakeholders external to the community timber projects in Acre and Rondnia included both government and NGO staff 33 However their views of the GPF initiative were fairly similar. There was a general skepticism about the capacity of the communities to control the coalition and also about the GPF as the medium to transform the projects into business enterprises. For the GPF to be successful, it would have to be financially self-sufficient and capable of providing the technical services that were currently offered by the partner organizations to the projects. For the external partners, the GPF coalition might prove to be too demanding on the already over-extended association leaders. Their participation would be weak and would be compensated for by an overly active coordinator controlled by one (or several) of the partner organizations. Since participation of association leaders in the decision-making of the projects was already limited, it was hard to think that it would be any different at the coalition level. The perception was that changes were needed, but that they should occur at the project-level: treating the projects as business enterprises that would require financial sustainability and building stronger local accountability of the association leaders. In spite of the general skepticism, external partners 33 I interviewed personnel from SEFE, WWF and also researchers from UFAC, the Federal University of Acre.

PAGE 164

152 expected the GPF to have some positive results, although for them these were still unclear and would be very different to the expectations of the coalition proponents. Perception Matrix The table below presents a summary of the views and perceptions from the different stakeholders. Table 5.1: GPF Perception Matrix Perception aspect: Partner Organization Community Association External Partners Definition GPF is a funding strategy to attract donors for needed aspects such as marketing development and organizational capacity building. It is a platform to foster learning across projects. It is a channel to offer certified community timber to certified wood buyers. The GPF is the group of all the communities involved in timber management in Acre and Rondnia. The GPF is a space for exchange of ideas and experiences among the communities The GPF could be a good platform for cross learning among projects. It runs the risk of representing just another way for NGOs and other agencies to expand their control Vision It will free the projects resources for use on the technical aspects such as ecological monitoring It will became an effective commercializing mechanism for community projects The GPF will become a permanent solution for the marketing and commercialization component of the project. The GPF will succeed because for them timber is a very secure product Success is unlikely. The only chance of success for the GPF would come from achieving financial self-sufficiency and striving to raise local control over decision making Challenges The main challenge is the standardization of the timber products offered by the projects so that they can be offered jointly. Finding financial support to start and maintain the coalition A current problem is that by 2002 not all association leaders, much less the members, understood either the idea of the GPF, or its objectives and activities. The lack of awareness regarding the time (and other resources) that will be required by this new structure The GPF would only increase dependency as the structure is too heavy for the communities to take control of it The changes must occur at the project level, raising local accountability of the associations and transforming the projects into business enterprises. I have divided their perceptions into three elements: their definition or concept of what the GPF was or could be; their vision and expectation of what it

PAGE 165

153 could achieve for them; and finally, the challenges that they believed were addressed by the coalition proposal. I will analyze the differences and similarities below. In general the concept of what the GPF should be was very clear across the stakeholders: a coalition of community associations to foster cross learning and to market the timber products jointly. The perceived benefits varied among the stakeholders between the more immediate interest in the GPF becoming an effective commercializing venue and the goal of fostering autonomy among the involved associations. While the GPF would later in 2004 (see next section) be able to make its first joint sale and had worked on organizational capacity building, results on this last issue would take a longer time to emerge. Regarding challenges, the most significant thing was the critical perspective from some of the external partners about the inadequate timing for establishing the GPF. For this group, the GPF needed the projects to be transformed into community enterprises first before a coalition of community timber management could develop. They did not see the GPF as a vehicle for this transformation. In contrast, both the partner organizations and the community associations trusted the idea of the GPF and instead identified other challenges such as the cost of participation or the difficulties in standardizing products and processes across the projects. Wrapping up this section on perceptions, I would like to point to the timing and significance of the GPF for all stakeholders. The issues of marketing and capacity building were acknowledged as serious threats to the viability of the

PAGE 166

154 community timber projects. Also, the idea of coming together to address these problems was well accepted and promising to all parties. In general it can be said that the idea was perceived as a good idea at a right time. The different emphases of the stakeholders seemed to complement each other, rather than potentially pulling the process in different directions. Most of the differences were related to the operationalization of the idea. While formally the POs presented the coalition as CAs initiative, CAs did not seem to be aware of this fact. In particular, they did not seem to understand the costs that leading and running the coalition would present. They all seemed to trust that the POs would cover those costs, just as they had in the timber projects. External agents seemed to be well aware of this problem. Although they pointed out that it might reinforce the dependency of the CAs on the POs, they thought the GPF was a good idea, capable of rendering positive impacts. Process Design In this section I study the evolution that the GPF went through from its initial formulation in 2001 (presented above) to its 2002 functioning structure. The information in this section is based on the interviews conducted in the summer of 2002 and my participant observation of GPF planning meetings during that period. I complemented this information with later interviews and a review of CTA documents on the GPF progress, after 2002.

PAGE 167

155 Innovation Innovation processes are described by Douthwaite (Douthwaite, 2002) as a series of interactive exchanges between proponents and users of a technology 34 In this process the initial invention presented by the proponents will be transformed by the users as they apply it to their own activities. During this process, the participation of the proponents will gradually diminish while the users will increase. For a period of time in the middle of the process they will interact in a fairly balanced way. The process is divided in four stages: development, start up, adaptation and expansion 35 Douthwaite saw the initial invention as an unfinished product that gets completed through the participation of the users. That participation guarantees the products social validity and dissemination. The GPF Timetable To analyze the GPF innovation process using Douthwaites model, I first developed a timetable of events based on the history of the GPF. This is presented in Table 5.2. This timetable summarizes the sequence of events in the development of the GPF initiative. 34 Douthwaite looks at farming technology. He mainly uses the words researchers and farmers that I have adapted to proponents and users to analyze the GPF innovation process. I extrapolate certain elements of Douthwaites model; however there is a broad difference between the technological innovations he studies and the type of social innovation of the GPF coalition. Among other differences; GPF is not an object (tool or equipment) being tested individually nor does it involve licensing or patenting. 35 Development is the initial stage when the proponents prepare the invention proposal. Start up is early sharing of the invention among a small group of users. Adaptation is the middle process where the interaction among proponents and users is more balanced and the product (idea or invention) is closer to being completed. Expansion is the last phase were the idea is scaled up for application by a larger number of users.

PAGE 168

156 Table 5.2: GPF Coalition Timetable Dates Events 1995-96 Beginning of timber projects in Porto Dias, Cautrio, Peixoto 1998 Projects put on stand-by because small scale licensing was not available from IBAMA Initial working group meeting for IEB-WWF-IMAZON supported community timber projects 1999 Start of timber projects in Cachoeira Oct: second community timber working group meeting supported by IEB-WWF-IMAZON 2000 Simplified small scale licensing available from IBAMA July: third community timber working group meeting supported by IEB-WWF-IMAZON First harvest for Cautrio and Pto. Dias In So Paulo, the NGO Friends of the Earth-Brazil organizes a coalition of certified wood buyers known as the Grupo de Compradores (Buyers Group). 2001 Cross-visits among the different community timber projects in Acre and Rondnia Visit to the forest ejidos in Southern Mexico Projects experience serious marketing problems; most are not making a profit to distribute among participants Dec: First community timber projects meeting in Acre; the GPF coalition idea is presented 2002 Cachoeira is Certified by FSC Mar: formal creation of the GPF coalition Jun: a consultant is hired to begin working towards the creation of the GPF. The goal was to analyze the operational process of each project to identify the issues facilitating and challenging the proposal to jointly offer the products from the different timber projects. Reelection of the Governo da Floresta (Acre state government) Nov: Porto Dias is Certified by FSC It shows when the proposal for the coalition came from an NGO and was presented at a meeting of community forestry projects. The context of this presentation was the marketing problems experienced by the timber projects. Parallel to the GPF initiative was the certification of the timber projects by a joint effort of the POs and the WWF. Another important event adding to the context was the reelection of Jorge Viana as governor of Acre. Vianas Governo da Floresta has shown a close interest in the GPF initiative. In the two-year period since this study was conducted the GPF has evolved in a very positive manner. In 2003 a new coordinator was hired for the GPF,

PAGE 169

157 based in CTA and supported by WWF-Brazil. Three new community projects were incorporated into the GPF, while Cautrio from Rondnia withdrew. Most of these new projects were initiatives supported by the Acre state government. In May 2004, the first joint sale of certified timber by the GPF coalition was made to a group of furniture industries from the south of Brazil. Also in 2004, another five communities became part of the GPF coalition, raising the total to eleven. All the new groups (including those added in 2003) were starting from scratch. None had any experience producing or selling timber. Innovation Process The significant thing about the initial proposal of the GPF was that it represented the final goal or perfect scenario. It showed what the proponents wanted to see in terms of values and structure. The ideal GPF would be made up exclusively of community associations, which would make decisions and direct the coordinators work, while the supporting organizations would remain outside. Local participation and control as well as collaboration were essential working principles in this ideal. The plan for how to get to that ideal GPF was never presented or developed. Planning the process was not a priority for the proponents for several reasons. First, for them, the GPF was responding to a more immediate need (the need to market timber) and second, in the NGO project culture, ideas are generally not developed thoroughly unless funding for them is secured. Planning costs are too high for NGOs that are normally under-staffed and under-funded. Planning is normally done on demand and according to the amount of funding expected. The NGO that proposed the GPF developed an initial two-page

PAGE 170

158 document and circulated it among their particular partner organizations and donor officials. The idea was moved forward once interest from WWF 36 was established. The level of support by WWF was small, just enough for the March 2002 meeting and to support a coordinator to start up the group. The GPF idea did not come as a result of discussion among participants in the community timber projects of Acre and Rondnia. It was presented to them as a solution to their marketing problems. The emphasis was placed on the solution it would provide not on the collaborative way it was proposing to do it. Just as the timber projects had been external initiatives into which the communities were brought, so was this new coalition idea. Following Douthwaite (2002), the development phase of the GPF was constrained by the typical limitations of NGO proposal development (lack of planning, urgency, donor driven priorities). The community timber producers meeting that formally and publicly established the GPF, in March 2002, launched the start up phase. Community associations became part of the GPF not as active co-creators but as endorsers of yet another NGO idea, very similar to the way they were brought into the timber projects. From March 2002 to late 2003 when the GPF made its first joint sale of certified wood, the GPF went through numerous changes with little to no community leadership or participation. Right after the GPF was established the discussion among proponents turned to the formalization of the GPF. The issue was whether it should be 36 As was mentioned earlier, WWFs interest in the GPF was motivated by the potential it had to consolidate the certified wood system they had been promoting. Given the small scale of the certified community projects, the way to attract certified-wood buyers was to offer more wood volume by aggregating the projects production. The GPF offered a venue to achieve this.

PAGE 171

159 structured as a cooperative (as had originally been thought), a community owned company, a producers union, or a federation. The issue became urgent as it was hoped that the new organization (whatever form it took) would be able to take care of the 2002 harvest. The formalization was required because part of the marketing problem in 2001 was caused by the associations legal prohibition on selling products directly. The discussion carried on until it was clear that nothing could be done for that years harvest. The issue of formalizing the GPF has not re-emerged. The second half of 2002 channeled WWF financial support to CTA. A GPF coordinator was hired, with a list of responsibilities that went from evaluating the production process in each project to assessing the organizational capacity of each community association and identifying potential markets for the 2002 production. The first person hired for the job was an outsider to the process and the region, who left after five months of work. In 2003 a new coordinator was hired who remained in the position from then on 37 This person was from Acre, and had plenty of experience in community development. He was able to consolidate the joint sale of the 2003 harvest and kept the GPF group of community participants meeting every month. The group grew from the initial four community groups to eleven, and only dropped one group, Cautrio from Rondnia. This group was originally included as part of the negotiation with WWF, although because of the distance to the rest of the projects it was clear that it would be difficult to stay involved in the GPF activities. 37 The information post-2002 is based on documents and interviews conducted with CTA staff.

PAGE 172

160 The GPF functioning structure that resulted from CTA work and WWF financial support is presented below in Figure 5.2. This diagram contrasts with that of Figure 5.1, which represented the GPF-as-planned. This one represents the GPF-in-the-making, the one actually resulting from the innovation process. This diagram reflects the central role played by CTA, and the leadership and control the coordinator had over the committee of local participants. A.Pt. DiasCTAEMBRAPAECOPORESEATERAPRUMAAGUAPEAMPPAE-CM Committee Coordinator Timber buyers New Markets SEFE new membernew memberGPF Coalition WWF Figure 5.2: GPF Functional Structure The GPF-as-planned and the GPF-in-the-making are not linked in an explicit manner. Since no roadmap was developed on how to get to that ideal GPF-as-planned, it is hard to determine if the current process-structure is taking the process in the direction of the ideal one. The duality between reality and discourse tends to create confusion and to project an image of manipulation of the communities by the NGOs. The underlying perception is that an explicit recognition of the gap would not be favorable for the image of either the GPF or the NGO. An image of a coalition that is internally strong and led by the local

PAGE 173

161 participants seems to fit the expectations the NGO has of its own work and the one accepted by other partners and founders. Whether the current functioning-GPF will evolve into the ideal-GPF remains to be seen. What is clear is that the NGO project culture (and the NGO agenda) affected its development during the start up phase, just as it did during the development phase (Douthwaite, 2002). One of the outcomes of this was that the innovation process moved at the pace of the partner organizations, limiting the CAs effective participation, as many decisions were taken without their input. If the initial proposal was that the GPF would be a coalition led by the community timber associations, these associations input to the functioning coalition had been meager. The consolidation of the first joint-sale of timber marked a step ahead for the GPF and placed the process in the adaptation phase (Douthwaite, 2002). The new coordinator had managed to consolidate a core group of local participants that met every month. The content of these meetings ranged from training in interpersonal communication to forest certification or setting prices for their products. I expected invigorating participation to have an effect on the design and functioning of the GPF, but further research would be required to follow up on this. In conclusion, since explicitly designing the GPF innovation process was not a priority for the proponents, a plan for how to get to that ideal GPF was never developed. The resulting functioning GPF was loosely linked to the ideal GPF model. It was hard to determine whether the real GPF in-the-making

PAGE 174

162 might lead to the ideal one. Given that the initial proposal was that the GPF would be a coalition led by community timber associations, the functioning coalitions falls short of that participatory vision. The GPF had remained an NGO initiative where the CAs participated but did not lead the process. Synthesis Stakeholders positively received the initial proposal for the GPF coalition to act as a marketing center. That proposal included the vision of the GPF as a horizontal and independent organization of CAs with support and resources coming from the POs and the external stakeholders. This ideal conceptual design was never backed by a planned process design, because in the NGO project culture, process design depends on the available funding. The premise within the NGO culture is that once a source of funding is close to being secured, a process design is developed (in the form of a project proposal) to meet the conditions of the potential donor. By not committing to a concrete process design, the promoters allow themselves the freedom to adapt the idea to different donors. Because major funding never materialized, a plan for how to get to the ideal GPF was never developed or promoted among the parties involved. The resulting functioning GPF was loosely linked to the ideal GPF-as-planned. In fact, the GPF idea was presented and promoted by an NGO that rallied for its support, among other POs and CAs, without handing over control of the idea to the other participants. Nor were they forced by the other participants to hand over control. This lack of collaboration in the creation of the GPF made the coalition look more like a typical NGO project and less like a collective action initiative, with only limited input by the CAs to the current functioning coalition. At

PAGE 175

163 best they were clients of the initiative; they did not behave as true coalition partners. They had a partial understanding of the benefits of the GPF, but no idea of the costs that this new second-degree organization would represent to them. On the other hand, external agents pressure had influenced the participation and timetable of the GPF by bringing in new communities to the group and selecting the training topics (financing only some activities). I suspect this pressure also reinforced the paternalistic collaboration pattern between the PO and CAs, because in order to show short-term results, the POs would expedite decisions that would have taken more time in the hands of the CAs. Examples of such decisions were the choice of a name for the GPF, the hiring of the GPF coordinator, and the decisions about who got invited to join the group. This was not a case of manipulation by an NGO, but a survival strategy that allowed the promoters of the coalition to match donors and activities, keeping the process going. This was primarily a structural problem caused by the established project-financing culture among development and NGOs and public and private donor agencies. In the next chapter I will integrate the analysis of the design component with the other two (internal, external factors) to assess the potential of the GPF coalition proposal for success, and to determine whether my hypothesis regarding the preeminence of the internal factors (organizational development, in particular) could be validated by the studys results

PAGE 176

CHAPTER 6 CONCLUSIONS This research aimed to contribute towards defining how communities could be linked to markets on their own terms. I argued that the development of a second-degree organization, such as the coalition, represented a self-organized response to deal with a rise in social complexity. Part of this complexity came from efforts to build a differentiated market relationship through forest certification and the connection to green markets. The experiences of Mexicos Yucatan Peninsula, Guatemala and Nepal strongly suggested that community forestry initiatives highly benefited from a second-degree organization that positively reinforced its base-group members. These consortium-like platforms provided a variety of services ranging from conflict resolution to product marketing, enabling members to consolidate their activity and sustain a long-term learning process that would transform community-market relationships. The GPF coalition of community timber projects from Acre and Rondnia left little evidence to confirm if coalitions were a secure structure for equitable market integration. The GPF was still at a very early stage compared to the cases of those other regions. My work contributed to developing instruments to analyze this early stage of collaborative organizations. It showed the existing gap 164

PAGE 177

165 between the desired GPF and the real one. It pointed to some of the causes of this gap and to the need to re-adjust the role that social learning played in building an ideal autonomous alliance of community timber enterprises. I begin by retrieving my hypothesis and assessing it with regard to the results presented in Chapters Three to Five. In the following section, I describe the degree to which the hypothesis was verified by the results, and also the limitations of this hypothesis to explain the complexities of the GPF coalition. I then reflect on the methods used, particularly the analytical framework, considering its usefulness in retrieving and analyzing data. I include a section (second to last) that looks beyond my hypothesis at some of the key issues emerging from my research framework. The final section is devoted to my final comments on and recommendations for the GPFs future process. The organizing element in this thesis is the three-tired analytical framework. This framework divides the factors conditioning the success of the GPF in three categories: internal, external and design. I hypothesized that while the external and design factors would prove to be positive to the success of the GPF, the internal factors would be the limiting factor. Within this last category, the limited management capacity of the community associations would be the main constraint for the GPF. Results as Related to the Hypothesis In the hypothesis I had stated that the internal and external factors would be positive for the GPF while the internal factors would be the constraining element. As it turned out internal, external and design factors affecting the GPF influenced each other within a complex system of relations, making it hard to establish the

PAGE 178

166 individual influence of each factor. I present the results using a matrix developed in Chapter Two that I have now completed (Table 6.1). This table has three columns: a list of variables per factor; a broad list of ideal conditions or benchmarks for coalition processes; and the main results (the nuggets!) from each research component. I continue then to discuss vertical and horizontal relationships within this matrix. Table 6.1: Integrating Matrix with Results Category Factors (Variables) Ideal Condition or Benchmark Case Studied At the Group level: o Participation o Leadership o Management Capacity Need for a coalition should be explicit among group members Time should be available to participate in coalition building Trust should be strong among members and partners. Coalition responded to immediate needs (built market alternatives) CAs had very little management capacity POs strongly controlled the projects. Internal: At the Project Level: Partner relationships Capacity development Appropriation community Projects should foster social learning Projects should encourage appropriation by the community Conflict management capacity should be a developed skill Projects had a paternalistic pattern of collaboration between POs & CAs Projects had a technical bias -concentrated on forestry issues and leaving out the business dimension of the timber activity Social learning was only beginning to get attention in some projects. External: State and federal government agencies, policies and laws on conservation, forestry development and social organization NGO and Aid agencies Forest products markets Laws and policies should secure land and resource use rights for the communities Credit and technical support should be available for the community enterprises Stable long term relations should exists with between communities and NGOs and government agencies Legal, policy and institutional environments for the GPF were very positive. As a marketing tool certification was an innovation, not a proven tool. Strong mutual dependency between higher scale innovation (certification, RESEX) and the lower scale (GPF). Coalition Design: Objectives Structure Leadership Resources Structural design should be flexible and open to change Initial proposal should provide room for adaptation by the potential members Concept of the GPF well accepted by parties: an independent, horizontal group of CAs focused on marketing CAs timber products. Initial process design lacking. Design: Innovation Process History Perceptions Adaptations Process design should provide room for adaptation by the potential members Process should have an explicit learning and reflection strategy with built-in mechanisms for monitoring and evaluation The real GPF differed from the conceptual one The real one resembled more a NGO-project than a collective action initiative As in a typical project, CAs behaved like clients, not as partners. External demand caused more attention to be given to achieving results than to the CAs social learning process.

PAGE 179

167 From this matrix we can point to the following key findings: The GPF coalition as an idea remains very attractive, because it seems an adequate intermediate mechanism in which to nest the community timber projects. It was expected that the coalition would strengthen the associations bargaining power with market and political agents. However, as predicted in my original hypothesis, internal factors, in particular the management capacity of the community associations participating in the GPF, hindered the chances of realizing the coalitions potential. My hypothesis fell short of estimating the multidimensionality of the GPF problems. Factors from both the external and internal categories constrained the process, too. In its current trajectory the GPF failed to break from the pattern of dependent-development that also affected community timber projects. Instead of moving out of the projects to let the community manage, partner organizations were perpetuating their role. The GPF was also a partner-controlled initiative. The existing external environment reinforced this dependency because agents such as government and donors had a large stake in the community timber projects. Larger scale innovations like sustainable forest management, tropical forest certification and, the governo da forestas neo-extractivist policies were themselves dependent on the success of the community timber projects and the GPF. This set of larger scale innovations, because of their own early stage of development, were very sensitive to the small-scale feedback of initiatives like the timber projects in Acre. For these reasons, the large-scale actors promoting the broader innovations intervened strongly in the community projects and the GPF. This intervention resulted in an emphasis on obtaining short-term quantifiable results, dismissing experimentation and social learning. These last aspects were key to raise the associations organizational capacity, but too long-term for most of the external partners agendas. Social learning (as well as experimentation) still happened during the projects implementation. However, its effects were hard to predict when such learning was not explicitly managed (monitored and evaluated). Social learning might lead to empowered communities that would turn to conflict as the only way to manifest such gained agency if the projects staff remained unaware of the social learning process. The data analyzed on the GPF process and the preceding community timber projects suggests this could potentially happen. Hypothesis The results seemed to support the hypothesis: both external and design factors were broadly positive for the successful development of the GPF, while

PAGE 180

168 the internal factors painted a more difficult picture. However, a closer analysis showed a much more complex picture. The various factors (internal, external and design) proved to be more interconnected than I expected, making it hard to isolate the effect of any individual one. In addition, the analytical framework was built based on NRM coalitions, which the proposed GPF model was expected to follow. The GPF seemed to be following more of an NGO project pattern, which might in fact be more appropriate for the internal and external circumstances of the parties involved in western Amazonia. Methods The framework proved valuable to organize the sources of data, both archival information and interviews. The framework was weak to help elucidate the cross-factor and cross-scale relations. It was adequate to compartmentalize but it did not facilitate integration. As a result, the main insights about the situation seemed to come from intuitively looking at the whole picture canvassed by the framework. It was difficult to discern whether insights were a result of the analysis or of the experience as a whole. Therefore, in trying to keep conclusions tight within the reach of the framework, many ideas generated in the process were left out. Emerging Issues This section presents a series of specific issues that responded to problems visible through the use of the analytical framework, but not adequately addressed by it.

PAGE 181

169 Economic and Cultural Transition It was pointed out, especially by external stakeholders, that there was a large cultural gap between the livelihood strategies of the community participants and those of the timber projects. This was largely because timber was a highly time-intensive industrial activity and because the projects were focused solely on economic benefits. The projects were also initially rejected in some communities because logging was seen as contradictory to the rubber-tappers struggle to save the forest. Early project evaluations pointed to the communities lack of organizational capacity and business skills needed to manage the timber activities. The timber projects tried to address these issues through different strategies: for example, the projects relied on formal training of community participants in order for them to take over the timber operation. While results from this study provide evidence of the above-mentioned cultural gap, it needs to be considered that the general situation of the rubber tappers demanded a substantial change in their livelihood strategies. The rubber-tapper communities were pressured to look for new economic alternatives, both for positive (creation of RESEX) and negative (collapse of rubber markets) reasons. In this context, the typical strategy considered by most individual rubber tappers was to leave the forest and move to the city, or raise cattle. These options was certainly more dramatic than the one represented by the timber projects. The projects cultural and social gaps had more to do with their implementation and less with their conceptualization. To begin with, the urgent need to show profits overshadowed the learning curve that was necessary before

PAGE 182

170 the local participants could adapt to the new activity. Secondly, the projects overlooked the organizational conditions of the communities, and assumed that the associations would be capable of taking on project tasks. Thirdly, business skills were neither assessed nor developed until after an almost general marketing crisis. As a result of these social and cultural gaps the projects have failed to appreciate what they have accomplished in terms of social organization and cultural identity building. The associations were significantly stronger after their participation in the projects than they had been before. When compared to associations in other RESEX and PAEs, the timber community associations stood among the strongest and more consolidated (Stone, 2003). Yet the projects had not measured this capacity development and therefore could not claim it as a direct result of the projects. In cultural terms, the word manejador 38 represented a forest identity-in-the-making (Cunha, 2002). This new identity could be interpreted as a challenge to the traditional rubber-tapper identity but also as a representation of a diversified-extractivist, somehow derived from Regos neo-extractivism (Kainer et al., 2003; Rgo, 1999). The cultural change represented by the timber projects could be interpreted not as opposite to the traditional rubber tapper culture, but as a way to build resilience in these communities. The timber projects provided tools to cope with a changing economic and political context. The timber projects were still at a pilot phase and their impact, both socially and economically, was 38 Manejador is the word used by the CA participants in the project to define themselves.

PAGE 183

171 very small. In fact there was the potential risk that if external funding ended, the projects would stop and community members would have to go back to their traditional survival strategies such as leaving the forest and moving to the city, or deforesting for cattle production. Community Enterprises The ideal situation would be to have the timber projects entirely managed and run by the community associations. To date the timber projects had been designed and implemented with other more immediate goals, such as training the participants in timber operation skills, or generating profits to directly benefit the families involved. Moving towards the community-business ideal seemed to require a reformulation of the projects, because the project staff found themselves overwhelmed by the technical and ecological aspects of the activity. The GPF coalition represented an effective way to work on the business dimension of timber. As was pointed out by some of the interviewees, the problem with the GPF was that it had been developed with the same project culture principles as the original community timber projects. Management, planning and design remained in control of the external partner. It seemed hard to think that more participation would result from using a template that had shown itself to be limited for these purposes. It seemed equally difficult to be promoting a business mentality through another project that would rely heavily on donor funds. In the same way that the timber projects had been expected to deliver more than they could (in terms of creating community businesses), the GPF seemed likely to run into the same problem.

PAGE 184

172 An important issue hardly addressed by the timber projects (and further from the reach of the GPF), was the accountability of local leaders. The associations represented a middle level organization immediately above the individual households. Most of the associations in this study were created by external initiatives, and membership corresponded to friendship and kinship. These local organizations had had little to no change in their leadership positions since their creation. Internal community conflicts were not always aired at the association meetings, and in some cases these led to community divisions and the creation of new associations. Project staff tended to interact strongly with very few community members (typically the leaders). The personal relationships built between project staff and community leaders could turn from an asset into a liability, should conflicts arise. In order to both increase participation and develop a business mentality in the timber management, local accountability had to be encouraged and fostered. This is turn required (among other things) that certain types and degrees of internal conflicts be openly dealt with and transformed into learning and reflection opportunities that consolidated internal democracy. Chapter 3 presented two examples of social learning processes that turned had turn into conflicts: first, the changes in the use of the colocaes as the management unit of the timber project. Second, the outsourcing of the transportation and other aspects of the timber operation. These two adaptations

PAGE 185

173 of the original plans could have been less costly to the projects if social learning had been an explicit and central element of the management strategy. Future of the GPF I will use these final paragraphs to comment briefly on the future of the GPF coalition. I will consider two aspects of the GPF coalition: its conceptualization and the design of the process. The gap between the GPF-on paper and the GPF-in-the-making is more than a difference between ends and means. The real GPF (GPF-in-the-making) responded to a different concept than that of the ideal GPF (the one on paper). The ideal was an alliance, a horizontal platform to share resources and take on collective challenges. The real GPF was an NGO project that brought together the group of manejadores from a set of different projects, and which served as a training and cross-learning platform. The chances of the current GPF leading to the ideal one were uncertain. To begin with, there was no process design for such a goal. Secondly, it was unclear whether the ideal was even desired by the local participants. It is clear from this study that their organizational capacity was very limited and probably not sufficient to lead to a bottom-up formal network. Third, there was the risk that active coordination and management by one of the partner organizations would create a dependent development. GPF would consolidate and even grow as long as the NGO that currently coordinated it played that role. This seemed to be the situation with the timber projects, whose transformation into community timber enterprises was at best uncertain.

PAGE 186

174 Another reason why the transformation of the GPF from an NGO-project to a coalition was uncertain was the lack of process planning and documentation. There was no roadmap for turning the current process into an effective coalition. There were no parameters to measure whether activities were leading in that direction or not. An important element for establishing such parameters would be to turn the GPF from its current marketing focus to a learning focus. This would require being aware of the structural and systemic constraints to social learning, such as the NGO-project culture or the issue of dependent development. Nevertheless, as stated by one of the CAs leaders a madeira vai dar certo. The enthusiasm and commitment of the GPF participants would go a long way towards finding the path to consolidate the GPF coalition as an effective organization for community timber management in western Amazonia.

PAGE 187

LIST OF REFERENCES Agrawal, A. (2001). Common property institutions and sustainable governance of resources. World Development, 29(10), 1649-1672. Alcorn, J. B., & Royo, A. G. (1999). Indigenous social movements and ecological resilience: Lessons from the dayak of indonesia.Unpublished manuscript, Washington D.C. Alcorn, J. B., & Toledo, V. M. (1998). Resilient resource management in mexico's forest ecosystems: The contribution of property rights. In F. Berkes & C. Folke (Eds.), Linking social and ecological systems, management practices and social mechanisms for buiding resilience (pp. 216-249). Cambridge: Cambridge University Press. Allen, T. F. H. (2004). A summary of the principles of hierarchy theory.Unpublished manuscript, Madison. Amaral, P., & Amaral, M. (2000). Manejo florestal comunitario na amazonia brasilera, situacao atual, desafios e perspectivas. Brasilia: IIEB. Amaral, P. H. C. (2001). Evaluacion de las condiciones, procesos y resultados del manejo forestal comunitario en la amazonia brasilea. Centro Agronomico Tropical de Investigacion y Enseanza, Turrialba, Costa Rica. Anderson, A., & Clay, J. (2001). Greening the amazon: Communities and corporations in search of sustainable business practices. Washington D.C.: Ford Foundation. Anderson, L. (1994). The political ecology of the modern peasant: Calculation and community. Baltimore: John Hopkins University Press. Babbington, A. J., & Carroll, T. F. (2003). Induced social capital and federations of the rural poor in the andes. In C. Grootaert & T. v. Bastelaer (Eds.), The role of social capital in development (pp. 234-278). Cambridge: Cambridge University Press. Baland, J.-M., Bardhan, P., & Bowles, S. (2002). Inequality, collective action and environmental sustainability. Los Alamos: Santa Fe Institute. 175

PAGE 188

176 Baland, J.-M., & Platteau, J.-P. (2002). Collective action on the commons: The role of inequality. In J.-M. Baland, P. Bardhan & S. Bowles (Eds.), Inequality, collective action and environmental sustainability. Los Alamos: Santa Fe Institute. Borrini-Feyerabend, G., Farvar, M. T., Solis, V., & Govan, H. (2001). Manejo conjunto de los recursos naturales, organizarse negociar y aprender en la accion. Heidelberg: GTZ-IUCN. Bray, D. B., Merino-Perez, L., Negreros-Castillo, P., Segura-Warnholtz, G., Torres-Rojo, J. M., & F.M.Vester, H. (2003). Mexico's community-managed forest as a global model for sustainable landscapes. Conservation Biology, 17(3), 672-677. Brick, P., Snow, D., & Wetering, S. V. d. (2001). Across the great divide, explorations in collaborative conservation and the american west. Washington DC: Island Press. Brown, K., & Rosendo, S. (2000). Environmentalist, rubber tappers and empowerment: The politics and economics of extractive reserves. Development and Change, 31, 201-227. Bryant, R. L. (1991). Putting politics first: The political ecology of sustainable development. Global Ecology and Biogeography Letters, 6, 164-166. Buck, L. E., Wollenberg, E., & Edmunds, D. (2002). Social learning in the collaborative management of community forestry: Lessons from the field. In E. Wollenberg, D. Edmunds, L. Buch, J. Fox & S. Brodt (Eds.), Social learning in community forestry (pp. 1-20). Bogor Indonesia: Center for International Forestry Research (CIFOR). Buckles, D. (1999). Cultivating peace, conflict and collaboration in natural resource management. Washington D.C.: World Bank, IDRC. Capossoli, M. J. (1999). Manejo florestal comunitario, segunda oficina. Maraba: Amigos da Terra WWF. Cardoso, C. A. S. (2002). Extractive reserves in brazilian amazonia. Hamshire: Ashgate Publishing Co. Clay, J. (2001). Community-based natural resource management within the new global economy, challenges and opportunities. Washington D.C.: Ford Foundation. Colchester, M., Apte, T., Laforge, M., Mandondo, A., & Pathak, N. (2003). Bridging the gap: Communities, forest and international networks. Jakarta: Center for International Forestry Research.

PAGE 189

177 Coughlin, C., Hoben, M., & Manskopf, D. (1999). A systematic assessment of collaborative resource management partnerships. University of Michigan, Ann Arbor. CTA. (2001). Central de associaes de produtores florestais do acre (pp. 3). Rio Branco: Central de Trabalhiadores da Amazonia. Cunha, M. d. S. (2002). Adaptive co-management, a case study: Pae porto dias, acre, brazil. Rio Branco: CIFOR-University of Florida. D'Arcy, D.-C. (2001). The reflective practitioners: Learning and teaching in community-based forest management. Conservation Ecology, 5(2), 15. Dasgupta, P. (2002). Social capital and economic performance: Analytics. In E. Ostrom & Toh-Kyeong (Eds.), Social capital: A reader (pp. 41). London: Edward Elgar. Devine, R., Hitz-Sanchez, A., Keenan, J., Leon, P., MacLeod, P., McGean, B., et al. (2001). Institutional. Washington DC: The Nature Conservancy. Douthwaite, B. (2002). Enabling innovation, a practical guide to understanding and fostering technological change. London: Zed Books. Douthwaite, B., Delve, R., Ekboir, J., & Twomlow, S. (2003). Contending with complexity: The role of evalutation in implementing sustainable natural resource management. International Journal of Agricultural Sustainability, Vol 1(No. 1), 51-66. Douthwaite, B., Haan, N. C. d., Manyong, V., & Keatinge, D. (2001). Blending "hard" and "soft" science: The "follow-the-technology" approach to catalyzing and evaluating technology change. Conservation Ecology, 5(2), 13. Edmunds, D., & Wollenberg, E. (2001). A strategic approach to multistakeholder negotiations. Development and Change, 26, 231-253. Edwards, K., Hideyuki, K., & Veer, C. (2001, April 9-12 2001). Facilitating support networks for community forestry development. Paper presented at the Training Workshop on Facilitating support networks for community forestry development, Bangkok, Thailand. Escobar, A. (1999). After nature: Steps to an antiessentialist political ecology. Current Anthropology, 40(1), 1-30. Fisher, L., Russell, V., & Ericson, J.. Coalition building for conservation: Latin american multi-stakeholder partnership. Paper presented at the XXIII International Congress of the Latin American Studies Association, Washington D.C. (2001, September 6-8, 2001)

PAGE 190

178 Folke, C., Carpenter, S., Elmqvist, T., Gunderson, L., Holling, C. S., Walker, B., et al. (2002). Resilience and sustainable development: Building adaptive capacity in a world of tranformation. Stockholm: Ministry of the Environment Sweden. Gomez, C. V. A. (2001). Dynamics of land in an amazonian extractive reserve: The case of the chico mendes extractive reserve in acre, brazil. University of Florida, Gainesville. Gunderson, L. H. (2003). Adaptive dancing: Interactions between social resilience and ecological crises. In F. Berkes, J. Colding & C. Folke (Eds.), Navigating social-ecological systems. Cambridge: Cambridge University Press. Holling, C. S. (1978). Adaptive environmental assessement and management. New York: John Wiley & Sons. Holling, C. S., Berkes, F., & Folke, C. (1998). Science, sustainability and resource management. In F. Berkes & C. Folke (Eds.), Linking social and ecological systems, management practices and social mechanisms for building resilience (pp. 343-362). Cambridge: Cambridge University Press. Holling, C. S., & Gunderson, L. H. (2002). Resilience and adaptive cycles. In C. S. Holling & L. H. Gunderson (Eds.), Panarchy, understanding transformations in human and natural systems. Washington DC: Island Press. Holling, C. S., & Meffe, G. K. (1996). Command and control and the pathology of natural resource management. Conservation Biology, 10(2), 328-337. Kainer, K. A., Schmink, M., Pinheiro, A. C., & Silva, M. J. D. (2003). Experiments in forest-based development in western amazonia. Society and Natural Resources, 16, 869-886. Kenney, D. S. (2000). Arguing about consensus, examining the case against western watershed initiatives and other collaborative groups active in natural resources management. Boulder: University of Colorado School of Law. Kopelman, S., Weber, J. M., & Messick, D. (2001). Factors influencing cooperation in commons dilemmas: A review of experimental psychological research. In E. Ostrom, T. Dietz, N. Dolsak, P. C. Stern, S. Stonich & E. U. Weber (Eds.), The drama of the commons. Washington DC: National Academy Press.

PAGE 191

179 Kornexl, W. (1997). Avaliao economica e financiera independente do projeto de manejo florestal de uso mltiplo asentamento extrativista porto dias (ac). Rio Branco Brasilia: Centro Nacional de Desemvolvimiento Sustentado das Populaes Tradicionais CNPT/IBAMA. Krishna, A. (2003). Creating and harnessing social capital. In P. Dasgupta & I. Serageldin (Eds.), Social capital, a multifaceted perspective. Washington DC: World Bank. Lee, K. N. (1993). Compass and gyroscope, integrating science and politics for the environment. Washington DC: Island Press. Macedo, D. S. (2000). Manejo florestal comunitario, iii oficina de manejo florestal comunitario. Rio Branco-AC: PROMANEJO. Margoluis, R., Margoluis, C., Brandon, K., & Salafsky, N. (2000). In good company: Effective alliances for conservation. Washington DC: Biodiversity Support Program. Margoluis, R., & Salafsky, N. (1998). Measures of success, designing, managing and monitoring conservation and development projects. Washington DC: Island Press. Mayer, E. (2002). The articulated peasant, household economics in the andes. Boulder: Westview. McCay, B. J. (2000). Emergence of self-organized cooperation. Paper presented at the Constituting the Commons, Conference of the International Association for the Study of Common Property, Bloomington Indiana. Monge, L. (2001). Coaliciones para la conservacion, una herramienta para mejorar capacidades en el manejo de los recursos marino costeros. Guatemala: PROARCA COSTAS (TNC-URI/CRC-WWF). Nygren, A. (2000). Environmental narratives on protection and production: Nature-based conflicts in rio san juan, nicaragua. Development and Change, 31, 807-830. Ostrom, E. (1990). Governing the commons, the evolution of institutions for collective action. Cambridge: Cambridge University Press. Ostrom, E. (2003). How types of goods and property rights jointly affect collective action. Journal of Theoretical Politics, 15(3), 239-270. Peluso, N. (1992). Rich forest, poor people: Resource control and resistance in java. Berkeley: University of California Press.

PAGE 192

180 Pender, J., & Scherr, S. J. (2002). Organizational development and natural resource management: Evidence from central honduras. In R. Meinzen-Dick, A. Knox, F. Place & B. Swallow (Eds.), Innovation in natural resource management (pp. 207-239). Baltimore: John Hopkins University Press. Poats, S. V. (2001). El consorcio carchi: Un espacio para aprender y actuar en favor de al conservacion de los paramos del norte del ecuador.Unpublished manuscript, Quito. Powell, W. W. (1990). Neither market nor hierarchy: Network forms of organization. Research in Organizational Behavior, 12, 295-336. Rogers, E. M. (1995). Diffusion of innovations (fifth ed.). New York: Free Press. Rgo, J. F. d. (1999). Amaznia; do extractivismo ao neoextractivismo. Ciencia Hoje, 25, 62-65. Rosendo, S., & Brown, K. (1998). Strategic alliances, partnerships and collective action: Rubber tappers and extractives reserves in rondonia, brazil.Unpublished manuscript, Norwich. Royo, A. G. (1999). The power of networking: Building force to navigate cross-scale turbulance where solo efforts fail. In J. B. Alcorn (Ed.), Indigenous social movements and ecological resilience: Lessons from the dayak of indonesia. Washinton DC: Biodiversity Support Program. Salafsky, N., Margoluis, R., & Redford, K. (2001). Adaptive management: A tool for conservation practitioners. Washington DC: Biodiversity Support Program. Salisbury, D. S. (2002). Geography in the jungle: Investigating the utility of local knowledge for natural resource management in the western amazon. University of Florida, Gainesville. Santos, V., Carreon., M., & Nelson, K. C. (1998). La organizacion de la union de ejidos productores forestales de al zona maya. Mexico D.F.: Red de Gestion de Recursos Naturales y Fundacion Rockefeller. Scherr, S. J., White, A., & Kaimowitz, D. (2002). Making markets work for forest communities. Washington D.C.: Forest Trends-Center For International Forestry Research. Schmink, M. (2004). Communities, forest, markets & conservation. In D. Zarin, J. R. R. Alavalapati, F. Putz & M. Schmink (Eds.), Working forest in the neotropics, conservation through sustainable management? (pp. 119-129). Cambridge: Cambridge University Press.

PAGE 193

181 Schmink, M., & Wood, C. H. (1987). The 'political ecology' of amazonia. In P. D. Little & M. M. Horowitz (Eds.), Lands at risk in the third world: Local level perspectives (pp. 38-57). Boulder: Westview. Segura, O., Kaimowitz, D., & Rodriguez, J. (1997). Politicas forestales en centro america: Analisis de las restricciones para el desarrollo del sector forestal. San Salvador: IICA-Holanda/LADERAS C.A., CCAB-AP, Frontera Agricola. Shrestha, N. K., Kafle, G., & Britt, C. (1998). Community forest user group networking and the emergence of a federation of community forestry users in nepal. Forest Trees and People, 32/33. Smutylo, T. (2001). Crouching impact, hidden attribution: Overcoming threats to learning in developing programs.Unpublished manuscript, Ottawa. Stone, S. S. (2003). From tapping to cutting trees: Participation and agency in two community-based timber management projects in acre, brazil. University of Florida, Gainesville. Susskind, L., McKearnan, S., & Thomas-Larmer, J. (1999). The consensus building handbook, a comprehensive guide to reaching agreement. Thousand Oaks: SAGE publications. Thomson, J. T., & Freudenberger, K. S. (1997). Crafting institutional arrangements for community forestry. Rome: FAO. Toni, F. (2003). La gestin forestal en los municipios de la amazonia brasilea. In L. Ferroukhi (Ed.), La gestin forestal municipal en amrica latina (pp. 145-178). San Jose: CIFOR-IDRC. USAID. (2002). Tools fo alliance builders. Washington D.C.: USAID. Veer, M. C., Victor, M., & Fisher, R. J. (1997). Overview and introduction. Bankok: RECOFT. Walters, C. (1986). Adaptive management of renewable resources. New York: Macmillan. Wenger, E. (1988). Communities of practice; learning, meaning and identity. Cambridge: Cambridge University Press. Wenger, E., & Snyder, W. M. (2000). Communities of practice: The organizational frontier. Harvard Business Review, January-February, 139-145.

PAGE 194

182 White, A., & Martin, A. (2001). Strategies for strengthening community property rights over forest: Lessons and opportunities for practitioners. Washington D.C.: Forest Trends. White, A., & Martin, A. (2002). Who owns the world's forest? Forest tenure and public forest in transition. Washington D.C.: Forest Trends Center for International Environmental Law. Wondolleck, J. M., & Yaffee, S. L. (2000). Making collaboration work: Lessons from innovation in natural resources management. Washington D.C.: Island Press. Wood, C. (2003). Land use and deforestation in the amazon.Unpublished manuscript, Gainesville. Yaffee, S. L., Wondolleck, J. M., & Lippman, S. (1997). Factors that promote and constrain bridging, a summary and analysis of the literature. Ann Arbor: Ecosystem Management Initiative, School of Natural Resources and the Environment, University of Michigan. Ann Arbor.

PAGE 195

BIOGRAPHICAL SKETCH Franklin Paniagua was born in San Jose, Costa Rica. He went to Law School and parallel to his studies did volunteer environmental work. In 1990 he began working at CEDARENA, the Environmental Law and Natural Resources Law Center, where he developed his professional career focus on coastal planning and legislation before turning to environmental conflict management. Based with CEDARENA, he coordinated the Mesoamerican Conflict Management Network a broad group of organizations in their use of collaborative approaches to natural resource management. The network served as a learning community to develop mediation tools from the ground up. Franklins later work involved assessing and facilitating environmental conflicts throughout Latin American in places that he learned to love, such as the Araucania, Baja California, Galapagos, Petn, and Acre. He is pursuing his Ph.D. at the University of Florida in Gainesville. 183


Permanent Link: http://ufdc.ufl.edu/UFE0010484/00001

Material Information

Title: Factors Conditioning the Development of a Community Forestry Coalition in Western Amazonia, Brazil
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0010484:00001

Permanent Link: http://ufdc.ufl.edu/UFE0010484/00001

Material Information

Title: Factors Conditioning the Development of a Community Forestry Coalition in Western Amazonia, Brazil
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0010484:00001


This item has the following downloads:


Full Text











FACTORS CONDITIONING THE DEVELOPMENT OF A COMMUNITY
FORESTRY COALITION IN WESTERN AMAZONIA, BRAZIL

















By

FRANKLIN PANIAGUA


A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF ARTS

UNIVERSITY OF FLORIDA
2005
































Copyright 2005

by

Franklin Paniagua

































In memory of Lubin Villalobos and Guillermo Marin















ACKNOWLEDGMENTS

I feel very fortunate to have the chance to thank the people whose love and

care have made this document possible. My parents, brother and sister, were my

strongest supporters. The teaching and guidance of my committee (Marianne

Schmink, Karen Kainer and Tom Ankersen) gave me the opportunity to learn and

grow both personally and professionally. I am deeply grateful for Marianne's

extraordinary patience and faith in my work. I thank Jon Dain and Bob

Buchbacher for their insightful comments and for becoming de facto committee

members. I would like to acknowledge Magna Cunha (in Brazil) for her openness

and dedication to work. Magna was the person who suggested the Grupo de

Produtores Florestais as a research topic. Through her, a network of friends

and colleagues facilitated my work. Among them are: Nivia Marcondes, Pedro

Bruzzi, Luis Meneses, Renata Texeira and Carlos Vicente; the community

forestry leaders from Peixoto, Cachoeira, Porto Dias and Cautario.

Finally, heartfelt thanks go to my friends and "family-of-practice" at UF:

Hannah, Margarita, Wanda, Myrna (CLAS-TCD); Lin (Editor-in-Chief), Alfredo,

Claudia (Senior Editor), Meredith, Wendy-Lin (Excited Editor), Tita (y Pablo), Ana

Cristina, Richard, John, Christine, Amy, Maria, Emilio, Daniel and Adriana. In

Costa Rica: Roberto, Sergio, Luis, Carlos, Llamil, Gaby, Ale, Andrea, Rolo and

"nutelly" to Vivi.














TABLE OF CONTENTS

Page

A C K N O W LE D G M E N T S ........................................ ......................... ............... iv

LIS T O F TA B LE S ....................................................................................... viii

L IS T O F F IG U R E S ......................................................................................... ix

LIST O F A B B R EV IA T IO N S ..................................... ....................... ............... x

ABSTRACT ............... .........................................................................xi

CHAPTER

1 IN T R O D U C T IO N ................................................................................... 1

Sociopolitical C ontext ................. ... .... .......... ...................................... 4
Natural Resource Management Institutions......................................... 7
H isto ry of the G P F ............................................................. .. .......... 10
Research Design and Methodology ......................................................... 16
Research Design ............................................................... ............... 16
H ypothe sis ................................................................. ........... 18
S ite s .................................................................................................. . 1 8
Instru m e nts ................................................................. .. .......... 19
F ie ldw o rk ....................................................................................... 2 0
D a ta A n a ly s is ..................................................... ................ . .......... 2 1
S y n th e s is ................................................................................................. . 2 1

2 ANALYTICAL FRAMEWORK.................................................................. 23

Natural Resources Management Coalitions............................................. 24
Approaches to Analyze Community Forestry Coalitions ............................ 26
Framework to Analyze Community Forestry Coalitions................................ 31
Internal Factors .................................................................. ............... 34
E xte rna l F a cto rs ................................................. ................ . .......... 4 2
Design Factors ................................................................... ............... 49
Integrating the Analysis ..................................................... ............... 53
Analytical Framework Limitations ...................................................... 58
S y n th e s is ................................................................................................. . 6 0









3 IN T E R N A L FA C T O R S .............................................................. ............... 6 1

T he G P F C coalition M em bers..................................................... ............... 62
P o rto D ias P roject........................ ....... ............................................ 6 5
Centro de Trabalhadores da Amaz6nia....................................... 67
Associacgo de Seringueiros de Porto Dias................................. 69
P e ixoto P roje ct .......................... ........ ............. ............. 7 0
Empresa Brasiliera de Pesquisa Agricola ................................... 73
Association of Rural Producers of Peixoto.................................. 74
C achoeira Project .......................... .. ... ................... ................. 75
Associacgo de Moradores e Produtores do PAE Chico Mendes .... 78
Cachoeira Project's External Partners......................................... 79
C a uta rio P roje ct ................................................................. . .......... 82
A g5o Ecol6gica G uapore.............................................. ............... 84
Associacgo dos Seringueiros do Vale do Guapore....................... 84
C om parative A analysis .............................................................. . .......... 86
P a rticipatio n ................................................................................... 8 7
Le ad e rsh ip .................................................................. ........... 8 9
M anagem ent C capacity ........................................................ ............... 92
P project R relationships ......................................................... .... ........ 96
P projects' T ra ining F ocus ...................................................................... 99
Projects' Appropriation by the Community Associations.................... 107
S y nth e s is ........................................................................................... 1 13

4 EXTERNAL FACTORS...... ........................... .................... 117

Law and P o licies ................................................................................ 119
Land T enure ................................................................................ 119
Forest M anagem ent ......................................................... .............. 122
Social O organization ................ ........................................... 124
Markets and Community Timber Projects ......................... .................. 127
External Stakeholders ............................................................ ............... 129
Federal Government ........................................................ ............... 130
Acre State Government ...... ................. .................... 131
W o rld W ild life F und .......................................................... ........... 135
Furniture Company AVER ...... ........ .................... 136
External Stakeholder Map ................ ........................ 136
C ross-scale Interactions ......................................................... .............. 137
Extractive R deserves ......................................................... .............. 138
P o litica l D ile m m a .............................................. ............... .. .......... 13 9
S y nth e s is ........................................................................................... 14 1








vi









5 DESIG N FACTO RS .............................................................................. 144

Concept and Design .............................................................................. 145
Perceptions about the G PF ...... ......... ....... .................... 148
Partner O organizations ...................................................... .............. 148
Com m unity Associations ....... ....... ..... .................... 150
External S takeholders ...................................................... .............. 151
Perception Matrix........ .... .............. .................. 152
Process D esign ..................................................................................... 154
Innovation ....................................................................................... 155
The GPF Timetable ......... .... ............. .................. 155
Innovation Process ............ ............. .................. 157
S ynthesis ...................................................................... ........ ............... 162

6 C O N C LU S IO N S ................................................................................... 164

Results as Related to the Hypothesis ....... ... ................................... 165
H hypothesis ...................................................................... ..................... 167
M e th o d s ..................................................................................................... 1 6 8
E m erg ing Issues ...................................................................................... 168
Econom ic and Cultural Transition...... .... .................................... 169
Community Enterprises ......... ... ....... .................. 171
Future of the G P F ................................................................................. 173

LIST OF REFERENCES ............................................... 175

BIOGRAPHICAL SKETCH ............. ............................... 183















LIST OF TABLES

Table Page

1 .1 : H isto rica l T im e lin e ..................................................... ............... . .......... 6

2.1: NRM Collaborative Organization Analytic Variables............................... 28

2 .2 : C F C A na lytica l Fram ew ork .......................................................................... 33

2.3: Elem ents of the Innovation Process ....................................... ............... 52

2 .4 : Inte g rating M atrix .................................................... ................ ........... 54

3.1: C om parative A analysis V ariables ............................................. ............... 87

3.2: Task Distribution by Projects ................ ......................... 108

3.3: Project Appropriation by the CAs...... .... ...................................... 112

5.1: G PF Perception M atrix ....... ......... ........... ..................... 152

5.2: G PF Coalition Tim etable...... ......... ........ ..................... 156

6.1: Integrating Matrix with Results................ .......................... 166















LIST OF FIGURES

Figure Page

1.1: G P F P ro b le m T ree .................................................. ............... ........... 13

1.2 : D rivers of the G P F Initiative .................................................... ............... 14

2.1: Hierarchical System of Factors Conditioning the GPF................................. 34

2.2: C FC Internal Agents ............................................................... .............. 35

2.3: Timber Project Community of Practice .................................................. 37

2.4: C om m unities of Practice M odel .............................................. ............... 38

2.5: CFC External Agents .............................................................. ............... 43

2.6: Historic Trajectory Modeled on the Adaptive Cycle. .............................. 46

3 .1 : S ta ke ho ld e r M a p ..................................................... ............... . .......... 6 4

3.2: Task Distribution by Project...... ...... ...... ..................... 109

3.3: Task Distribution Overall ....................................................... ............... 111

4.1: Policies Supporting Timber Extraction and the GPF Coalition................ 127

4.2: Community Timber External Stakeholders ..................... .................. 137

4.3: Hierarchical Web of Innovations ....... ... ....................................... 140

5.1: G PF C coalition Structure...... .......... .. ........ ..................... 146

5.2: GPF Functional Structure ....... ............... ..................... 160














LIST OF ABBREVIATIONS


CA

CFC

CIFOR

CNS

CTA

EMBRAPA

FSC

GPF

IBAMA

IMAC

INCRA

NRM

NTFP

PESACRE

PO

SEATER

SEFE

WWF


Community Association

Community Forestry Coalition

Center for International Forestry Research

National Rubber Tappers Council

Amazonian Workers Center

Brazilian Institute of Agricultural Research

Forest Stewardship Council

Forestry Producers Group

Brazilian Institute of Environment

Environmental Institute of Acre

National Institute of Colonization and Agrarian Reform

Natural Resource Management

Non-Timber Forest Products

The Acre Group for Agroforestry Research and Extension

Partner Organization

Secretariat of Rural Extension and Technical Assistance

Secretariat of Forestry and Extractivism

World Wildlife Fund














Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Arts

FACTORS CONDITIONING THE DEVELOPMENT OF A COMMUNITY
FORESTRY COALITION IN WESTERN AMAZONIA, BRAZIL

By

Franklin Paniagua

August 2005

Chair: Marianne Schmink
Department: Latin American Studies

My study examines how forest communities might be linked to markets on

their own terms, analyzing particularly the development of a community forestry

coalition in western Amazonia, Brazil. I built on growing interest in pro-grassroots

coalitions that integrate organized communities, NGOs, government and

international agencies as a strategy to successfully link communities to markets. I

used a multi-layered assessment that integrates internal, external, and design

factors conditioning the development of a coalition among a set of four

community timber projects from the states of Acre and Rond6nia, in the Brazilian

Amazon region. Developing a second-degree organization (such as the coalition)

represents a self-organized response to the social and organizational challenges

of executing community timber extraction projects.

A major challenge was to try to meet market demands while responding to

local social and political constraints. The coalition followed an NGO-project









pattern instead of a collaborative organization pattern, creating a gap between

the original idea of the coalition and the actual process. My findings support the

need to re-adjust the role that social learning plays in building an autonomous

alliance among community forestry projects, and contributes to the development

of instruments for analyzing collaborative organizational innovations. Because

this coalition was still at a very early stage compared to cases in other regions,

such as southern Mexico's Sociedad de Ejidos, it is still difficult to know whether

the coalition is a secure structure for equitable market integration.














CHAPTER 1
INTRODUCTION

How can communities be linked to the market on their own terms?

(Schmink, 2004), This powerful question lingers as an emerging challenge for

community forestry. Responses from the tropics reflect changing trends in policy

that have opened a window of opportunity to build more equitable relationships

between the rural poor and the market. Among the changing trends: almost 25%

of the global forest estate is owned or administered by indigenous and rural

communities (White & Martin, 2002); democratization is stimulating an increase

in local people's participation in decision making; and global markets are

increasing the demand for tropical timber with a premium on certified social and

ecological products1 (Scherr et al., 2002).

Nevertheless, these trends have diverse effects on the tropics as regions

undergo transformation in different ways and at different paces. Some authors

suggest that a key factor in successful market strategies is the development of

pro-grassroots coalitions that integrate organized communities, NGOs,

government and international agencies (Alcorn & Toledo, 1998; Brown &

Rosendo, 2000; Fisher et al., 2001).


1 A certified product is one for which certain production characteristics (conditions, procedures,
byproducts) are overviewed by a third party who guarantees to the consumers of the product, that
certain production standards are maintained. In the case of timber, the system used is the Forest
Stewardship Council (FSC). This is an international body that establishes the standards to
certify wood based on environmental and social considerations. FSC has a body of certifierss",
conducts the assessments, and provides the credential to the production unit.









The state of Acre in Brazil has undergone intense transformation over the

past 15 years since the initial struggle of the seringueiros (rubber tappers) (led by

Chico Mendes and others) drew international attention, and watershed major

political changes encouraged forest-centered development (Kainer et al., 2003).

The result of these policy struggles was the creation of the Extractive Reserves

(RESEX) and the Agro-Extractivist Projects (PAEs) that secured land tenure

rights for the rubber-tapper communities. Currently, "the Forest Government",

with national and international support, is engaged in generating a suitable policy

environment for sustainable community-centered development.

In an effort to diversify traditional "non-timber forest products" (NTFPs) such

as Brazil nut and rubber, a number of agencies have set up community timber

extraction projects. Each project paired an NGO or government agency with a

local community association. Until the summer of 2002, these projects had had

partial success, experiencing difficulties consolidating production and identifying

stable markets for their timber products. The projects did not have the capacity to

produce sufficient quantity or adequate quality for the national and international

markets interested in sustainable community timber.

In response to this situation, several groups proposed the creation of a

consortium of community forestry producers (Grupo de Produtores Florestais

Comunitarios, GPF). This second-degree organization was to act as a platform

for joint marketing of the projects' timber. According to the proposal, the

community associations participating in the timber projects would lead the

coalition. The aggregated volume of wood from the different projects would









make it more attractive to sustainable timber buyers from cities like Sao Paulo

and Rio de Janeiro in southern Brazil.

Other regions with community forestry development, such as Central

America (Segura et al., 1997), Nepal (Shrestha et al., 1998) and especially

southern Mexico (Bray et al., 2003) have experienced successful second-degree

organization development2. These larger-scale inter-community innovations

have, in turn, secured and strengthened timber production throughout their areas

of influence (Colchester et al., 2003). Protected areas and watershed

management are also fields where this phenomenon of stakeholder collaboration

has grown in recent years (Wondolleck & Yaffee, 2000)

The capacity to build a regional coalition and to have it render positive

feedback to the local timber activity is dependent on a complex set of factors.

These factors range from the social capital and organizational capacity of the

potential member groups, to state and federal policies that shape the technical,

and the organizational context. The purpose of my study was to untangle the web

of factors that condition the success of this organizational innovation. I used a

hierarchical approach that analyzed three categories of factors: internal, external

and design. The analysis of these factors was then integrated through a series

of matrices, to contrast and compare the effects of each category in the coalition

development process. The information resulting from this research is meant to

contribute to the ongoing process of providing a picture of the emergent

2 A second-degree organization is a formally established group made up of other organizations.
Typically the member organizations will be of a lower scale of order. For example, a federation is
a second-degree organization of associations. Associations could come from specific
communities while the federation would represent an entire region.









constraints of the GPF initiative, and to give insights towards alternative paths for

overcoming these constraints.

Sociopolitical Context

The state of Acre has an area of 153.149 km2 and a population of

approximately 600,595 in 2000. The administrative structure of Acre is divided

into three major provinces, 5 regions, and 22 municipalities. Forest conversion in

the state has been fast, grouped and dominated by the establishment of cattle

ranches. Yet by 2001, Acre still retained 89% of its original forest cover.

In the 1880s Acre was colonized as part of the first rubber boom to supply

rubber to ports in Manaus and Belem. This boom was instigated by both Charles

Goodyear's discovery of the vulcanization process, through which rubber is made

chemically stable by exposure to sulfur and heat (1839); and demand for tires for

bicycles and cars. Between the years 1840 and 1860, the price of rubber nearly

tripled (Bakx 1988), at the height of the rubber boom in 1899, the territory of Acre

supplied 60% of the rubber from the Amazon (Hall 1997). During this time,

rubber barons and laborers from northeastern Brazil heavily colonized Acre. In

1903, Acre was officially annexed to Brazil after a war called the Acre Revolution.

The fall of the rubber trade occurred around 1910, when rubber trees grown

in Southeast Asia began to produce. Increased production in Asia caused rubber

prices to fall in Brazil and led to the eventual collapse of Brazil's export market

(Hall 1997). A smaller, second rubber boom in Brazil during World War II when

the Allies lost control of the Asian rubber plantations. In 1942, the United States

and Brazil signed a five-year agreement to support increased rubber production









in Brazil and ensure their access to Brazilian rubber (Schmink and Wood 1992),

which revived the traditional rubber tapping system known as the seringal.

Because of the declining rubber market between the 1950s and 70s, many

rubber estate owners abandoned their landholdings, leaving local rubber tappers

to begin tapping rubber for subsistence and trade purposes (Kainer et al. 2003).

With another influx of land speculators in the 1970s, under the support of

Brazil's military dictatorship, many rubber tappers were thrown off their lands. In

response, they began to organize, under the leadership of such figures as Chico

Mendes, and fight for their land. The creation of extractive reserves in 1989 was

an outcome of this movement, following the fall of Brazil's military dictatorship

and the support of national and international environmental groups, who saw

extractive reserves as an ideal "grassroots" conservation and development

initiative.

Table 1.1 summarizes the main historical events that have molded the

current socioeconomic and political situation in Acre. The table pairs the events

from Acre with those at the national level (Brazil). The table follows the rubber

boom-and-bust cycle from the mid eighteen hundreds to the mid nineteen

eighties. Starting in the late eighties, signs of a re-organization process

emerged, with the transformation of the old rubber estates into common property

extractive reserves. This reorganization phase continued through the nineties

and to the present.

The social and political struggles of the 1980s paved the way for most of

the resource management institutions were created. The degree of violent










conflict, especially in the Alto Acre region of Xapuri and Brasileia, was significant

although not as extreme as in the eastern Amazonian state of Para. Wilson

Pinheiro and later Chico Mendes were among the rubber tapper leaders killed at

the hands of cattle ranchers.

Table 1.1: Historical Timeline
Year Acre Brazil
1852 Annexed to the province of Amazonas
1877- Large immigration from the northeast to Drought in the northeast
1880 the Amazon
Aviamento system
1898 Bolivia creates the "Departamento de
Acre"
1903 Acre Revolution: armed conflict between
Brazil and Bolivia that resulted in the
creation of the State of Acre as part of
the Brazilian Federation. Prior to the
conflict the territory as part of the
Bolivian Republic
1904 Petropolis treaty, creation of the
state of Acre of Brazil.
1910-20 Decline of the rubber economy due to
competition from Southeast Asia rubber
plantations.
1920-45 "Dormant" economic period
1945-49 Resurgence of the rubber economy and Treaty between Brazil and USA
the "Seringaf' system governments to supply rubber to
allied forces during WWII
1950- Building of rubber tappers social Military dictatorship in Brazil (1964-
70's organization; deforestation; development 1984). Land struggles in Amazonia
1980 Land tenure conflicts in the southwest Gradual democratization
region of the state
1989 First extractive reserves Legalization of extractive reserves
1998 Jorge Viana's Forest Government is
elected
2001 BR-317 pavement by the state Avanga Brazil: infrastructure for the
government) Amazon

It is interesting to note that while the same violence that affected Acre

happened in other places (Rond6nia Para) the result was significantly different.

A number of factors transformed the "path of violence" in an unexpected way for

Acrean history. Instead of cattle ranchers expelling other social groups and

taking over their land, they were defeated through the support of the government.









Cattle expansion was halted in Acre through the creation of extractive reserves

(RESEX). Although originally presented as an alternative means for rubber

tappers to secure land tenure, extractive reserves became an innovative

conservation and development strategy that was championed internationally.

Along with the leadership and organizational capacity of the rubber tapper

movement, the international support for the RESEX was key in changing

attitudes of the Brazilian government. With the creation of the RESEX, violence

de-escalated, and an opportunity for reorganization emerged. The success of

this social movement led Jorge Viana to win the gubernatorial race in 1998.

Under the banner of the "Governo da Foresta" (the forest government), Viana's

progressive government has institutionalized most of the ideas that flourished

during the conflict years. Extractivist activities and community forest

management are strongly supported by the state government.

Currently, there are three extractive reserves in Acre, including the Chico

Mendes Extractive Reserve, which is the largest in Brazil, and eight agro-

extractivist projects (PAEs). The success of extractive reserves has come under

scrutiny in recent years, due to increased deforestation within their boundaries,

largely due to the fact that extraction of non-timber forest products, alone, is often

not sufficient for making ends meet. Cattle ranching and logging have increased

in Acre, and with the paving of BR-317 in 2001, supported by Acre's Governo da

Floresta, the landscape will continue to change.

Natural Resource Management Institutions

In Acre, natural resources are managed through a complex web of

individual, communal, state and federal level institutions. At the base of this









management lies the property rights system. RESEX and PAE (Agro-extractivist

settlements) land is owned by the Brazilian government at the national or state

level, and granted to a community as a concession, normally for thirty years.

Since extractive reserves are managed by IBAMA, the Brazilian environmental

protection agency, and PAEs by INCRA (Land Reform Agency), communities are

required by IBAMA and by INCRA to create plans for the sustainable

management of natural resources within the reserves. Property rights are held

by individuals, as well as communally in extractive reserves, with rubber trails

providing the base for private access to natural resources. As rubber prices

decline, and the value of harvesting Brazil nuts increases (along with the illegal

harvest of timber or conversion of forested land to pasture within the reserves),

the allocation, use and control of forest resources within this land tenure system

is bound to change.

Land cover change decisions promoted by landowners must be approved

both by IBAMA and IMAC, the federal and state environmental agencies. Forest

extraction goes through a particular process called "licenciamento," or licensing,

which requires the development of a Management Plan that must be approved

by a forest engineer. Timber, once harvested, requires special transportation

permits and for the producer to pay a series of state and federal taxes. The cost

and duration of all these legal procedures is beyond the reach of most small

landholders, who then turn to illegal logging practices. In extractive reserves the

alternative has been to obtain this legal authorization as a group, through the

development of collective timber extraction projects. The banner community









timber extraction project is in the community of Cachoeira, just outside of the city

of Xapuri, in the Alto Acre Region.

Illegal logging is controlled by both IMAC and IBAMA, but the actual legal

investigation corresponds to the Minist6rio Puiblico (Attorney General). Illegal

logging cases do not commonly come up. Of the few cases that are brought to

court, even fewer get sentenced because in most cases there is not sufficient

evidence. A particular problem at the Minist6rio Piblfico is the issue of the

reservea legal". This is a minimum percentage of property that must remain

forested, according to federal land tenure laws. In the Amazon region 80% of

each private property must remain in forest cover (90% for RESEX). This

limitation has also become an incentive for illegal logging.

A significant alignment of policy actions has been occurring in the years

since the transformation generated by the rubber tapper social movement and

emergence of extractive reserves. These elements are

* The zoneamento econ6mico-ecol6gico This state-wide economic and
environmental planning strategy brought together a large set of information
sources into a comprehensive document on the status of the environment in
Acre. This study represents a significant baseline for tracking changes in
land-use and establishing sustainability thresholds.
* IMAC this state environmental agency represents a significant source of
information and decision making that was less active before the
establishment of the governoro da floresta"
* The consolidation of UFAC, the Federal University of Acre as a source of
environmental research and training of professionals. UFAC researchers
have been key in the development of the MAP (Madre de Dios, Peru; Acre,
Brazil and Pando, Bolivia) network that later became a large-scale
interstate and international effort
* The development of learning networks between research organizations like
UFAC and EMBRAPA (Brazilian Institute of Agricultural Research) with
development NGOs (PESACRE, the Acre Group for Agroforestry Research
and Extension; CTA, Amazonia Workers Center), local organizations and
government agencies on issues such as fire management, forestry and









agro-forestry. These networks are also linked to international as well as
national research centers and funding sources.

History of the GPF

The emergence of the Grupo de Produtores Florestais (GPF) must be

understood within the broader process of community timber extraction

development in Western Amazonia.The fifteen-year period since the

establishment of extractive reserves (RESEX) and agro-extractivist settlements

(PAEs) has been fruitful but challenging at the same time. The struggle of the

rubber tappers' social movement has rendered not only the necessary legal

institutions to secure their land tenure, but also a significant improvement in the

political role and social stature of the extractivist communities in Amazonia and in

Brazil, in general. This has resulted in small but significant improvements in the

quality of life of the rubber tapper families. However, the opportunity provided by

secure land tenure came with its own set of problems, some of them due to

changes in the broader political and economic context. Others were unintended

consequences of the social movement's success.

Community timber extraction can be considered a "bitter fruit" of this

process. It is a fruit because the regained land rights allowed extractive

communities to take advantage of this highly valuable resource. It is bitter

because the original intention of RESEX-far from cutting trees-was to

conserve the forest for traditional extractive activities (e.g., Brazil nuts, rubber).

The failure of traditional extractive products to sustain a viable livelihood

motivated the search for alternative activities and land uses, among them cattle

ranching and timber extraction.









Cattle ranching came almost naturally to rubber tapper families, as it

already formed part of their livelihood strategies. Surplus from other activities

would be invested in cattle as a secure savings. Increasing the number of head

of cattle per family requires changing land cover from forest into pasture, a

process that has increased in recent years (Gomez, 2001; Salisbury, 2002).

Although legally each family has a right to clear 10% of their land for pasture

and/or agriculture, the increased economic role of cattle drives some families to

deforest beyond that threshold. The cumulative effect of larger numbers of

families following this "cattle" strategy is uncertain, but it does call into question

the existence of RESEX as a means of maintaining forest cover, since forest and

pastures are mutually exclusive land covers.

On the other hand, timber-if sustainably harvested-could be

complementary to the RESEX model. However, it would require a major

institutional and cultural change in Western Amazonia. The process of directing

this change was spearheaded by NGOs that had played a supporting role to the

rubber tappers' movement through the eighties and nineties. These groups

transformed themselves into forest managers to lead the development of

experimental community timber projects. These experiments began around

1996-97. The projects struggled with a variety of obstacles, including direct

opposition from significant parts of the rubber tapper community (Kainer et al.,

2003). Interest in the activity grew with the election of Jorge Viana-trained as a

forest engineer-as governor of the state of Acre in 1998, and has increased

significantly with his reelection in 2002. Expectations regarding the benefits of









timber extraction for extractivist communities have increased to the point of

becoming a mainstream idea for the future of RESEX and PAE communities.

Economically, however, timber extraction has not shown itself to be as

successful for the rubber tapper families as they had expected. Until 2002, most

of the projects were far from generating stable profits or handing over complete

operational control to the communities (Stone, 2003). The cultural, political, and

institutional transformations required for timber extraction to serve as a viable

economic alternative for RESEX are still in their developmental stages. Timber

project leaders recognized this situation and proposed an articulation among the

different timber projects as a strategy to cope with the increasing complexities of

the timber harvest activities (CTA, 2001)

The timber project leaders presented this proposal for the Central de

Associagdes de Produtores Florestais do Acre, a union of the timber producing

associations, at the first meeting of community timber projects of Acre and

Rond6nia in December of 2001. The justification for the idea was presented in

the manner described in Figurel .1. The Central, later called Grupo de

Produtores Florestais (GPF), would vertically integrate the current community

timber projects to facilitate timber sales. This integration would alleviate the

different marketing problems experienced by some of the projects, as well as

channel the increase in production from new or existing community timber

projects. A vertical integration of projects would take advantage of larger

volumes of woods to link regional, national, and international markets. Control of

the Central would remain in the hands of the community-based associations,









guaranteeing that profits would return directly to improve the livelihood of the

extractivist families.







GPF
Coalition




Marketing limitations of
community timber
projects

Figurel.1: GPF Problem Tree

The Amazon Workers Center (CTA)3-proponents of the GPF idea-

detailed the reasons and expectations of the coalition, in the manner illustrated in

Figure 1.2. The main motivation for the proposal came from the problems

projects experienced in selling their timber. These problems were related to

issues such as: (1) the legal limitations on community-based associations to sell

products4, (2) the imminent reduction in funding from donors, and (3) the general

difficulty of demonstrating significant success in achieving the group's

sustainability goal.


3 Centro dos Trabalhadores da Amazonia is an NGO based in Rio Branco, the capital of the state
of Acre. Created in 1981 and with a long history of accompanying the rubber tappers social
movement, the group originally conducted education programs. In 1996, the organization started
the first Brazilian Amazonia community timber project in the agro-extractivist settlement (PAE) of
Porto Dias.

4 Community-based associations are legally limited to conduct social activities, not commercial
enterprises. For this reason timber sales can only be signed off by individual association
members. Normally it is the president of the association who is in charge of signing the sale.






















Ecor


SEntreposto
Selling post
GPF
Coalition


Marketing limitations
of community timber
Projects


Ecological Social

Figure 1.2: Drivers of the GPF Initiative


While these other problems were independent of the issue of marketing,

solving the problem of selling the timber would bring about a solution to them,

according to the GPF proposal's supporters. The GPF was envisioned mainly as

a "selling post" (entreposto da venda) that would receive timber from all the

projects and be responsible for marketing and selling. According to the plan, the

GPF would be organized as a cooperative of the different community

associations and, in this way, solve the legal limitations impeding the sale of

products by the social organizations. The entreposto would manage more

volume than the individual projects could, and would therefore be able to sell to

larger markets and take advantage of the FSC forest certification promoted by

VWWF-Brazil5(CTA, 2001). International and national buyers interested in





5 VWF: World Wildlife Fund -Brazil is a Brazilian NGO, part of an international network of WWF.
In the case of Acre and Rondonia, it has supported the Cautario community timber project in
Rondonia, and has promoted the certification of the community timber projects and the creation of
the GPF.


_WUNMAL*_ JL









certified wood would be interested in obtaining it from the community only if they

could guarantee a significant and stable volume.

In 2001, a large group of individuals involved in community timber

management and forestry policy in Acre participated in an exchange with the

ejidos of Quintana Roo, Mexico. This exchange provided participants with a vivid

example of a successful coalition of community forestry enterprises. The GPF

was modeled on the example of the Asociaci6n de Ejidos6. It was this closer

contact with the Mexican ejidos that helped to articulate the question of whether

Western Amazonia had the "necessary conditions" to develop a similar coalition.

Despite significant similarities such as the pre-existing extractivist culture, there

were some equally important differences between southern Mexico and western

Amazonia. One of the most important differences was the strong support

provided by the Mexican Government and some international agencies through

the Plan Piloto Forestal, for over fifteen years. Another difference was the strong

social capital shared by Ejido communities, that strengthened local leadership

and cultural identity.

Recognizing these differences, the proponents of the coalition had the

challenge to adapt the "Sociedad de Ejidos" model to the situation in western

Amazonia. This research represented an effort to inform the GPF proponents

about the existing factors conditioning the coalition initiative and the alternative

paths of development. For this reason it was designed and conducted


6 This association of ejidos is presented and discussed in Chapter Two.









collaboratively with the GPF proponents, as will be detailed in the following

section on the design and methodology of the study.

Research Design and Methodology

Research interest in collaboration, as a catalyst for change in NRM, is

relatively recent. Tools and approaches to study collaboration are still in the

making. Researchers in fields such as communal property or political ecology

have been more interested in the results or impacts of collaboration than in

looking at its process (Alcorn & Royo, 1999; Mayer, 2002; Ostrom, 1990).

Studies concerned with the way collaboration is developed are scarce. Most of

the analysis corresponds to development and conservation organizations such

as the United States Agency for International Development (USAID), Canada's

International Development Research Center (IDRC), or the United Nations Food

and Agriculture Organization's Forest, Trees and People Program (FTPP).

Different conceptual and applied publications have resulted from their field

projects (Buckles, 1999; USAID, 2002).

Research Design

The objective of this research is to yield insights for improving multi-

stakeholder collaboration platforms, similar to that attempted in the Brazilian GPF

initiative, the focal case of this thesis. The idea for the research developed from

conversations with a forestry researcher involved in several of the community

timber projects and also in the creation of the GPF. An immediate practical need

was identified to understand the conditions present for the GPF, both within the

groups and in the policy environment. The research would provide a deeper

understanding of these factors.









Based on literature review I organized these factors into three groups:

internal, external and design7. Internal factors refer to the organizational

conditions (leadership, management capacity, participation) and collaborative

practices at each of the community forestry projects. A total of four projects were

considered in the research. Each community forestry project consisted of a

community association (CA) and a partner organization (PO). Partner

organizations were both governmental and NGO types. External factors refer to

the policy and market environment in which the proposed initiative is embedded.

These factors are separated into laws, policies and institutional behaviors. State

and national (federal) laws and policies are of special interest, as well as the

funding policies of international aid agencies and NGOs. Design factors refer to

the conceptualization of the proposed coalition, and the way it has been adopted

by the different timber communities and their NGO partners.

The analysis of these factors was intended to provide a picture of the

potential success of multi-stakeholder coalitions. To establish a reference for

what would constitute a successful coalition process, I developed different

indicators derived from a selection of similar studies. The main source for these

indicators was the articles on The Union of Forest Ejidos (Union de Ejidos

Forestales) of southern Mexico by Santos and Bray in their respective

publications (Bray et al., 2003; Santos et al., 1998). Other studies done on

sustainable businesses in the Amazon basin (A. Anderson & Clay, 2001; Clay,

2001).


7A detailed description of the research framework is provided in Chapter 2.









Hypothesis

I hypothesized that my results would reveal a scenario in which both the

design and external factors would be positive for the emergence of the coalition.

By contrast, the internal factors would be the source of major constraints. I

developed this hypothesis based primarily on the literature that suggested

internal factors such as management capacity of the local groups and social

capital as the weak factors leading to the failure of community timber projects, in

analyses conducted by CIFOR and WRI-Forest Trends (Colchester et al., 2003;

Scherr et al., 2002). I also based it on my personal experience working in the

development of natural resource management coalitions. I had worked in

coalition efforts around watersheds and protected areas in several countries of

Latin America, such as Ecuador, Costa Rica, Mexico, Panama and Peru.

Sites

Four sites were defined for the study, corresponding to the four timber

projects involved in the first GPF meeting held in December 2001. Three of them

were in the state of Acre (Porto Dias, Peixoto and Cachoeira) and one was in

the neighboring state of Rond6nia (Cautario). I visited these four sites at least

once, during the five-week period defined for fieldwork. At the time of my arrival,

a new person from outside the region was hired to spearhead the GPF initiative.

She participated in the definition of the research plan and joined me for the field

visits and interviews. We worked as a team, conducting all the interviews and

analyzing each session together. However, the final analysis was done

separately.









Instruments

Three open-ended questionnaires were prepared, as well as research

guidelines for the analysis of relevant laws and policies. We were to interview

the NGO or public agency project directors (or coordinators) and at least two

members of community association boards from each of the four projects.

Where available, more project staff and community members would be

interviewed, as well as other relevant individuals not directly involved with the

timber projects.

The first questionnaire was directed to the group of external stakeholders

that included Acre state government officials, NGO coordinators and other

knowledgeable individuals not directly involved in the timber projects. The

second was geared towards the PO project staff, and the third to the community

association leaders. The first one focused on the perceptions of the government

officers on three topics: community forestry in Acre, issues and future of current

community timber projects (including the idea of the GPF), and the effectiveness

of timber extraction as a conservation policy. These topics were laid out in a set

of twenty questions. The other two questionnaires were oriented to sketch a

profile of each organization (NGO, government agency or community

association) involved in the project. These profiles included general basic

information about the groups (foundation, membership, organization) and a

particular look at their relationship with the partner organization in the timber

project. This meant asking the association about the NGOs and vice versa.

Both groups were asked about the management capacity of the associations and

their degree of "appropriation" of the timber project. In this context,









appropriation refers to the degree of control and operational autonomy the local

group had over the timber activity. Perception of the GPF proposal was a

crosscutting topic in all questionnaires. These two instruments had forty-two

items for discussion with each respondent.

Fieldwork

Data gathering included other sources such as laws, policy briefs and

organization bylaws, project reports and evaluations. Fieldwork extended from

June 24 to August 16, 2002. The first four weeks were devoted to the initial

contacts and research strategy design. From week four to week eight, interviews

were conducted and documents were collected. A total of forty interviews were

carried out, including sixteen community association leaders, fourteen NGO or

agency officials from the timber projects, seven state or federal officials, and

three researchers from the local Federal University of Acre (UFAC). Related

personnel from NGO and agencies were selected according to their involvement

in the timber projects. All project coordinators (or directors) were interviewed,

along with at least one other field-staff personnel from each PO. From the

associations, all presidents and treasurers (administrators) were consulted.

Other association members were also interviewed. University researchers were

included to gain a broader perspective on the role of community timber projects

in western Amazonia development.

Interviews took 1-2 hours on average; they were conducted as informal

conversations. Guiding questions or topics were suggested according to the

evolution of the discussion in a flexible manner, giving the most freedom to the

interviewee to lead the dialogue. All discussions took place in a friendly and









relaxed environment. All of the participants agreed willingly to be taped. Time

constraints allowed for only one visit to each respondent; only in very few cases

was there a chance for a second contact.

Data Analysis

Data from both primary and secondary sources were analyzed using

qualitative methods, such as stakeholder maps and comparative matrices.

Archival information was assessed to identify critical themes and develop

timelines of keystone events and processes. These data were critical for cross-

referencing comments from the interviews and the meeting notes. A stakeholder

map was developed initially as an instrumental tool to guide the analytical

process. Themes selected prior to the interviews were reviewed and used to

organize the data into a series of matrices that will be presented and discussed

throughout the thesis. Stakeholders external to the timber projects were

separated and grouped in a single matrix. Both project personnel (from NGO

and government agencies) and association leaders were brought together in

three different matrices: one regarding the transfer of project decision making

(appropriation); a second on the management capacity of the local associations;

and a third clustering their perceptions on the relationship dynamics within each

project. Perceptions about the GPF idea and its process so far were assembled

into a matrix including all the stakeholders.

Synthesis

This chapter introduces the emergence of the GPF coalition as the issue of

research of this thesis. It frames the Grupo de Produtores Florestais as a

response to the broader challenge of how to justly integrate rural communities to









global markets. The chapter describes the sociopolitical context of community

forestry in western Amazonia. It describes how the creation of extractivist

communal areas like RESEX and PAEs provided the opportunity for communal

timber projects to be established in the mid nineties. Solving the marketing

problems confronted by these projects was the main reason behind the initial

GPF proposal. The history, of this initiative is detailed in this chapter. The last

section of the chapter presents the structural pillars of this research. Beginning

with the hypothesis it describes the research design and strategy that was

supported by the theoretical framework presented in Chapter Two.














CHAPTER 2
ANALYTICAL FRAMEWORK

In 1993, Kai Lee's seminal book Compass and Gyroscope introduced the

concept of adaptive management, from the original ideas of C.S. Holling (Holling,

1978) and Carl Walters (Walters, 1986), to the wider conservation community.

Lee compared the current challenge of dealing with environmental change to that

of the sixteenth century explorers, who were acutely aware of the limited capacity

of their navigation gadgets and their knowledge. Also, they dealt with uncertainty

on a daily basis and generated practical knowledge, vital for their immediate

survival (Lee, 1993). The situation of practitioners working in the development

of multi-stakeholder collaboration efforts in natural resource management (NRM)

is no different. Most of these NRM initiatives have started spontaneously,

sparked by a major social or ecological disturbance. In other situations, a long

history of failures or conflict has left parties with collaboration as a last option. In

most of the cases, efforts have run on intuition and the goodwill of the parties

involved, more than on systematized knowledge. This issue represents a

limitation to collaboration's potential to achieve effective conservation and

development.

This chapter aims at contributing to the crafting of that knowledge by

providing an introduction to the study of collaboration in NRM, with particular

emphasis on the study of community forestry multi-stakeholder collaboration in

the form of formal second-degree coalitions. As sources for this review I have









considered the available, but scattered, literature on the subject, along with

works in related fields such as political ecology, adaptive management, social

learning and innovations. I have woven this diverse literature into an analytical

framework to assess the conditions for the emergence of multi-stakeholder

coalitions. I detail the components of this framework in the latter part of this

chapter and use it in the following chapters to evaluate the situation of the GPF

initiative in the western Amazon Basin, Brazil.

Natural Resources Management Coalitions

Coalitions and other types of multi-stakeholder collaboration innovations in

NRM have gradually drawn attention from researchers as the practice multiplies

in a diversity of ecological contexts. In each site where multi-stakeholder dialog

has grown into substantial collaboration and later into new forms of organization,

motivation seemed to lie within each participant's experiences and interests. The

process of collaboration is an expression of the self-organizing quality of complex

living systems (Holling et al., 1998). Neither theory nor enforced external policy

seems to have played a direct or substantial role in the development of this trend

(Brick et al., 2001; Wondolleck & Yaffee, 2000). The parties' needs assessment

plays a key role in crafting these collaboration efforts, but is accompanied in most

cases by a nested set of institutions and external factors that both constrain and

facilitate this bottom-up drive. These other factors, such as development

projects (and their founding theories), governmental policies and institutions, and

donor agendas, condition the expression and extension of collaboration.

In many cases collaborative fora are platforms where actors representing

formal social "structure" (NGOs, government agencies) come together with social









movement actors. As such, these "platforms" run the risk of being interpreted

solely as mediating devices to serve the powerful groups (Edmunds &

Wollenberg, 2001; Kenney, 2000; Nygren, 2000).

The emergence of a myriad of collaborative initiatives has occurred without

any thoughtful scholarly analysis of the conditions under which such efforts would

be successful. Many of the studies done thus far have been based on North

American experiences (Brick et al., 2001; Wondolleck & Yaffee, 2000), with only

a scattered number of case studies written on coalitions in the tropics (Fisher et

al., 2001; Poats, 2001) and even fewer that look at community forestry networks

(Rosendo & Brown, 1998; Shrestha et al., 1998).

Conservation organizations such as the Biodiversity Support Program

(BSP, 2000) and the Center for International Forestry Research (Buckles, 1999),

as well as The Nature Conservancy (Fisher et al., 2001; Monge, 2001) and the

International Union for the Conservation of Nature (Borrini-Feyerabend et al.,

2001), have publications promoting the development of collaborative resource

management, with documented cases and guides for establishing similar groups.

Even within this small amount of emergent "applied" literature, the complexity of

the phenomenon of collaboration is evident. The variety of titles (networks,

alliances, co-management, coalitions, federations, etc...) used to name the

collaborative structures are as diverse as the contexts in which the networks

arise. Fisher et al. underline the need for an objective and critical study of

collaborative experiences, given the risks of promoting alliances, coalitions, etc.,

as "the new magic bullet"for conservation (Fisher et al., 2001).









Community Forestry Coalitions (CFCs)

According to the FAO, the potential of forests and their resources to reduce

poverty depends upon, among other factors, support to participatory processes,

multi-party learning and enhanced interagency collaboration (Thomson &

Freudenberger, 1997). Community forestry, defined as the management of

forest resources by local people for their own benefit (Hood et al., 1998; Veer et

al., 1997), should rely strongly on developing the capacity to collaborate

effectively. In fact, forestry has transformed itself from a "technical" activity, into

an "interdisciplinary" field in which economic, social and political understanding is

crucial (Colchester et al., 2003). Community forestry is among the most sensitive

areas, where social and human skills play an important role. Community forestry

practitioners have to deal with the forest issue as it encounters market pressures,

land-right disputes, legally mandated participation and other current phenomena

produced by the rapidly changing political environment (Alcorn & Royo, 1999;

Escobar, 1999; Peluso, 1992).

Approaches to Analyze Community Forestry Coalitions

In this section I review different approaches that have been used to analyze

collaborative efforts in natural resource management. Guided by the definition of

CFCs from the previous section, I selected studies of collaborative efforts that

shared similarities with CFCs, so that their methods and tools could be

extrapolated for the analytical framework I designed to study the GPF in Brazil.

From this review a scattered set of elements from different sources was

identified, and three frameworks were compared to establish the final study

variables.









Researchers from a variety of disciplines have analyzed collaborative

resource management. Anthropologists and sociologists interested in social

movements and collective action have a long tradition of looking at peasants and

rural social movements (L. Anderson, 1994). Political ecology brought this same

attention to look at social groups and social movements with closer ties to their

environment, such as indigenous and extractivist societies. In this field,

collaboration and conflict between social actors at different scales remains an

area of interest for research (Bryant, 1991; Edmunds & Wollenberg, 2001;

McCay, 2000; Peluso, 1992; Schmink & Wood, 1987). Interest in linking social

and ecological systems has also sparked an interest in collaboration and

common property, leading to a set of interdisciplinary studies of collaborative

NRM systems (Folke et al., 2002; Holling & Meffe, 1996; Lee, 1993; Royo, 1999).

Social capital studies led by political scientists and economists also consider the

role of collaborative organizations. Community's organizational capacity, social

networks and second degree organizations (federations) are studied as forms of

structural social capital (Babbington & Carroll, 2003; Dasgupta, 2002; Pender &

Scherr, 2002).

These studies have applied a diversity of methodologies from quantitative

to ethnographic. The focus of research has leaned towards the results and the

impact of collaboration, not so much on the process. The conditions leading to

collaboration, or the organizations for collaboration, have received little attention.

Both social capital and common property (Kopelman et al., 2001; Ostrom, 2003)










represent new frameworks for which process and organizational characteristics

are increasingly the center of attention.

The efforts to understand NRM collaborative processes also evolved from

conflict management and environmental conflict management research (Buckles,

1999; Susskind et al., 1999). Conflict management researchers have mostly

relied on case study methodologies to extract the collaborative process

characteristics (Brick et al., 2001; Fisher et al., 2001; Margoluis et al., 2000;

Yaffee et al., 1997). In Table 2.1 I compared three of these cases across the set

of critical variables for collaborative research identified by each study.

Table 2.1 NRM Collaborative Organization Analytic Variables
Coughlin et al., 1999 Edwards et al., 2001 Fisher et al., 2001
* Location Function Goals and objectives
* Issues Membership Size and composition
* Participants Objectives Geographic range
* Outcomes Management Degree of formality
* Decision Authority Leadership Leadership
* Connection to existing Facilitation Management
procedures Information exchange Supportive policies and
Elements of process Skill development institutions
structure Communication Threats to conservation
Scientific basis for planning, Resources Process
decision-making, Levels (scale of members) Structure
implementation and Results
monitoring
Level of support/ opposition
Level of experience/
knowledge
Funding
Time frame (when initiated/
meeting frequency
Scale of projects
Land ownership

The first case, from Coughlin et al, is itself a comparative study of

conservation partnerships in the midwest of the United States. The second is

from the East-Asia Community Forestry Network (RECOFT), based on

characteristics identified by participants of a forestry network workshop in 2001.









In the third, Fisher et al. provide a conceptual framework for conservation

coalitions, based on protected areas cases in the United States and other tropical

countries. For this comparison I use the work of Arun Agrawal (2001) comparing

common property regime (CPR) institutions as a model. Agrawal concludes that

developing lists of conditions for CPR institutional analysis is useless (Agrawal,

2001). Nevertheless, I believe that such an approach is appropriate because this

analysis is targeted towards feeding back practical information, rather than

employing the case study merely as an anchor for a theoretical analysis of

institutional development. Moreover, as Agrawal points out, the list of

categorized variables can be calibrated and used in cross-case analysis with the

adequate database.

Despite being geared towards different types of collaborative efforts the

variables identified could be applied regardless of the type of structure

(partnership, network or coalition) (Table 2.1). Biophysical aspects were

included in variables such as location, geographical range, threats to

conservation, and resources managed. Background and contextual aspects

were incorporated in factors such as time frame, decision authority, scale of

members and, supportive polices and institutions. All three approaches

considered strategic aspects such as objectives and outcomes of collaboration.

Operational aspects were common to all cases, as well (structure, leadership,

management, resources, process).

These overall variable categories-biophysical, contextual, strategic and

operational-can also be broken down into two more general areas: internal









variables and external variables. Internal variables consider issues within the

groups involved in collaboration, in addition to the process and purpose of the

cooperation itself. External variables refer to all aspects outside the

collaborative process that provide the surrounding environment to the effort,

including biophysical, socioeconomic and political aspects.

In order to adapt these variables to evaluate the GPF initiative, I had to take

into account the issue of temporal change within collaborative efforts. While the

majority of studies have examined already-consolidated coalition processes, the

GPF initiative was still in a pre-emergence phase. References regarding the

conditions leading towards collaboration were made in retrospect and only for

successful cases (Colchester et al., 2003; Wondolleck & Yaffee, 2000).

From the set of variables considered above, the "external" set could be

applied directly to the GFP case, but the "internal" variables needed to be

readjusted. Since there was no existing coalition for which to consider these

variables (decision-making process, leadership or structure), I redirected these

variables to analyze the experiences of the potential coalition members. The

assumption was that the past and current organizational behavior of the potential

members would inform the behavior of these groups in the future coalition. By

analyzing current collaborative practices we could forecast their behavior in the

future coalition. To do this, I relied on the concept of communities of practice

(Wenger, 1988; Wenger & Snyder, 2000) as a guiding metaphor to study the

potential coalition-members' internal organizational issues, including factors such









as internal conflicts, leadership, and decision-making and organizational

capacity.

To analyze the GPF in its pre-emergent state, I decided to consider it as an

idea that was still being presented to parties and discussed. Therefore, I decided

to incorporate a particular set of variables that would assess the process of

design, diffusion and adoption. For this purpose, I adapted elements from the

study of technological innovations (Douthwaite, 2002; Douthwaite et al., 2001;

Rogers, 1995) that will be explained in a later section.

First, however, I detail the analytical framework that resulted from the

review of the literature on collaborative research discussed above. I go into more

depth regarding particular aspects of the literature as they inform sections of the

framework. This framework built on my previous work included coordinating a

cross-site learning coalition project for an international conservation organization

(cases in Panama, Mexico, Brazil, Peru and Panama). For this project I

prepared a self-evaluation tool for coalitions, which is related to this framework.

Previously, I had also participated in the documentation of three unrelated natural

resource management coalition cases (Galapagos National Park, Ecuador; Osa

Biological Corridor, Costa Rica; Serra do Divisor, Brazil)

Framework to Analyze Community Forestry Coalitions

Thus far, I have delineated the theoretical trajectory that led me to the

design of the analytical framework I used in my research. I used a hierarchy

theory approach (Allen, 2004) to structure my analysis of the factors providing

conditions for the emergence of a CFC. These factors were drawn from the

literature studying different types of conservation coalitions, and organized









through a hierarchical approach. Hierarchy theory posits that any level of

examination is constrained by the levels immediately above while controlling

those immediately below; thus, all three levels must be considered in order to

understand causes and consequences at the level of interest.

Figure 2.1 presents the components and boundaries of the hierarchical

system to study. The middle component (design factors) represents the main

scale of interest. This represents the level where the organizational innovation is

occurring. Design refers to the structure and process definition of the community

forestry coalition. This scale contains the lower scale of "internal factors" that

encompasses the groups that are interested in participating in the coalition. At

the highest level is the "external factors" component. This scale contains the

factors that constrain the activities of the interest groups (potential coalition

members).

By developing a hierarchical approach to CFCs, I pursue the integration of

the seemingly paradoxical aspects of coalitions, such as the manifestation of

both human agency and sociopolitical structure. Agency is contemplated within

the "internal factors" and structure in the "external ones". While the hierarchical

approach allows the mapping of the GPF's significant components, its main

dynamics might remain hidden. I focus on two key dynamics: first, on the cross-

scale interactions that will be analyzed mainly in Chapter Four (external factors)

and secondly, in the social learning process occurring within the timber projects

and through the GPF innovation process. Chapter Three (internal factors) and

Chapter Five (design factors) both discuss the learning dynamics.










Table 2.2: CFC Analytical Framework
Factors
Categories (Variables) Indicators (Sources) Collection tools Analysis tools
At the Group level: Documents of Questionnaires Stakeholder
o Participation activities, results Interviews with analysis and
o Leadership Organization key group map
o Management members' leaders
Capacity perception Archival research
Internal: At the Project Level: Project documents: Questionnaires Stakeholder
o Partner relationships reports, proposals, Interviews with mapping
o Capacity evaluations key project Comparative
development Project participants participants Matrix to elicit
o Appropriation by perception Archival research the common
community collaboration
pattern
State and federal Laws, policy Questionnaires Stakeholder
government agencies, documents, project Interviews with analysis and
policies and laws on proposals, key project mapping
External: conservation, forestry regulations, donor participants
development and social policy documents, Archival research
organization NGO reports,
NGO and Aid agencies Key official
Forest products markets perception about
the issues
Coalition Design: Documents Questionnaires Graphs
o Objectives detailing the Interviews with visualizing
o Structure coalition proposal key project coalition
o Leadership Promoters' vision participants structure
o Resources and perception Archival research alternatives
Design: Innovation Process Promoters' vision Questionnaires Innovation
o History and perception Interviews with process
o Perceptions Potential key project timeline
o Adaptations participants' participants Coalition
perception Archival research proposal
External supporting perception
agents' perception matrix



Table 2.2 shows the three main categories of factors: internal, external and

design (column 1). The remaining four columns detail the factors within each

category, the indicators for each of those factors, and the relevant collection and

analytical tools8.


8 It should be noted that this matrix organizes the factor categories differently than Figure 2.1.
The chosen sequence that starts with describing the actors (internal) then the context (external)
and finally the innovation process (design) is meant to be more accessible to the reader. It is also
a sequence that resembles more closely the research process that when from the actors to the
context and finally looked at the coalition innovation.









The next sections of this chapter will describe each of the three general

categories, starting with each category's objective and then moving horizontally

across the table, discussing the variables, the indicators and the tools. A

subsequent section integrates the three components. The final section presents

some of the caveats and limitations of the analytical framework.



SPolitical Ecology



/ Diffusion of Innovations




Communities of practice,
Social learning


Figure 2.1: Hierarchical System of Factors Conditioning the GPF9

Internal Factors

The internal factors category summarizes the conditions of the

organizations interested in the formation of a coalition. In the case of the CFC

cases reviewed, these groups were mainly community-based associations

directly involved in community timber enterprises. In this analytical framework, I

consider not only these groups but also the NGOs or government agencies that

jointly developed the timber activities with the communities (Figure 2.2).

Although formally, CFCs have not included entities other than community-based


9 Stone (2003) uses a similar diagram in her analysis of local participation in two community
projects in Acre.









groups, I believe the supporting groups play a very important role in the

emergence of this type of second-degree organization, particularly in the case of

Acre and Rond6nia. In these states, community timber activities are conducted

through projects financed by international donors and executed by these non-

community groups. Both the community associations and the partner

organizations described in this figure conform the middle component of this

study's hierarchical system.

Community Forestry Community Forestry
Projects organizations
Support NGO or Community-based
Gov Agency Organization














Figure 2.2: CFC Internal Agents

Unlike Mexico, Nepal or Guatemala where the community-based groups

run the enterprise, in western Amazonia community timber extraction is still at an

experimental stage. The autonomy of the community associations is very

limited. This explains why I chose to use the forestry project as my unit of

analysis. Each project was composed of a community-based group and a private

or public agency that supported the activity technically and financially (Figure

2.2). In a material sense, the community forestry coalition would incorporate the









forestry projects, which were composed of the two types of organizations:

community associations (triangles) and partner organizations (diamonds).

The objectives of this component of my research were three-fold. First, to

describe the basic characteristics of each organization. Second, to assess the

internal collaboration characteristics by looking at leadership, participation and

management capacity. Third, to compare the projects in terms of partner

relationships, degree of appropriation by community, and training focus. I

expected to integrate these three aspects to sketch out the common

collaboration pattern for the GPF. I define this pattern as the behavioral template

or collaboration model that would implicitly guide the formation of the GPF as a

second-degree collaboration platform.

My perspective is that the traditional CFM view of two distinct sets of

actors with "insiders" (community members) and "outsiders" (NGO personnel) in

community projects is misleading. Instead, when these groups work together,

they form a functional community. Just as any other, this community has

hierarchies, boundaries, and functions. It exists parallel to and in conjunction

with the more "traditional" communities of place.

Researchers (P. H. C. Amaral, 2001; Stone, 2003) have pointed to the fact

that not all members of the community are involved in the timber projects. In the

same way, not everybody in the NGO or government agency participates in the

projects, either. The functional community (or community of practice) of the

timber project tends to be small and includes a diversity of individuals with

different backgrounds and roles. Understanding this functional community is








critical for evaluating the potential success of the proposed coalition. The figure

below represents this emergent community made in the intersection between the

land-based community association and the partner organization staff.


Partner Organization

\ Community ofpractice
(those involved in the
S timber project)





Association



Community (land based)


Figure 2.3: Timber Project Community of Practice

A useful metaphor is provided by the theory of communities of practice

developed for business management by organizational specialists (Wenger,

1988). Wenger notes that the idea of "communities of practice presents a theory

of learning that starts with this assumption: engagement in social practice is the

fundamental process by which we learn and so become who we are. The primary

unit of analysis is neither the individual nor social institutions but rather the

informal 'communities of practice" that people form as they pursue shared

enterprises overtime..." 10. The fundamental idea behind a community of

practice is that identity, trust and sense of belonging derive from the activities in


10 Text in the first page presentation (Wenger, 1988).








which an individual engages rather than from the formal organization (Figure 2.4,

below). If one looks at community forestry projects as such an activity, one can

study the communities that build around them.


(Communit)

Identity






Practice Learning
(Timberextraction, (Challenges, conflicts,
timber business) adaptationsinnovations)


Figure 2.4 Communities of Practice Model
I was interested in several aspects of such communities. The issue of

trust was of particular interest, as it is fundamental for any collaborative effort.

This is distinct from to the concept of trust in social capital research, where it is

identified and assessed for quantifiable purposes (Babbington & Carroll, 2003;

Dasgupta, 2002).

My interest was in the practice of collaboration, the activity that depends

upon and reproduces social capital. This is why my purpose was to delineate the

common pattern of collaboration by looking at the different organizations'

experience. To operationalize this concept I adapted a set of variables from

other studies on collaboration (Table 2.1) to the particularities of an emergent









coalition. I provide the definition of these variables and their analysis in the next

section.

Variables and analysis

Variables for the internal factors analysis are clustered at two levels: first,

the organizational level that includes participation, leadership and management

capacity of both the community associations (CA) and partner organizations (PO)

of each project, and secondly, the projects level, which includes analyzing the

relations among project partners (CA and PO), the degree to which the CA has

appropriated the timber operation, and the training focus used by the PO in each

project.

To assess participation I used the framework developed by Stone (2003),

which emphasizes the complexities of communities involved in timber projects

and establishes a gradient of participation according to a larger set of external

and internal conditions. My interest in the issue of community participation in the

timber projects is much more focused than Stone's. To operationalize

participation, I considered the number of individuals involved in the community

associations, and the assessment by both the CA leaders and the PO staff of

the quality of involvement of those participants in association activities.

Leadership was considered in both a material and a formal sense, since

not all leaders of the groups were heads of the organizations11. In the formal


11 Heifetz (1997) associates leadership with the idea of adaptive work, which is the learning
process required to approach the conflicts between people's values and the challenging reality.
Leadership is about learning new ways of behaving. Another useful idea of this author is the
notion of leading with-authority and leading without-authority. This refers to the fact that
sometimes the individuals guiding the adaptive work do not hold a formal leadership position, and
that at other times, the individuals in such positions avoid this type of work.









aspect, I wanted to look at changes in formal leadership positions. In a

community group, the lack of change in leadership might mean a lack of broader

participation and a concentration of power. Such conditions might cause the

NRM project to reinforce internal inequalities, which in turn would have an effect

on the long-term sustainability of the natural resource base (Baland et al., 2002;

Baland & Platteau, 2002). In contrast, the effect of frequent changes in the

partner organization leadership (NGO, government agency) is often negative

because it directly affects the execution of the project and prevents the

consolidation of trust among participants. Changes in project leadership usually

represent changes in technical orientation of the projects and in the way

community participation is understood and incorporated.

For management capacity, I concentrated on eliciting a qualitative

assessment of the activities conducted by every CA. By management capacity I

am referring to the activities and tasks that the different organizations are

capable of doing. This issue is more relevant for the CA than for the PO. I

wanted to identify the complexity of these activities, the duration, the number of

people involved, and the reasons why the activities were abandoned, if that was

the case.

At the project level, I began with a basic description that included duration,

objective, activities and results. I then concentrated on the key aspects of

relations, training and appropriation.

With regard to project relations, I first focused on analyzing conflicts within

groups. Conflicts can represent a defining event that solidifies a group's identity.









They can also create a negative environment that prevents any collaboration

from occurring. Conflicts between the partner organization and the community

group can also have a dual meaning. They can be the expression of the

community's agency as its members strive for greater control of the "NRM

project". Conflict can also be the expression of the individual agendas of leaders

reluctant to loosen their grip over the project resources. For the purpose of

assessing the potential for collaboration, the way in which conflict was managed

was more important than the conflict itself. Although it was not expected that

projects would have explicit mechanisms to deal with internal conflicts, conflictive

experiences should build member's skills and set organizational patterns. How

those skills and patterns constructed the organizational culture was of primary

interest in this analysis.

My interest in the projects' training focus was linked to the concept of

capacity development. This refers to the process of acquiring new organizational

skills that allow groups to increase the number and complexity of their activities.

A project's approach to training is a good indicator of the role social learning

plays. Approaches that are more explicit about social learning tend to foster

experimentation and encourage collaboration between the PO and CA

(Wondolleck & Yaffee, 2000).

Appropriation refers to the degree to which the PO had successfully

transferred the control and operation of the project tasks to the community-based

groups. Stone discussed this aspect as a dimension of community participation.

Since I was interested in delineating the common collaboration pattern, I looked









at this issue in terms of process and as a direct indicator of the collaboration

between POs and CAs in each project. In order to give greater emphasis to this, I

treated appropriation separately, and not as a part of participation.

To analyze these internal characteristics, I used matrices to contrast the

opinions and views of individuals from the groups involved. Graphic displays,

such as stakeholder and conceptual maps, facilitated the integration of

information across all the variables.

Once processed, the information from the needs assessment and the

collaborative practice analysis had to be integrated with the other components of

this framework. This integration needed to be matched to some reference state,

which defined successful collaboration and adequate conditions for coalition

establishment. Both the strong context dependency of CFC processes and the

paucity of literature on the subject limited the composition of such benchmarks

for this study. Since part of this analysis refers to intra-organizational dynamics

and conditions, assessment tools geared toward institutional evaluations of non-

profit organizations can also be used as benchmark references (Smutylo, 2001).

The Nature Conservancy's Institutional Self-Assessment provided a good

example (Devine et al., 2001).

External Factors

The objective of this component was to identify key external constraints to

the coalition and the main cross-scale interactions shaping its emergence.

Based on the CFC cases analyzed above in this chapter, coalitions appear to be

driven in a top-down manner by community-support groups (government agency,

NGOs), not the result of strict bottom-up community initiative. Whether this is









positive or negative would remain to be defined in each case, but undoubtedly

external agents, institutions and markets have a significant role in the shape and

purpose of CFCs.

Policies and legislation delineate the use and appropriation of resources

and also set the basic financial, fiscal, and structuring rules for social

organization. Regulations represent the hand of the government (federal and

state) over community activities. International agents such as foreign

governments became part of the External Factor arena by providing funding for

projects and markets for forest products. Figure 2.5 represents the array of

external agents and the diverse links that tied them to the community forestry

activity. This model is based on similar cross-scale, hierarchy-based

methodologies proposed to study NRM in Amazonia (Schmink & Wood, 1987;

Wood, 2003). In Chapter Four, I will complete this diagram with the actual

stakeholders from the GPF case.


Figure 2.5: CFC External Agents









Variables and analysis

I selected three categories for studying the external environment for the

GPF: policies and laws, markets, and stakeholders. I identified three variables

of law and policy that I found relevant for my case: land tenure, forest

management and social organization. Land tenure legislation defines the

appropriation of the natural resources. This is the key structuring element in the

resource use system. Forest management regulations define the resource use

process, and are the principal instrument to implement government conservation

policies. Finally, social organization laws and policies provide the framework for

collective action, setting economic and political limits to what community

associations can do.

Market analysis in this study was limited both conceptually and

operationally. Conceptually, I am referring exclusively to the market for

sustainably managed timber coming from the community projects. Operationally,

I have limited myself to the same dataset that I used for the rest of the study,

consisting mainly of the interviews with parties involved in the GPF initiative and

other organizations related to community forestry. Due to time constraints I did

not look into the actual financial market information. I constructed a vision of the

timber markets based on the interviews and the literature, which did not directly

address timber market conditions.

Asking the GPF-involved individuals from the PO to tell me who were the

key external players in the community forestry activity, I identified external









stakeholders12. I describe the role of the relevant stakeholders and diagram their

role in the GPF initiative using a stakeholder map.

Data for this component were collected through interviews. A specific

questionnaire was designed for the external stakeholders, and questions about

their role and the impact of policies and regulations were included in the

questionnaires for the POs and CAs. Information from documents such as laws,

policy papers and project proposals were also be examined. For the analysis,

each variable will be considered and presented separately (laws and policies,

markets, stakeholders). The interactions between scales will be inferred from the

interviews as well as from the analysis of the other factors (internal and design)

in the analytical framework.

The analysis of the external factors focused on how external variables

constrained the GPF initiative that was located at the scale below it. This

unidirectional effect was assumed to constitute the main link between the

external factors (laws, policies, markets, stakeholders) and the internal (timber

forest projects). I hypothesized that external factors encouraged the GPF

initiative (occurring at the internal level). Not only did I expect to find an influence

from the larger scale aspects to the lower scale, I expected this influence to be

positive for the formation of the GPF coalition.

This expectation was based on my interpretation of the socio-ecological

evolution of the state of Acre, based on Holling's (2002) model of the adaptive


12 This could be described as following a snowball-like methodology. I asked my interviewees
who else I should interview regarding community forestry and the GPF initiative that were not part
of the timber projects. A total of 15 individuals were mentioned, of which I interviewed 10.









cycle. According to my interpretation of this cycle for the case of Acre, the

situation in the late nineties and early twenty-first century was one that was very

open to innovations coming from the ground up. This was the case with the

RESEX that rose from a social movement proposal in the late eighties into a

nationally sanctioned and internationally acclaimed institution during the nineties.

I anticipated that the conditions that enabled the development of the RESEX

would still be present in 2002.


k
2.The Bust lethargic years
Asian Rubber 1910
4. Contesting Integration

/0r Md6-Th


Res 989

Rube Tapers


Pacific


Vulcanization 1844 State Dev.Plans

r 1.Starting point: the Boom 3.State Driven Integration


Figure 2.6: Historic Trajectory Modeled on the Adaptive Cycle.

Figure 2.6 displays the main events, as detailed in the timeline in Chapter

One, using the adaptive cycle diagram developed by C.S. Holling to describe the

trajectory of complex socio-ecological systems. The figure begins with the

invention of vulcanization, which made possible the commercial use of natural

rubber in industrial and commercial products. The subsequent industrialization of

the late nineteenth century (and the development of automobiles in particular)

increased the demand for rubber, bringing international attention to the Amazon.









This attention resulted in the booming of the local economies in Amazonia,

through the expansion of the rubber extraction operations. This expansion

relates to what Holling describes as the accumulation phase of the system; in the

case of Acre (and the rest of Amazonia) it lasted until the 1910s. At the end of

this decade, rubber plantations in East Asia took over the market, signaling the

beginning of the "bust" period in the Amazonian region's economy. Here the

system flipped downward, after the economic collapse, and remained in a

disorganized state until the late sixties when the military government began its

strong intervention to develop the Amazon. An important point to make here is

that the war between Brazil and Bolivia that resulted in the creation of Acre,

significant event as it was, did not represent enough of a perturbation to bring

about a change in the socio-ecological system's trajectory. The land use pattern,

as well as the economy, remained the same.

In the late sixties, under the influence of the government's policies, the

system began slowly to reorganize. However, the dramatic impact of these

policies on the western Amazonian landscape was not truly felt until the eighties.

By this time the political context (global, national and regional) was no longer

favorable for the military government's plans. Democracy had been re-instated,

international concern for the environment was growing, and local, grassroots

social movements were articulated around the protection of the forest and the

populations that used it (the rubber tappers). The intense cross-scale

interactions of this period were expressed in the form of social conflict for land.

State and international mediation to resolve this conflict resulted in the









emergence of innovative land tenure institutions such as the RESEX. The

innovations in land tenure opened an opportunity for a different land use pattern

and a new economy to emerge, bringing the system back to begin a new r-phase

in its cycle. This new land use type would have standing forest as the source of

a diversified economy of sustainable products and services. However, this

"opportunity" had yet to deliver substantial results.

The undertow of the cattle-centered development promoted by the national

government since the sixties was very strong. Infrastructural improvements

necessary for any kind of regional development seemed to reinforce the cattle

economy more positively than the forest one. The state government found itself

in a very difficult position: on the one hand, creating new infrastructure, while on

the other building a new institutional capacity capable of enforcing forest

economy policies. Whether the government succeeds and interest in cattle

production decreases, remains to be seen. It is at this point that the figure ends

- with the question about how the impacts of the road building plans will affect

the future trajectory of the next accumulation phase. This is the context in which

the community timber activities were taking place.

As with the internal component, the characteristics constituting an enabling

environment that would foster a successful coalition process have not yet been

defined. However, these characteristics can be extrapolated from successful

examples such as the Mexican "Plan Piloto Forestal" a state government strategy

that took advantage of the federally sanctioned Ejido common property system to









successfully promote community forestry enterprises. The state level policy

worked alongside other initiatives, like a long-term technical aid program.

Design Factors

This component of the research was based on the assumption stated by

Krishna that "the design of the institutions delivering public goods can influence

the level of social capital' (Krishna, 2003). In my case the institution was an

organization (a coalition) and I believed its design would influence the amount

and the quality of collaboration (i.e. social capital) among the involved parties.

Here, I was interested in how the way the coalition was designed and

implemented would affect collaboration among parties.

Powell (1990), in his study of network forms of organization, pointed out

that while "hierarchical organizations were the result of human design and

markets were the result of human action (not of design); 'network' forms of

organization (like coalitions) were the result of both action and design" (Powell,

1990). So far in this framework I have presented the "actions" that could

potentially shape the GPF coalition. This includes actions by both the internal

and external stakeholders, in a variety of categories: participation, leadership,

relationships, management, training, policies, laws, enforcement, and funding

among others. This third, and last, component of my analytical framework

studied the coalition's design, as the context in which these actions took place.

The design would affect the coalition, but not the laws policies, etc, which were

external.









Variables and analysis

Design, according to Powell (1990), refers to the way in which a network is

planned and made. Based on this definition I considered two dimensions of

coalition design: concept and process. Concept design refers to the idea, or

definition of the coalition, and considers mostly structural and strategic aspects

such as objective, goals, members and structure. Process design is the strategy

used to promote the coalition idea; this includes looking at the how the idea was

presented to the interested parties and how it evolved as it was appropriated (or

not) by those parties.

To identify the concept design elements in the GPF case, I used interview

responses and the few documents that had, at the time of my fieldwork, been

written about the GPF. Process design elements were collected in a similar way.

To analyze the information regarding concept, I developed an ideal model of the

GPF structure and contrasted this with another model based on its actual

functioning. While the ideal model was extracted from the documents and

answers given by the GPF promoters, the real model was built from my

participant observation of the functioning of the GPF and the critical assessment

provided by some of the interviewees, particularly those external stakeholders

familiar with the GPF proposal but not directly involved in the process. To present

the evolution of the GPF process used a timetable matrix.

I used an innovation system perspective to analyze the process design of

the GPF coalition. An innovation system perspective is a view of innovation

processes, which sees innovation as being produced by networks of actors that









co-evolve with the technologies (in this case social technologies) they generate.

Co-evolution occurs as a result of iterative experiential learning between the

actors involved. Successful innovation results from strong interactions and

knowledge flows within the network (Douthwaite et al., 2003).

In the context of innovation systems, the idea of a coalition encompasses

several things. First, it is an invention-an idea or set of ideas that sparks the

innovation process. Then it becomes an innovation process as the idea starts to

be socialized. An innovation process is driven by how the proposal serves

immediate problems confronted by parties. The second element is its legitimacy,

which is dependent upon the qualities of the proponents and the way they

promote the idea. The important aspect to remember is that the initial proposal

design and the success of its diffusion (or up-scaling) go hand in hand. This

requires that design be viewed both in its structural and procedural dimensions.

The guiding principle is that "in order to have up-scale success you need to

understand process" (Ashby, 2003).

New ideas about NRM organizations such as adaptive management

(Gunderson, 2003; Lee, 1993; Salafsky et al., 2001) and social learning (Buck et

al., 2002) provide a strong characterization of what an ideal coalition process

and structure should look like. The emphasis of these frameworks is on how the

organization's elements foster learning and deal with uncertainty and change in

their natural and social environment. Although learning and adapting abilities are

typical characteristics of successful coalition cases (Coughlin et al., 1999; Fisher









et al., 2001; Monge, 2001), purposefully designed cases for these characteristics

are rare. Nevertheless, different efforts on a variety of scales are being made.

As an example, the Biodiversity Support Program (BSP), a consortium of

conservation organizations, prepared a detailed guide to design adaptive and

learning projects based on field research (Margoluis & Salafsky, 1998). Similarly,

the Center for International Forestry Research (CIFOR) conducted research on

adaptive co-management of community forestry projects to profile such

characteristics (Cunha, 2002). One of the sites for the latter project served as a

case for my research, as well. I planned to adapt the CIFOR project-scale

guides to evaluate the GPF coalition ex-ante.

A more detailed set of assessment criteria was provided by Douthwaite

(2002) based on his guidelines for a learning-selection innovation model. These

criteria look at the relationship between innovation promoters and their target

adopters (Table 2.3).

Table 2.3:Elements of the Innovation Process
* Start with a plausible promise
* Find a product champion
* Keep it simple, stupid (KISS)
* Work with innovative and motivated partners
* Work with pilot site or sites where the need for innovation
is great
* Set up open and unbiased selection mechanisms
* Don't release the innovation too widely too soon
* Don't patent anything unless it is to stop attempts trying
to privatize technology
* Realize that culture makes a difference
* Know when to let go
(Douthwaite, 2002)









According to the author, a flexible initial design will facilitate the engagement

of the interested parties. As these parties go about changing and transforming

the initial idea, they begin to appropriate it. The level of appropriation will signal

the degree of success of the idea. Douthwaite also points to the fact that the

proposed innovation must demonstrate its usefulness upfront. It should solve an

immediate problem or provide a direct benefit to the adopters.

Integrating the Analysis

In order to study the factors that conditioned the successful

emergence of a coalition I laid out a three-layered framework based on a

hierarchical view of the coalition phenomenon (see Table 2.2). On a higher level,

I clustered the institutions and policies that constrained the community timber

management project. The focus here was on the specific top-down links (laws,

policies, etc...) conditioning collaboration among projects stakeholders. In the

middle of this system, I placed the study of the groups and functional

communities that were interested in becoming part of the proposed coalition. I

drew specific attention to characterizing the collaborative practices occurring

within and between these groups. Below, I placed the analysis of the coalition

itself; looking at the conceptual proposal and actual process, I considered the

structural and functional design of the GPF and its early evolution stages.

Bringing the information together to form an adequate picture of the

situation of a coalition required considering the cross-relations between the

factors. For this purpose, Table 2.4 presents a grid that brings together the

organized variables by category, along with a set of ideal conditions or

benchmarks to identify the quality of each category of variables. A final column










should include the findings from the particular case analyzed. For this chapter I

have left this last column empty, to be filled in the conclusion chapter with the

data from the fieldwork.

The matrix tool, once completed with the information from the studied

case (or cases), was meant to facilitate the analysis of relationships between the

different components. The variety of these relationships depends on the type

and number of cases available, as well as the quality of the information available.

Table 2.4: Integrating Matrix
Factors Case
Categories (Variables) Ideal Condition or Benchmark CaStudied
At the Group level: Need for a coalition should be explicit among
o Participation group members
o Leadership Time should be available to participate in coalition
o Management Capacity building
Trust should be strong among members and
Internal: partners.
At the Project Level: Projects should foster social learning
o Partner relationships Projects should encourage appropriation by the
o Capacity development community
o Appropriation by Conflict management capacity should be a
community developed skill
State and federal government Laws and policies should secure land and resource
agencies, policies and laws on use rights for the communities
conservation, forestry Credit and technical support should be available for
External: development and social the community enterprises
organization Stable long term relations should exist between
NGO and Aid agencies communities and NGOs and government agencies
Forest products markets
Coalition Design: Structural design should be flexible and open to
o Objectives change
o Structure Initial proposal should provide room for adaptation
o Leadership by the potential members
o Resources
Design: Innovation Process Process design should provide room for adaptation
o History by the potential members
o Perceptions Process should have an explicit learning and
o Adaptations reflection strategy with built-in mechanisms for
monitoring and evaluation


Although some of the relationships will not be clear until the matrix is

completed, it is important to point to some foreseeable links, starting with the

comparison between the ideal condition and the information from the case study.









To clarify, ideal conditions (column three, Table 2.4) are presented in a synthetic

form, using key qualifying words by which they should be observed.

The comparison between the ideal and the real conditions required

attention to the context and history of each coalition. Although some general

stages can be identified in the evolution of a coalition (Edwards et al., 2001), not

all cases follow the same order; therefore they cannot be associated with the

same set of characteristics. Once this is considered, a more limited comparison

can still provide valuable information.

Another important relation to keep in mind is between aspects of

coalitions that bring together variables from different categories and at different

scales. An example could be the role inequality (or equality) could play in

shaping the proposed coalition. In order to assess this question, different scales

of action must be analyzed: intra-community; intra-organization; between

community and organizations; between the "project champions" and those more

passive or skeptical; along gender lines.

Although inequality might be fostered through structural aspects such as

laws, policies and finances, at a coalition level it is an issue less to do with

structure and more to do with process. There is no coalition structure that would

guarantee full participation or equity. These are issues that need to be overseen

constantly, through monitoring and evaluation. The goal is to aim constantly for

more equity and participation. Just how process and structure should respond to

existing inequalities is contingent upon the particular case.









One of the goals of the analysis was to provide feedback on an on-going

process with the analysis results. I was examining the conditions for the

emergence of a coalition in a range of time when that proposal had already been

discussed and some actions have been taken to build it. At that moment existing

proposed structures of the group needed to be tested, along with the

programmatic aspects such as objectives and activities.

The information about external and internal factors should help to

establish how the proposed operative design could be adapted to deliver the

purpose of the coalition. The adequacy of the current design can be assessed

based on two elements: the efficiency of the design (and the coalition idea in

general) to satisfy concrete needs of the participant groups, and the adaptive

capacity (resilience to unexpected change) of the current process-structure

design.

The coalition's purpose has to respond to the concrete reality of the

participants, solving relevant problems and providing direct tangible benefits.

While demonstrating that it is efficient in reaching its purpose, its structure has to

be flexible, adaptive, and open to change. One way to achieve such flexibility is

by purposefully leaving things undefined. These aspects will be filled out in the

process of putting the original idea into action. This could include things like

leaving open the type of activities the coalition could assume or testing different

decision-making structures. The valuable thing is the awareness with which

those things are left undefined. This facilitates the social learning process and









the accounting of these "loopholes" as strategic decisions in an evaluation

process.

In an organic process, unrecognized loopholes will require adaptive

responses as well, but the process of generating those adaptations could be very

costly. The cost might come in the form of internal conflicts among participants

or as unbudgeted issues that distort available funding. Personal and group

emotional conditions harnessing uncertainty and surprise play a key role in how

adaptations are generated. Under a rigid coalition structure an unexpected

event may cause conflict and undermine collaboration. Conflict dynamics

normally include different degrees of blaming and scape-goating that, depending

upon the collective emotional resilience, could destroy the coalition or severely

hamper reciprocal trust.

Under a negative emotional environment structural, causes of problems

are overshadowed by personalized accusations or explanations. An issue such

as the rigidity of decision-making structures is hard to bring up in the middle of a

polarized dispute. Even in the aftermath of conflicts, the opportunity to learn is

missed because the group might not have the adequate "social learning

platforms"(Buck et al., 2002) to process it. In a coalition, learning platforms and

emotional awareness are essential, given that the added value of participating in

a coalition should be its efficiency to foster dialog, joint problem solving and

interactions.

An open design also allows appropriation to take place. As with

technological innovations, part of the success comes from participation. The









greater chance interested parties have to define what the coalition becomes, the

more successful the collaborative proposal will be.

Analytical Framework Limitations

Thus far, I have identified three major limitations of this analytical

framework. First, biophysical aspects were mentioned but not developed into a

full variable. Second, market analysis was considered in very narrow terms. And

third, technical forestry and timber management's impact on the organizational

behavior were only indirectly contemplated.

Despite the need to incorporate the study of biophysical aspects in the

assessment of external conditions to the emergence of the coalition, I was

unable to develop this variable in the analysis. I was not able to identify which

biophysical aspects and at what scales needed to be incorporated. Literature

analyzing collaborative efforts from an environmental conflict management

perspective presented the same void (Brick et al., 2001; Wondolleck & Yaffee,

2000). Even among literature on community conservation and conservation

project development the role of biophysical aspects in organizational

development has not been well documented (Borrini-Feyerabend et al., 2001;

Clay, 2001). I reproduced this handicap without shedding light onto what could

be the potential bridges towards this connection between ecological environment

and scale of social organization. Indirectly it is clear that there is a relationship,

evident in the current trend to shift from a site-based NRM to an ecosystem- or

landscape-scale management. Establishing this connection extends beyond

the capacity of this study.









The effect of market dynamics on the potential success of a CFC is very

clear. Community forestry activities are mostly oriented towards selling forest

products to local, regional or international markets. Nevertheless the

methodology of this framework does not explore the correspondence between

the characteristics of the market and those of the organizations. Empirically, it is

known that part of the motivation to collaborate among community forestry

enterprises comes from the need to take advantage of the market benefits of

belonging to a larger group. The methodology's emphasis on process issues

and interpersonal relations overlooks the role of material conditions such as

markets. The problem of incorporating market factors adequately is

methodological: what information to retrieve? How to relate to the emergence of

the coalition? How does its relation to the market affect the coalition?

The degree to which the analytical framework is tailored towards community

forestry was given by the literature review. Literature on community forestry

organizations establishes some explicit links between issues such as property

rights and type of product extracted from the forest. Nevertheless, the most

important aspect defining structure of organization is function, and function is

defined by the characteristics of the activity undertaken. The way an NTFP

enterprise is conducted is substantially different from a timber project-the

structure of organization should be as well. I would assume that those

differences in operation and organization would affect the way a CFC would be

organized if it were composed of NTFP or of timber projects. However the

framework does not account explicitly for these differences. I expected that the









management activity's effect on the organization would come from the data, but

that may or may not be the case-the framework does not target this issue

directly.

Synthesis

This chapter began by characterizing the concept of community forestry

coalitions. It also presents the analytical framework built to analyze the case of

an emerging coalition: the GPF in Acre and Rond6nia. The analytical framework

was presented in terms of the factors within three components: internal, external

and design. The variables of each component were then described, as were the

analytical tools used. A final section presented the integration of the research

framework. It is through the integration of the research components that the

research hypothesis on the hampering condition of the internal factors will be

tested. It will provide a complete vision of the constraints and opportunities for

the GPF coalition success. The next three chapters present the results from

each of the components. The first to be addressed are the internal factors

affecting the GPF initiative.














CHAPTER 3
INTERNAL FACTORS

As a first step to map the conditions for inter-project collaboration among

the community timber projects in Acre and Rond6nia, I started out by identifying

the characteristics of intra-project collaboration. I was interested in identifying the

different groups' collaborative behaviors, as these were critical to identify the

common pattern of collaboration that would inform the GPF initiative. I had stated

in my hypothesis that internal factors would be the source of the major

constraints for the development of the GPF, while the external (chapter 4) and

design factors (chapter 5) would be positive for the emergence of the coalition.

I used the theory of communities of practice as a metaphor to guide this

assessment. I considered each community-NGO (or government agency) as an

actual community-of-practice of collaboration, given that this was the type of

activity I was interested in studying. Through this theoretical framework, I chose

to see each project as a particular system, defined by its own experience of

collaboration.

I evaluated the effectiveness of each case, and also the similarities among

them, to assess the role that the collaboration experience would have in the

success of the GPF. To do this, I interviewed individuals from all organizations

involved in the timber projects: sixteen from the four community groups and

fourteen from the partner organizations. Based on my analytical framework I

focused the interviews on six variables: leadership, participation, management









capacity, project relationships, project training focus and project appropriation by

the community. While the first three considered each organization as a whole

and included its involvement in the timber projects, the second three focused

exclusively on their experience with the timber projects.

The results from this chapter confirmed my hypothesis about the weak

management capacity of the community associations. The complexity of the

timber operations seemed to overwhelm the community participants, and their

learning curve seemed to be longer than was expected in the projects' original

plans. The external pressure to show results (particularly profits) led the partner

organizations to maintain a stronger control over the operation, creating a

negative paternalistic-bias in the projects' collaboration pattern. However a

conscious social learning process in some of the projects was leading them to

experiment with broader community participation in the period from 2001 to 2002

when this research was conducted.

This chapter is organized in three larger sections. The first describes the

stakeholders involved in the GPF. The second presents a comparison of the four

projects' organizations on the set of key variables (leadership, participation,

management capacity, project relationships, project training focus and project

appropriation). The last section discusses the results and synthesizes the main

findings of the chapter.

The GPF Coalition Members

The timber project community interested in participating in the GPF initiative

consisted of four projects: Porto Dias, Peixoto and Cachoeira from the state of

Acre, and Cautario from the state of Rond6nia. The Porto Dias project involved









The Center for Amazonian Workers (CTA, Centro dos Trabalhadores da

Amazonia) and the Association of Rubber Tappers of Porto Dias (ASPD:

Associagdo de Seringueiros do Porto Dias). The Peixoto project was managed

by EMBRAPA, Brazilian Agricultural Research Corporation, along with the

association of timber managers of Pedro Peixoto (APRUMA). In Cachoeira the

local residents association (AMPPAE-CM or AMC) had been supported by two

state government agencies: SEFE (State Secretary of Forestry and Extractivism)

and SEATER (State Secretary of Rural Extension). In this case the local group

also had a close relationship with two timber processing companies, the Polo

Moveleiro and AVER. Cautario, the timber project in Rond6nia, was a joint effort

of ECOPORE, a local NGO, and the AGUAPE rubber tappers association. In this

case the Rubber Tappers Organization of Rond6nia (OSR) also played an

important role. I will describe these projects and organizations in more detail in

the following sections.

A synthesis of the potential GPF members is presented below in the form of

a stakeholder map (Figure 3.1). It important to appreciate that the GPF was an

initiative to build a platform of interaction among four different community timber

projects, but these projects were in themselves collaborative initiatives that each

involved at least two very different organizations. Every case included a

grassroots community association and also a set of partner organizations

providing technical assistance. As will be discussed later, these support

organizations had a large stake in these projects and exercised a high level of

control over them. Although projects were presented as "community" timber









projects, and the GPF was presented as a "community coalition", the partner

organizations actually had much initiative and decision making power. To

represent this situation, figure 3.1 presents each project with both their local and

technical partner, and the GPF coalition as an initiative that would bring these

different partnerships into a broader "umbrella organization".

Community Forestry Community Forestry
Projects Organization

Support NGO or Community-based
Gov Agency Organization











lachoeira Cautar*


Figure 3.1: Stakeholder Map

Following in this section are the profiles from each of these potential

member-projects. Each profile presents the timber project and the main partner

organizations that teamed up for each project. The profile describes briefly the

history of the project, its basic information (type, years, finance, administration)

and the significant traits that each project (and each organization) had in terms of

leadership, collaboration and management experience. The following section

aggregates the information from the four cases to establish the baseline

collaboration model on which the GPF would potentially be developed.









Porto Dias Project

Port Dias was the first community forestry project to be established in the

region. It was started in 1996 with the financial support of PPG7 and ITTO13. The

PAE Porto Dias was an INCRA agro-extractivist project created in 1989 with an

area of 22,145 Ha; approximately ninety families live in the settlement. The

timber project aimed at developing timber extraction as an economic alternative

for rubber-tapper families in the PAE Porto Dias. It was established as a

groundbreaking project that could later be replicated in other PAEs and

RESEX14. The idea for this project came from CTA and had already been

rejected in other rubber tapper areas because of the negative perception the

social movement had about timber extraction. The rubber tappers' movement

had fought to save the forest and cutting trees seemed an aberrant

contradiction.

The success in getting the Porto Dias Association to agree to the project

came from various internal factors. First, Porto Dias had not been an area of

strong conflict over land during the eighties. Therefore it lacked the strong social

movement background presented in other areas such as the Alto Acre Region.

This fact made people more open to trying timber harvesting as an alternative

activity. At the time CTA approached Porto Dias with the idea, economic

hardships had caused some families to abandon the PAE, and also more

13 PPG7 is a fund for the conservation of forest in the Amazon basin established by the countries
of the G7 group (the 7 most industrialized countries in the world). The Fund is administered by
The World Bank and IBAMA (Instituto Brasilero de Meio Ambiente). ITTO is the International
Tropical Timber Organization.
14 A second community project supported by CTA in SAo Luis do Remanso was certified for
timber and non-timber products by the FSC in 2004









agriculturally oriented families had come in. This last fact plus the recurrent

illegal logging that went on had brought back the discussion of whether this PAE

should be transformed into a regular agrarian settlement.

Given the institutional abandonment in which INCRA had left the PAE after

its creation, the timber project not only represented an economic alternative but,

in the short term, it represented an apparently strong external institution

supporting rubber tapper families and their way of life. This strengthened the

Rubber Tapper Association and their interest in maintaining the extractivist

category for this settlement.

The timber project had its basis in the reduced impact logging developed by

IMAZON and the Tropical Forest Foundation (FFT) in Eastern Amazonia, with a

cutting cycle of 30 years and very low volume extracted per hectare (1 to 2

trees)(Capossoli, 1999; Kornexl, 1997). This technical approach was

accompanied by a strong capacity building component that aimed at eventually

transferring the whole operation to the community. Each community participant,

known as a "manejador", had to sign, and commit to, a letter of principles that

included: keeping their kids in school, participating in training and applying what

was learned, willingness to work in other people's areas, and agreeing that part

of the profits should go the association. To guarantee this approach, as well as

to take advantage of potential market benefits, the project was submitted for

sustainable forestry certification (FSC) and received its seal in 2002.









As early as 1997 a project financial evaluation (Kornexl, 1997) done by the

CNPT15 found that the financial and economic viability of the project was low.

The high initial investments as well as the "pre-investments" (government

infrastructure) required made the project a bad investment when compared to the

establishment of a Brazil nut processing plant in the site. The evaluator

nevertheless supported the project given its innovative characteristics and the

potential model that it represented for the rest of the Amazon region and its rural

communities. The evaluation assumed a potentially fully implemented project,

and did not consider the cost of bringing things from their status quo to a state of

full project implementation. It did remark that the costs to move from the initial

conditions were significantly high, but it did not quantify these costs. Community

organization skills as well as business skills were underlined as areas that

needed significant attention in order to have the project running. This

evaluation, as well as the project's proposal, relied mostly on training (conducting

workshops) to develop these managerial capacities.

Centro de Trabalhadores da Amaz6nia

Founded in 1981 as a grassroots, adult education organization to provide

support to the rubber tappers social movement on health and education issues,

the CTA turned its attention to timber management in 1996 with the first

community timber project in Amazonia (Porto Dias). This change came as a

result of the need to develop economic alternatives for the rubber tapper families

living in the recently created RESEXs and PAEs. The multiple forest use project

15 Conselho Nacional de Povos Tradicionais, the subdivision within IBAMA in charge of the
RESEX









was designed with a strong forestry focus concerned primarily with assessing the

forest timber potential and preparing the legally required documents such as

"pianos de manejo" (management plans to be approved by IBAMA).

At the time of the design and initial implementation of this multiple forest

use project, CTA was still a small organization led by forest engineers who

decided to get the project running as soon as possible. Although the project

proposal called for strong community participation throughout the whole process,

the complexities associated with the timber activity, and the limited capacity of

the community to get involved, left much of the control and execution in the

hands of CTA.

This established a noticeably vertical relationship between CTA and the

community project participants. A few interviewees mentioned that the strong

leadership style of the initial project coordinator resembled the behavior of the old

patron (seringalista) toward whom the manejadores (project participants) reacted

very passively. During this time the project confronted several external

challenges such as the lack of IBAMA's approval, which took over a year to

obtain leaving the project on hold. IBAMA had not differentiated between small

and large timber projects and it was not until 1998 that it established the small-

scale timber project guidelines. The negative perception of timber management

within the rubber tapper movement also affected the project. As a result of

changes in coordination, and significant internal conflicts, the initial technical and

paternalistic approach later changed to be more community development

oriented.









At the time of this study, CTA's technical team comprised a cohesive group

of forest engineers who shared a similar view of the project. They were close

friends beyond their work activities. They had gone to college together and came

to Acre at almost the same time, around the year 2000.

Based on a commitment to improve the rubber tappers' living conditions, to

the pioneering work developing economic alternatives such as innovative NTFPs

and community timber management, CTA was a well-respected organization.

Beyond the Porto Dias project work, they were involved in a variety of fora,

including the Amazonian Community Forestry Working Group, the "Grupo de

Trabalho Amazonico"(GTA) a consortium of NGOs and other grassroots

groups following Amazonian development policies.

Associagao de Seringueiros de Porto Dias

The Associagdo de Seringueiros de Porto Dias (association of rubber

tappers of Porto Dias) was established in 1987 with the purpose of consolidating

the Porto Dias agro-extractive settlement (PAE). The association bylaws

established a series of objectives for the group among them: the promotion of

collective action (associativismo), commercialization of the member's products,

and conservation of natural resources in the PAE. Membership was restricted to

rubber tappers that lived inside PAE Porto Dias. The association had a general

assembly that met annually, and a board composed of a president, a secretary, a

treasurer and three overseers (fiscales), who were all elected for a period of two

years.

Among the group's key activities was the management of a community

store that received local products, such as rubber and Brazil nuts, from the









members in exchange for house goods and food. The treasurer of the

association managed the store. This store reported small gains for the

association during the first half of 2002. The group was also in charge of several

oxen and mules used to transport rubber, Brazil nuts and timber.

It is important to note that only ten members of the association directly

participated in the timber project. This group included most of the members from

the association's board. As discussed by Stone (2003) the range of local

participation in the timber project varied widely. Of the ninety families reported to

be living in the settlement (Cunha, 2002), the association had only 24 members16

(typically male heads of households, that supposedly represented their families).

Of this number, only ten association members participated in the timber project.

In the six years of the project, a few families left the project. However, not until

late 2002 were there discussions regarding bringing new families into the project.

Peixoto Project

The Peixoto timber project is located in the southeastern corner of the

state of Acre between the municipalities of Acrelandia and Senador Guiomard.

Pedro Peixoto was one of the largest agrarian development projects in Brazil,

with an extension of 378,395 ha and a total of 3000 families. The initial area of

the timber project covered 440 ha, but with the addition of new families, this area

expanded to 1000 ha. The participants of the projects were mostly farmers;

some were originally rubber tappers, but most migrated from the south of the

country.

16 The number of members varied slightly between 21 and 24 members. The association
treasurer and the secretary mentioned twenty-four during my summer 2002 interview.









The project was designed and managed by EMBRAPA, the Brazilian

Agricultural Research Corporation, to work within the forested area of each

family's lot. According to the Brazilian Forest Code each landowner in the

Amazon must leave an area of 80% under forest cover. This limitation

represented a problem for the small settler because it significantly reduced the

land they could use for agriculture. The timber project's objective was to develop

sustainable logging as an economic alternative for these forest reserve areas.

The strategy was to identify appropriate low impact practices, and technologies

that could be adopted by the farmers without requiring large initial investments.

The project did not aim at transforming the farmer into a timber-manager, since

the potential timber to be exploited from the small 40 ha family lots was very

small. The idea was to complement the families' income with an activity that

could be conducted with available human resources.

This project was also designed with a forestry research focus, with a

control area and a well-established system for monitoring activities and

ecological responses in the treatment plots (initially ten). Financial studies were

conducted, which estimated the average income increase for the participant

families at 700 Reais per year (about $ 250). As part of the project, EMBRAPA

provided training for the participants in tree inventories, cutting and transporting

timber, and equipment maintenance. An association of participant families was

created in order to transfer some of the equipment (chainsaws, truck, etc) and to

facilitate the creation of a platform for marketing products.









After seven years the project had increased the number of participant

families to 24 and had trained several teams to conduct the full logging operation

from inventory to transportation. Some families had incorporated timber

production into their activity calendar and relied on this activity for a significant

part of their income. The association (APRUMA) had achieved a degree of

autonomy and management capacity. However, the project was also challenged

by some problems. On the economic side, the small scale of the project

threatened its sustainability. The project provided all the necessary monetary

pre-investment for the farmers, including the equipment and the legal

authorization, yet as EMBRAPA prepared to withdraw, it was uncertain who

would cover these costs.

Timber sales also represented a problematic dimension of this project. To

begin with, the quality of the timber available was very low due, among other

reasons, to pre-project illegal logging exploitation. Timber industries were

interested in larger volumes that compensated for the high transportation costs;

therefore individual families offering 40m3 of timber per harvest had little chance

of getting a significant price. Although the APRUMA represented a platform to

sell the harvest jointly, the cost of organizing and maintaining an association was

very high in terms of time and organizational skills. The project required a

professional forester to oversee the operation. EMBRAPA had provided these

services so far. How would the costs of a forester be covered? Certifying the

timber production through the FSC was sought as a means to guarantee a higher









price; however, who could pay for the costs of maintaining this certification was

not clear.

Empresa Brasiliera de Pesquisa Agricola

The Brazilian National Agricultural Research Agency was created in 1973

to provide scientific support for extensive agriculture and cattle ranching. Forest

management became part of EMPRAPA Acre's agenda only a few years before

this study. The Acre research station was one of thirty-two distributed around

the country. This station focused on providing economic development for the

ecosystem and the agro-industrial complex in the state. Sustainable forest

management and silviculture were specific areas of action.

Under this mandate, the Peixoto project was a prominent initiative to

promote low impact, sustainable forest management in family agricultural areas.

As a typical EMBRAPA research project, its financial support came both from the

agency and from international organizations (ITTO17). The project had a principal

investigator who had led the research since the very beginning. Project

leadership appeared to be concentrated in this person as he made most

administrative and strategic decisions for the project. He was responsible for the

results of the project as well as for the administration of the funds and the

personnel. He was also the contact person with the community, the person who

initially proposed the idea and under whose leadership the farmers had agreed to

participate.


17 International Tropical Timber Association









As with all projects dealing with forest management, it had both a

technical-ecological dimension that looked at environmental impacts and

technological approaches used, as well as an economic dimension that related to

the viability of the project. Research results had to integrate these two aspects.

However, the successes of this project were more in the technical and ecological

than in the economic aspects. Beyond the simple economic viability of the

project, staff recognized the need for further training of the participants in issues

dealing with the business aspects of timber. Also, the imminent withdrawal of

EMPRAPA from the project would require APRUMA, the community association,

to be able to take over the leadership of the project, raising its organizational

capacity. So far this association had passively submitted to the vertical

management style of the research project, showing very little self-management

capacity. This was recognized by the research team at EMPRAPA as a serious

limitation for the future of the project.

Association of Rural Producers of Peixoto

The association of rural producers of the Peixoto agrarian development

project (APRUMA) was created in 1998 to provide a platform to group the

participants in the timber project. EMBRAPA was able to transfer some of the

project equipment to APRUMA, especially the larger timber processing plant and

the truck, which were meant to help the farmers add value to the products they

were offering. This association would also help organize the collective work that

the timber operation required, since it had to move from one participant's plot to

the next. The association started with eight founders and, as noted above, had

twenty-four active members by mid-2002.









The association required that 2% of the income generated by each

participant from the sale of timber go to the group. The group met every two

months according to the project schedule set by EMBRAPA. A truck from

EMBRAPA picked everybody up and later brought them back after the meeting.

The group had a board of directors comprising 14 members, with a president, a

secretary, a treasurer, two vice-presidents and 8 overseers (fiscales). The group

had all its legal documents updated and kept a well-organized set of books

regarding activities and financial matters. The group had had two presidents

since their creation. Recently a crisis had challenged the organization as one

member was expelled after failing to contribute the 2% from the timber sale, that

corresponded to the association. Despite their very positive relationship with

EMBRAPA, the transfer of the project's equipment was seen as a difficult

process. The maintenance of this machinery would demand more collective

responsibility than the group had developed so far.

Cachoeira Project

The Cachoeira community forestry project was located in the municipality of

Epitaciolandia in the upper Acre river valley, on the border with Bolivia. It was

based in the Chico Mendes agro-extractivist settlement also known as PAE

Cachoeira. This PAE covered an area of 24,898ha. It had 68 families that lived

mainly on subsistence agriculture and the sale of Brazil nut and rubber. The

closest city was Xapuri, the center of the rubber tapper social movement during

the eighties. The area had a long history of political engagement and collective

action. Many of the families that lived in Cachoeira had participated in the









empates18 to stop the forest from being cut by the ranchers. As a result of that

struggle, the PAE was created by INCRA through the acquisition of the area and

it was reassigned as a place for rubber tapper families.

The timber project in Cachoeira originated as part of the initiative to

establish a wood products industrial park in Xapuri. The "Polo Moveleiro," as it

was known, was a joint endeavor of the Catholic Church and the Municipality of

Xapuri with strong support from the area's state congressman.19 The idea was

to establish a furniture-making school and open a manufacturing site in the city

for wood-related companies. Italian groups linked to the Catholic Church directly

supported the furniture-making school. The timber project in Cachoeira was to

provide the raw materials for these industries.

The mayor of Xapuri and the local state congressman had strong political

and personal connections to the community of Cachoeira. These key players

facilitated the introduction of the idea, although its approval by the community

was a gradual process. CTA was initially brought in to start the project by

establishing the initial forest inventories, but they were later replaced by staff

from the recently created SEFE (State Forestry and Extractivism Secretary)20.

The timber project was directly managed by the Associagdo de Produtores

e Moradores do PAE Chico Mendes (The Association of Producers and

18 During the eighties, empates were peaceful demonstrations organized by the rubber tappers
who physically put themselves between the forest, and the forest clearing machinery and workers
brought in by cattle ranchers to clear the forest for pastures.
19 Ronaldo Polanco, from the Partido dos Trabalhiadores (PT).

20 In 1999 the PT won the state government election. Their platform, as discussed later in this
chapter, was to center Acre's economy in the valuing of forest products such as rubber, Brazil nut
and timber. As part of this effort SEFE was created to conduct that policy.









Neighbors of the Chico Mendes PAE). A total of nine families from the

association participated. The project took place in each family's colocagdo

(traditional family subdivision within the PAE). The project was based on the

idea of "Jardinagem Florestal" developed by Virgilio Viana, a professor from the

University of Sao Paulo (and later Secretary for the Environment for the state of

Amazonas) who initially acted as the scientific supervisor for this timber project.

Jardinagem Florestal is a low impact logging strategy that used a simple

language to facilitate its adoption by community members. For example, forest

reproduction was presented by using the metaphor of "mae, filha e neta:" mother

plant, daughter plant and granddaughter plant.

The project used animal traction to transport the logs within the forest. A

truck provided by the municipality of Xapuri later picked these up. Most of the

wood was sold to the Polo Moveleiro in particular to the AVER company, a

designer furniture company from Sao Paulo. In contrast to the other timber

projects in Acre, Cachoeira benefited from having a secure buyer for their

product. The income from the wood went directly to the family from whose land

the timber was harvested, although 10% of the income went to the association.

This 10% was divided in two: half for education and health care, and half for the

project itself.

In 2002 SEFE left Cachoeira, and the State Secretary of Extension

(SEATER) took over support of the project's forester. Throughout its

development the project had counted on several sources of funding, routed

through the State Government, among them ITTO, IDB and the PPG7 fund.









The timber initiative in Cachoeira confronted several challenges in the four

years between 2000 and 2004. One of these challenges was the enforcement of

the management rules established jointly by SEFE and the community

participants. Internal control systems were weak and monitoring was not

followed up. Another challenge was the demand from other community members

to participate in the timber endeavor. The process of including new families

required thinking about introducing new buyers to take up the increase in

production.

This timber project saw its technical staff change every two years as a

result of conflicts among the staff, the community and the scientific coordinator.

Changing the staff created a lag in the operation, which affected production.

Formally, hiring new staff or bringing in new institutions solved these conflicts.

Associagao de Moradores e Produtores do PAE Chico Mendes

The association of residents and producers of PAE Chico Mendes was

founded in 1995 with the support of the local state congressman to help attract

projects to the PAE. Fifty families were associated with the group. The

association represented the community at the municipal and state level and was

in charge of several other projects besides the timber initiative. According to

Stone (2003) the group had strong support from the community. The group had

what are known as "delegados sindicais" (union delegates) that represented the

association in the furthest corners of the PAE. They acted as liaison officers of

the association and at the same time solved local conflicts and reported on

residents' needs.









The group had a long history of relationships with external organizations,

given the active role the community had during the 1980s social movements. At

the time of the research the association had strong ties to the Xapuri municipal

government and the State government of Acre both from the "PT' workers

party. The Association was also a part of the Extractivist Cooperative, CAEX,

based in Xapuri and the National Rubber Tapper Council (CNS).

As noted earlier, nine members of the association participated in the

timber project. This group had a lot more interaction than the rest of the

members. They participated in the project's decision making as well as in training

workshops and community forestry meetings at the state level. After SEFE left

in 2002 the Association took control over the complete timber operation from

acquiring the permits to selling the wood. Progressively they used the income to

cover the cost of equipment maintenance and other associated costs like

transportation and technical assistance.

Divisions had started to appear within the community, especially between

those living close to the center of the community and those far away. This

division also corresponded to kinship differences. In 2002 a group of association

members, together with other non-members living in the far south corner of the

PAE, began forming another association. One of their proposals was to start a

new timber project.

Cachoeira Project's External Partners

A number of organizations had played the role of external partners in

Cachoeira. Among them were: CTA, the NGO involved in the Porto Dias

Project; SEFE, the State Secretary of Forestry and Extractivism and SEATER,









the State Secretary of Rural Extension. The role of the latter two was limited

when compared to the way EMBRAPA or CTA had participated in their

respective projects. The strength of the local association, and the particular

origin of the project's link to the Polo Moveleiro project, were identified as the

explanation for the lesser role of the external partners (Stone, 2003).

Certainly the history of collective action and political activism placed

Cachoeira on a different level than other rubber-tapper communities. The social

and political capital were strong enough to suggest that their relationship with

external organizations in a development project would be more balanced. Their

capacity to organize themselves and state their demands required a more

horizontal approach from the external agents. This was recognized by SEFE

staff who were involved in the project between 2000 and 2002. However, the

origin of the timber project in this case was just as foreign to the community as in

the other cases studied. An external agent came in and proposed the idea,

which was accepted by the rubber-tappers after a period of doubt and debate.

Just as in the other cases, the project was limited to a small number of

association members, most of whom sat on the board of directors and shared

other connections among themselves (e.g. kinship and friendship).

In Cachoeira the project was proposed by two political figures from outside

the community: the mayor of Xapuri and the local state congressman. A third

proponent of the project was Virgilio Viana, the university professor and later

Environment Secretary of the state of Amazonas who served as scientific

supervisor of the project. Although these three people did not constitute an









organization in the way EMBRAPA or CTA did, they played a similar role to these

organizations but in a personal capacity. These figures were sought after

whenever the project had any type of crisis. The scientific supervisor frequently

visited the project and played a role in the few project crises regarding technical

assistance.

Cachoeira's timber project involved a very small group of people from the

community. This minority not only benefited from this project, they were also the

same group that implemented most of the other community initiatives. There

were friendship and kinship ties among them, too. The association had had only

two presidents since its creation and the rest of the members of the board had

rotated more than they had actually changed. Even within this small group of

people, the actual leaders of the group were a few individuals. These individuals

controlled access to information and also represented the community in relation

to external agents. This was a similar situation to that of the other community

timber projects. The difference in Cachoeira seemed to lie in the capacity and

vision of this very small group of leaders, particularly to maintain decision-making

control over what went on in the community.

Formally, this group of individuals had progressively gained more control

over the timber project, making the political role and power of the external

founders of the project (the state congressman and particularly the scientific

supervisor) very difficult to assess. Yet the conflicts discussed above showed

evidence of the strong role they still played. It was hard to determine the

intensity or extent of that influence, because the nature of the relationship









between the external founders and the community leaders was more personal

than professional. Most of the exchange that went on was hard to follow. It was

uncertain whether the community leaders had the same capacity to challenge

these external players as they showed with SEFE or CTA.

Cautario Project

The Cautario community timber project was a joint venture of the NGO

Ecopore; the Organization of Rubber-Tappers from Rond6nia, OSR; and the

Association of Rubber Tappers of the Valley of Guapore, AGUAPE. The project

was located in the Rio Cautario Extractive Reserve. This was a state sanctioned

extractive reserve as opposed to the federal reserves like the ones in the state of

Acre. The Cautario RESEX was located in the municipality of Costa Marquez on

the border with Bolivia on both sides of the Caut6rio river. The RESEX had a

total area of 146,400ha and a population of 210 individuals.

It was created in 1995 by the state government of Rond6nia as a result of

the PLANAFLORO project. This was a statewide initiative supported by the

World Bank to re-orient Rond6nia's development to a more sustainable path.

PLANAFLORO was implemented after the PLANOROESTE initiative (also from

the World Bank), whose environmental impact originated an international

movement against World Bank-supported development initiatives.

The timber project started in 1995, harvesting a total area of 964 ha, with a

30-year cutting cycle. It was established as a selected logging project using

mechanical transportation within the forest and building a sawmill facility that was

run by the AGUAPE association. The project was supported originally by the

World Wildlife Fund (WWF) and later by the PDA-PPG7 fund, administered by









the Brazilian Government. The project's main objective was to develop timber

management as an economic alternative.

The project also represented a strategy to defend the RESEX from the

continuous invasions by illegal loggers and land speculators. Since 1995,

AGUAPE had sued over 10 invasions to the extractive reserve. These invasions

were made more complicated by the lack of action of the state government to

defend the areas. In some cases the state government went as far as cutting

deals with the squatters, allowing their presence in the RESEX under the

condition that they stop clearing land. The project provided some limited

capacity to patrol the area's borders and react early to potential incursions.

After seven years the project had achieved a strong level of community

participation in all field operations; however, community control over the sawmill

had confronted various problems. The most significant were the management

and marketing difficulties, given the lack of business skills of the participants.

Other issues had to do with the long time required for legal permits, and the

inadequacies of the machinery scale (too small for the type of job).

Most of the timber offered in Rond6nia came from legal deforestation (to

clear for agriculture or pasture) or illegal logging; this kept prices low, affecting

the viability of the project. For this reason, the project had focused on completing

a business plan and had hired new personnel for the marketing of the products.

A new road into the reserve would greatly benefit the project, but contrary to the

situation in Acre, the government in Rond6nia had little interest in sustainable

development or community forestry. Through the support of WWVVF, ECOPORE









was pursuing FSC certification of the enterprise, hoping this would bring a plus to

the timber selling prices.

Agco Ecol6gica Guapore

ECOPORE was a non-governmental organization with thirteen years of

experience working for the sustainable development of the extractivist

communities in Rond6nia. Along with the Organization of Rubber-tappers from

Rond6nia (OSR), they participated in the early movement to stop the

PLANONOROESTE development plan, and pressured the government for the

creation of State RESEXes like Cautario. In this partnership OSR did most of the

political work while ECOPORE provided the scientific information and the

technical assistance.

The relationship continued in the formulation of the forest management

project for the Cautario RESEX. The strong articulation among the groups

generated high levels of trust among the different organization's leaders and

representatives. Originally WVVF-Brazil was one of these partners, but

progressively it withdrew to a more indirect role.

ECOPORE was a small technical organization led by an agronomist, who

was the team coordinator. Along with this coordinator three other professionals

worked in the Cautario project: a forester, an administrator and a marketing

specialist.

Associacao dos Seringueiros do Vale do Guapore

The AGUAPE association was founded in 1992 as the rubber tappers from

the Guapore river valley, on the west side of the state of Rond6nia, formally

organized the group that would later cover three different extractive reserves:









Cautario, Courrralinho and Pedras Negras. The main objective of the group was

to defend the socioeconomic interest of its members through mutual assistance

and the development of activities to strengthen the quality of life of the

communities. The association had three bodies: a general assembly, a

coordination team and a directory board. The Board was the governing body that

was elected by the general assembly every two years. The coordination team

was composed of representatives of the three extractive reserves and oversaw

the equitable development of activities.

AGUAPE had over 365 members, and partnerships with a series of NGOs

and government agencies in different projects, from schools and adult education

centers to health and ecotourism facilities. One of these projects was the

Cautario timber project, in which the association participated directly in all field

operations and also ran the sawmill. Through this project they were able to open

twenty-nine kilometers of roads within the Cautario reserve; however, the project

required much more road investment to secure its financial sustainability.

AGUAPE was continually challenged by the technical complexity of the

timber project. Rubber-tappers were not familiar with the activity, and there was

a serious cultural gap between their traditional lifestyle and the running of the

project. Low levels of formal education among the rubber-tappers were also a

contributing element. Nevertheless, steady progress had been shown as

community control had increased throughout the existence of the enterprise.

The community endured despite the intense involvement required by the

project, even though very little profit had been generated. However, the rules









collectively established to manage profits defined that 62.5% would stay in the

hands of the association for re-investment (37.5%) and association activities

(25%). The other 25% gross profit was distributed among the participants

working in the project, and 12.5% corresponded to the OSR. This arrangement

was very different from that of the other community timber projects in this study.

In the other three, most of the profits went to the rubber-tappers in whose area

the logging was taking place, and a small proportion went to the association. In

this case, not only did most of the profits go to the association, but some of it was

targeted to re-investment in the project.

Comparative Analysis

The purpose of this chapter is to identify the internal factors affecting the

collaboration experience both at the organizational and the project level, as I

expected this organizational experience to inform the effectiveness and

significance of the GPF as a collaborative platform. So far I have introduced and

described the four projects and their respective organizations involved in the

GPF initiative. From this pool of stakeholders I made a deeper analysis of the

variables more closely related to collaboration (Table 3.1)

I took these variables from the literature on conservation and community

forestry coalitions and used them to elicit the common collaboration pattern. I

defined this pattern as the implicit template that modeled the style and

effectiveness of collaboration in the GPF. In each section I start by describing

the results from my data collection. I aimed at assessing the condition of each

variable across the projects, to identify the common elements. I also mention









some of the relevant differences among the projects, emphasizing the ones that

might pose a future threat (or an opportunity) for the GPF process.


Table 3.1: Comparative Analysis Variables
Level Variable
Organizational: Participation
Community Assoc (CA) Leadership
Partner Organizations Management Capacity
(PO)
Projects: Partners' Relationship
(Usually represent a Project's training Focus
collaboration between a Appropriation of the
CA and a PO) Project by the Community

Participation

The four community timber projects analyzed in this study were designed

originally as "community participatory" projects. In practice these labels were

hard to assess, as the projects were neither completely open to every member of

the community nor did they explicitly exclude intra-community groups. The

projects did take place in a community setting, and their direct beneficiaries were

poor rural households living in communities. All of them were designed as "pilot

projects" with the aim of developing a production model that could later be

extended within the community and to similar communities. In the four to six

years since these projects began, this scaling-up had not yet occurred on a

significant scale. In fact, a few participants from each project had abandoned it.

The projects ranged from 10 to 25 participants21. Three of the four projects

used the individual areas of the participants as the area for the timber project,

establishing a direct relationship between what was cut from an area and what

21 Project staff referred to the number of families involved, but the direct participants in the
projects (referred to as "manejadores") were mostly the male heads of household.









the owner of that specific area received. The experimental nature of these timber

projects explained the small number of participants. The expectation remained

that once consolidated they would include the rest of their communities.

Participation in a project context also refers to the involvement of the

beneficiaries in the decision-making process. In the cases analyzed this kind of

participation was an explicit goal. The timber projects aimed at building capacity

in the community to run the timber operation and manage the resulting

enterprise. The goal of transferring the "project" to the community required

extended participation by the locals, and progressive decision-making autonomy.

I discuss this issue in a later section on project appropriation.

Association members were typically the male heads of household, with little

participation from women, elders or youth. Associations were created as a result

of an external institutional intervention: either the creation of the settlement or the

timber projects. In periods that varied from four to eight years, the boards of

directors had experienced little change in membership. Board members

changed positions with each election (every two years) but very few people

joined as new members. Typically only two (three in one case) individuals had

become association presidents in each association. Presidents tended to return

after one period or less than that if the group experienced a crisis. Kinship and

friendship tended to play a role in who got elected to the boards and, incidentally,

participation in development projects (like the timber management) tended to

include those board members first and then other affiliates.