1 INTERPRETING LEGITIMACY THROUGH A MULTI LEVEL GOVERNANCE LENS: THE CASE OF ENVIRONMENTAL FORUMS IN BOLIVIA By LAURA F. KOWLER A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFIL LMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2013
2 2013 Laura F. Kowler
3 To the students in the School of Natural Resources and Environment and the Tropical Cons e rvation and Development Program as well as social organizations across the world
4 ACKNOWLEDGMENTS I would first like to thank my amazing mother and father for their unconditional love and support throughout my life and educational career. I g ive special thanks to my committee members Dr. Christine Overdevest, Dr. Christopher McCarty, and Dr. Grenville Barnes who encouraged my understanding and application of theory, research design and methods, as well as the development of my research instrum ents. I also would like to thank my closest colleague, Jennifer Arnold, for her valuable support and insights throughout my graduate experience. Most importantly, I would like to thank the many individuals who participated in this research and shared their time and enthusiasm with me. I am especially grateful to the Villegas and Bilbao de l a Vieja family for providing a very comfortable and loving living environment during my time in La Paz the base city for my fieldwork In particular, I thank my Bolivian parents, Fer and Goi (Begonia de Alta Gracia) and my Bolivian grandparents, la Na a and el Tata, for their warmth and encouragement. Back in Gainesville, I would like to thank my wonderful neighbors Sandy and Henry for their continuous support throughout my graduate experience. In addition, I am very grateful to Alexander Boswell Ebersole for his support, encouragement, and editing skills that came in handy during the final stages of my dissertation writing process. This research was made possible by the N ational Science Foundation and partial
5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ ............... 4 LIST OF TABLES ................................ ................................ ................................ ........................... 9 LIST OF FIGURES ................................ ................................ ................................ ....................... 10 LIST OF ABBREVIATIONS ................................ ................................ ................................ ........ 11 ABSTRACT ................................ ................................ ................................ ................................ ... 13 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .................. 15 Background and Statement of the Problem ................................ ................................ ............ 16 Purpose of the Study ................................ ................................ ................................ ............... 17 Significance of the Study ................................ ................................ ................................ ........ 19 Primary Research Questions ................................ ................................ ................................ ... 20 Structure of Dissertation ................................ ................................ ................................ ......... 21 2 THE SHAPING OF RECENT POLITICAL HISTORY AND THE EMERGENCE OF THE INDIGENOUS VOICE ................................ ................................ ................................ .. 23 Revolution and Organization in the Andes ................................ ................................ ............. 24 Emergence of Social Movements in Lowland Bolivia ................................ ........................... 27 Decentralization Reforms ................................ ................................ ................................ ....... 29 Law of Popular Participation ................................ ................................ ........................... 29 INRA Law ................................ ................................ ................................ ....................... 31 Contemporary Political and Social Movements ................................ ................................ ..... 33 The Rise of MAS ................................ ................................ ................................ ............. 33 ................................ ................................ ..... 34 The Indigenous Rights Movemen t of Today ................................ ................................ ... 37 Conclusions ................................ ................................ ................................ ............................. 40 3 BACKGROUND LITERATURE ................................ ................................ .......................... 42 The Emer gence of Participatory Governance ................................ ................................ ......... 42 Collaborative and Network Governance ................................ ................................ ................ 45 Deliberative Democracy ................................ ................................ ................................ ......... 49 Empirical Inquiry ................................ ................................ ................................ .................... 52 This Research ................................ ................................ ................................ .......................... 59 4 FIELD SITES ................................ ................................ ................................ ......................... 62 Inventory of Cases and Case Selection ................................ ................................ ................... 62
6 Geography ................................ ................................ ................................ ............................... 63 Coordinator of Affected Communities of River Desaguadero, Ur Ur y Poop (CORIDUP) ................................ ................................ ................................ ........................ 63 Site Location and Context ................................ ................................ ............................... 63 Legal and Environmental Context ................................ ................................ ................... 64 Creation of CORIDUP ................................ ................................ ................................ .... 65 Block of Campesino and Indigenous Organizations in the Northern Bolivian Amazon (BOCINAB) ................................ ................................ ................................ ........................ 69 Site Location and Context ................................ ................................ ............................... 69 Legal and Environmental Context ................................ ................................ ................... 71 Creation of BOCINAB ................................ ................................ ................................ .... 73 Integrated Management Area of Apolobamba (ANMIN A) ................................ .................. 77 Protected Areas and Co management in Bolivia ................................ ............................. 77 Site Location and Context ................................ ................................ ............................... 78 Legal and Environmental Context ................................ ................................ ................... 80 Creation of the ANMIN A ................................ ................................ .............................. 81 5 METHODOLOGY ................................ ................................ ................................ ................. 88 Research Design ................................ ................................ ................................ ..................... 88 Sampling Frame ................................ ................................ ................................ ............... 89 Addressing Validity and Reliability ................................ ................................ ................ 89 Instrumentation ................................ ................................ ................................ ....................... 90 Pre testing Instruments ................................ ................................ ................................ .... 92 Interviews with Key Informants ................................ ................................ ...................... 93 Social Network Survey ................................ ................................ ................................ .... 93 Ne twork level measures ................................ ................................ ........................... 95 Node level measures ................................ ................................ ................................ 96 Legitimacy Scale ................................ ................................ ................................ ............. 98 Follow up Interview ................................ ................................ ................................ ........ 99 Macro Level Interviews ................................ ................................ ................................ 100 Participant Observation ................................ ................................ ................................ 100 Data Processing and Analysis ................................ ................................ ............................... 101 Social Network Analysis ................................ ................................ ............................... 101 Linear Regression Analysis ................................ ................................ ........................... 102 Qualitative Analysis ................................ ................................ ................................ ...... 102 Fieldwork Essentials ................................ ................................ ................................ ............. 1 02 Gaining Support ................................ ................................ ................................ ............. 102 Research Team ................................ ................................ ................................ .............. 103 Participant Accessibility ................................ ................................ ................................ 104 Informed Consent ................................ ................................ ................................ .......... 104 Conclusions ................................ ................................ ................................ ........................... 104 6 DEVELOPMENT OF LEGITIMACY SCALE ................................ ................................ ... 108 Scale Development ................................ ................................ ................................ ............... 108 Generating an Item Pool ................................ ................................ ................................ 108
7 Pre Testing ................................ ................................ ................................ .................... 110 Reliability Test ................................ ................................ ................................ .............. 111 Data Collection and Analysis ................................ ................................ ............................... 111 Item Analysis ................................ ................................ ................................ ................. 111 Discrimi natory Power Analysis ................................ ................................ ..................... 112 Factor Analysis ................................ ................................ ................................ .............. 112 Reliability Test ................................ ................................ ................................ .............. 113 Conclusions ................................ ................................ ................................ ........................... 114 7 SOCIAL NETWORK ANALYSIS ................................ ................................ ...................... 118 Network Level Analysis ................................ ................................ ................................ ....... 119 Node Level Analysis ................................ ................................ ................................ ............ 122 Meso level: Degree and Betweenness Centrality ................................ .......................... 122 Forum 1: CORIDUP ................................ ................................ ............................... 124 Forum 2: BOCINAB ................................ ................................ .............................. 128 Forum 3: ANMIN A ................................ ................................ .............................. 130 Macro level: Degree Centrality ................................ ................................ ..................... 131 QAP Correlation ................................ ................................ ................................ ............ 133 Conclusions ................................ ................................ ................................ ........................... 135 8 INTERNAL LEGITIMACY AS A FUNCTION OF NETWORK VARIABLES ............... 144 Transformation of Legitimacy Scores ................................ ................................ .................. 145 Regression Model ................................ ................................ ................................ ................. 146 Qualitative Model Results ................................ ................................ ................................ .... 149 Relationship with Supporting Institutions ................................ ................................ ..... 149 Leadership Experience ................................ ................................ ................................ .. 155 Conclusions ................................ ................................ ................................ ........................... 157 9 CROSS LEVEL LINKAGES AND EXTERNAL LEGITIMACY ................................ ..... 162 M eso Macro Relations: Government Responsiveness to Forum Concerns ......................... 163 CORIDUP ................................ ................................ ................................ ...................... 163 BOCINAB ................................ ................................ ................................ ..................... 169 ANMIN A ................................ ................................ ................................ ..................... 174 Institutional constraints ................................ ................................ ................................ 177 Section Conclusion ................................ ................................ ................................ ........ 179 Meso Micro Relations ................................ ................................ ................................ .......... 179 CORIDUP ................................ ................................ ................................ ...................... 180 BOCINAB ................................ ................................ ................................ ..................... 184 ANMIN A ................................ ................................ ................................ ..................... 185 Section Conclusion ................................ ................................ ................................ ........ 187 10 CONCLUSIONS ................................ ................................ ................................ .................. 188 Internal Legitimacy ................................ ................................ ................................ ............... 189 External Legitimacy ................................ ................................ ................................ .............. 190
8 Conclusions ................................ ................................ ................................ ........................... 192 Contributions ................................ ................................ ................................ ................. 194 Limitations ................................ ................................ ................................ ..................... 195 Future Research ................................ ................................ ................................ ............. 196 APPENDIX A IRB NOTICES ................................ ................................ ................................ ...................... 199 B INSTRUMENTATION ................................ ................................ ................................ ........ 201 C MACRO INSTITUTIONS USED IN DATA COLLECTION ................................ ............ 209 LIST OF REFERENCES ................................ ................................ ................................ ............. 211 BIOGRAPHICAL SKETCH ................................ ................................ ................................ ....... 224
9 LIST OF TABLES Table page 4 1 Inventory of forums. ................................ ................................ ................................ .......... 84 4 2 Characteristics of forums selected for this study. ................................ .............................. 85 5 1 Operationalization of network variables. ................................ ................................ ......... 106 5 2 Key literature used for evaluating procedural dimensions of internal legitimacy. .......... 107 6 1 Item tot al correlations for the 15 items included in the final legitimacy score. .............. 115 6 2 Eigenvalues for items used to determine number of salient scale components. .............. 116 7.1 Out degree and In degree for CORIDUP members. ................................ ........................ 136 7 2 Out degree and In degree for BOCINAB members. ................................ ....................... 137 7 3 Out degree and In degree for ANMIN A members ................................ ....................... 139 8 1 Linear regression models of network variables and socio demographic variables on forum legitimacy perceptions. ................................ ................................ ......................... 160 8 2 Comparison of mean and standard deviation of legitimacy, socio demographic variables, and network variables by case. ................................ ................................ ........ 161
10 LIST OF FIGURES Figure page 3 1 Conceptual map displaying the cross level linkages involved in constructing internal and external legitimacy within a multi level governance system. ................................ ..... 61 4 1 Map of Bolivia. ................................ ................................ ................................ .................. 86 4 2 Vicua, an endemic species found in the protected area of Apolobamba. ........................ 87 4 3 Alpaca, the domesticated species of vicua found in the protected area of Apolobamba. ................................ ................................ ................................ ...................... 87 6 1 Scale items plotted onto rotated space used to determine distinct components. ............. 117 7 1 Degree centrality of CORIDUP members. ................................ ................................ ...... 136 7 2 Node betweenness of CORIDUP members. ................................ ................................ .... 137 7 3 Degree centrality of BOCINAB members. ................................ ................................ ...... 138 7 4 Node betweenness of BOCINAB members. ................................ ................................ .... 138 7 5 Degree centrality of ANMIN A members. ................................ ................................ ...... 139 7 6 Node betweenness of ANMIN A members. ................................ ................................ ... 140 7 7 Histogram of CORIDUP macro data. ................................ ................................ .............. 140 7 8 Histogram of BOCINAB macro data. ................................ ................................ .............. 141 7 9 Histogram of ANMIN A macro data. ................................ ................................ .............. 141 7 10 Macro degree centrality for CORIDUP members. ................................ .......................... 142 7 11 Macro degree centrality for BOCINAB members. ................................ .......................... 142 7.12 Degree central ity of BOCINAB members. ................................ ................................ ...... 143 7 13 Macro degree centrality of BOCINAB members. ................................ ........................... 143 8 1 Percentage of males and females by group. ................................ ................................ ..... 15 8 8 2 Average age of forum members by group. ................................ ................................ ...... 158 8 3 Highest level of education completed by forum members by group. .............................. 159
11 LIST OF ABBREVIATIONS ANMIN A rea Natural de Manejo Integrado Nacional Apolobamba (Integrated Management Area of Apolobamba) BOCINAB Bloque de Organizaciones Campesinos y Indigenas en el Norte Amazonico de Bolivia (Block of C ampesino and Indigenous Organizations of Bolivia) CEPA Centro de Ecologa y Pueblos Andinos (Center for Ecology and Andean Peoples) CIDOB Central Indgena del Oriente Boliviano (Organization of Indigenous of the Bolivian Orient) CONAMAQ Consejo Nacional d e Ayllus y Markas del Qullasuyu (National Council of Qollasuyo Ayllus and Markas) CORIDUP Coordinadora en defensa de la cuenca del Ro Desaguadero, los lagos Ur Ur y Poop (C oordinator of Affected Communities of the Rivers Desaguadero, Ur Ur y Poop ) CSUTCB Confederacin Sindical nica de Trabajadores Campesinos de Bolivia (Confederation of Campesio Worker Unions of Bolivia) INRA Instituto Nacional de Reforma Agraria (National Agrarian Reform Institute) LPP Ley de Participacion Popular (Law of Popu lar Participation) MAS Movimiento al Socialismo (Movement Towards Socialism) MNR Movimiento Nacionalista Revolucionario (National Revolutionary Movement) NGO Non governmental organization OTB Organizaciones territoriales de base (Territorial b ase o rganiza tion ) PCA Principal c omponent analysis SERNAP Servicio Nacional de Areas Protegidas (National Protected Areas Institute ) SNA Social Network Analysis TCO Territorios comu nitarios de origen (Indigenous communal territory )
12 TIPNIS Territorio Indgena y Parqu e Nacional Isiboro Secure (Indigenous Territory and National Park Isiboro Secure) VME Vice Ministerio del Medio Ambiente y Agua (Vice Ministry of the Environment and Water)
13 Abstract of Dissertation Presented to the Graduate School of the University of F lorida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy INTERPRETING LEGITIMACY THROUGH A MULTI LEVEL GOVERNANCE LENS: THE CASE OF ENVIRONMENTAL FORUMS IN BOLIVIA By Laura F. Kowler May 2013 Chair: Chri stine Overdevest Co chair: Grenville Barn es Major: Interdisciplinary Ecology With the rise of participatory governance new spaces for grassroots political participation have increasingly enabled a wide range of civil society actors to influence the polic y process at a number of different levels. The emergence of such governance arrangements has inspired questions regarding the ir legitimacy, the opportunity of ci tizens to express their voice, and government responsiveness. I developed a multi level governa nce framework to examine the legitimacy of several environmental forums in Bolivia in which civil society actors are engaged in resolving natural resource governance issues emerging at the local level and through policy frameworks at the national level. I used a comparative case study design to explore how legitimacy is constructed at the meso (forum) level, while incorporating the effects of the micro (local or communit y) and macro (national) levels. In this study, legitimacy is conceptualized as both in t ernal and external legitimacy, where the former refers to perceptions of the democratic quality of forums and the latter refers to the recognition and support of forums by outside actors I developed a scale to measure internal legitima cy a nd employed social network analysis to operationalize interactions among forum members at the meso level and their interactions with
14 actors situated at the micro and macro levels. I used e thnographic methods to contextualize the social network data and to examine how forums engage actors acr oss these levels, which provides an understanding of the Findings indicate that the dynamics at the meso level intimately relate to the cross level linkages to the m icro and ma cro levels t hus demonstrating the need for greater integrat ion of these linkages to achieve a more legitimate form of governance T he most central network actor s or forum leaders are most influential in shaping internal and external legitimacy through th eir control over communication and decision making. Findings pertaining to external legitimacy also highlight the significance of the broader socio political context, which affects the responsiveness to forum demands lity to voice demands at the national level. This study provides insights into emerging collaborative networked arrangements involving civil society actors in participatory governance.
15 CHAPTER 1 INTRODUCTION In response to international political pressu res and the intensifying social movement number of decentralization reforms in the 1990s. These reforms were instrumental in opening new spaces for grassroots po litical participation and increased opportunities for indigenous and local organizations to participate in local decision making (Medeiros 2001). One of these reforms, the Law of Popular Participation, and the political forces it encouraged locally and nat ionally, marked a significant step in the process of democratization and the strengthening of civil society in Bolivia (Chaplin 2010; Thede 2011). Greater opportunities for participation in the policy process have enabled a wide range of civil society acto rs to influence policy at different political levels. traditional roles of government and non government sectors, which has inspired the development of both informal and for mal governance structures (Connelly et al. 2006, 268). A number of scholars argue that s civil society organizations direct access to previously remote decision making processes as equal also Hajer and Kesselring 1999 ). This research provides an analysis of the cross level interactions involved in grassroots efforts to resolve issues of natural resource governance emerging at the local le vel and those introduced through policy frameworks at the national levels. This research offers an important illustration of multi level governance, wherein decision making authority is held at multiple levels as opposed to one particular level
16 Backgroun d and Statement of the Problem Scholarship on participatory problem solving and decision making addresses the inherent shift from government to governance that results from decentralization reforms and also presents decentralized governance as an alternati ve to failed systems of hierarchical policy making. As many decentralization scholars argue, however, decentralization does not automatically promote democratic participation and resolve issues of accountability (Agrawal and Ribot 1999; Nygren 2005) or rep resent the interests of local communities (Ribot et al. 2006; Tacconi 2007). While many decentralization scholars have focused on the expected role of local governments in producing positive decentralization outcomes, some scholars argue that any given go vernance system depends on the relationships among a wide range of actors at different levels rather than asserting that one single actor or governance structure is superior to another (Andersson and Ostrom 2008). Increasingly, scholarship in deliberative democracy has turned to issues of equality and inclusion of marginalized voices in processes of democratic governance. Collaborative governance s cholars point out how collaborative and network arrangements address the governance challenges presented in co ntemporary politics, as these arrangements are understood centered modes of governing (Ansell and Gash 2008; Huxham and Vangen 2005). Although collaboration theory and practice suggest that de cisions or actions produced through strong engagement processes will be fairer and efficacious (Emerson et al. 2012; see also Innes and Booher 1999; Sipe and Stiftel 1995; Susskind and Cruikshank 1987) as opposed to more exclusive processes a limited amou nt of research addresses the quality of such participatory processes (Emerson et al. 2012, 12; see also
17 I n response to the need for coordinated policy making to address intractable and cross cutting problems, 1 new structures of participation or new governance arrangements have questions of how citizens express voice and how to ensure institutional responsiveness and accountability have become paramount (Corwall and Gaventa 2001). S ome scholars have questioned t he legitimacy of these new governance arrangements as it is considered a necessary qu ality for effective policy making (Goodwin 1998; Shortall 2004; Connelly et al. 2006). In addition, the literature on collaborative governance and network governance considers legitimacy a necessary condition for effective collaborations; however, a gap ex ists in the empirical knowledge regarding the legitimacy of such governance arrangements in the context of participatory governance. In this research, I consider the role of several forums where civil society actors are engaged in resolving natural resour ce governance issues emerging at the local level and introduced through policy frameworks at the national level. My study contributes to the literature on collaborative and network governance by examining the legitimacy of these forums through the lens of multi level governance. Purpose of the Study Participation (LPP), spaces and opportunities for civil society actors to voice their concerns have expanded. Occurring alo ngside the emergence of a strong indigenous movement and the 1 Across both traditional institutional divides and the fragmented organizational landscape which resulted from the 990s (Connelly 2006: 268 ; see also Lowndes and Sullivan, 2004; Skelcher et al. 2006 )
18 increasing political support for indigenous rights and environmental integrity, social organizations, communities, and other civil society actors have joined together to represent local concerns and to advocate for policy changes at the national level. With the adoption of a new state constitution in 2009, the current government administration began to involve civil society actors in the process of reforming existing laws, including environmental laws considered obsolete. In this research, I use a comparative case study design to examine the legitimacy of three forums by examining how governance is constructed at the meso (forum) level, while incorporating the effects of the micro (local or commu nity) and macro (national) levels. 2 these forums were initiated by, and are composed of, representatives of different rural communities and social organizations. The representatives established these forums with the purpo se of resolving issues of natural resource governance at the local level as well as engaging in advocacy designed to influence emerging environmental policy reforms at the national level. The different populations represented by forum participants are situ ated at the micro level, whereas national level institutions operate at the macro level. As actors are embedded in patterns of shared relations (Ernstson et al. 2008), I use social network analysis to conduct group level and individual level analyses. Whi le I use whole network analysis to operationalize the interactions at the meso level, I also use personal network analysis and ethnographic data to measure the scalar influences (or relations between levels) between forum participants and institutions at t he macro level and the populations represented at the micro level. The forums themselves set up the boundaries of the whole network. Each forum 2
19 is composed of approximately 15 to 20 members serving as representatives from local populations and social organ izations. The social network data helps explain the structural components of these relations within forums, and ethnographic data collected at the micro and macro levels allow for the contextualization of the social network data and the examination of the feedback between the meso level and the macro and micro levels. I expect that focusing on the meso level (forums) while incorporating effects from the micro and macro levels will offer an alternative approach to analyzing barriers to legitimate forms of g overnance considering my focus on cross level interactions as opposed to one single level. Moreover, this research is designed to capture the interactions across the micro meso macro levels, which I argue influence the legitimacy of the forums. Significan ce of the Study T his research focuses on the legitimacy of spaces created to involve affected populations in decision making processes related to natural resource governance. It is expected that the creation of legitimate spaces for multi stakeholder invol vement offers greater opportunities for participation in the policy process by engaging citizens in democratic practice and increasing the quality of information on citizen needs and preferences available to government. Bolivia provides a particularly uniq ue context in which to conduct this research given the progressive decentralization reforms of the 1990s and the new state constitution adopted in 2009. These reforms and the new constitution are founded on principles of cultural and legal pluralism, and regulation systems as well as indigenous rights in the legal structure of national environmental policy. This study has theoretical importance and practical applica tions for those seeking to understand the increasingly complex relationships between civil society and government actors
20 involved in constructing public policy in areas marked by poverty, inequality, and natural resource depletion. The use of social networ k analysis serves as a powerful methodological tool to analyze the relationships among group structures, processes, and the associated mechanisms that influence legitimate governance systems. Primary Research Questions The following research questions and proposed hypotheses 3 are tested to better improve our understanding of the influence of key social network variables (whole network structure, network position, and cross of legitimacy while controlling for demographic variables. legitimacy as perceived by participants? H1: A highly cohesive forum will be associated with a higher average legitimacy scor e than a less cohesive forum. perceptions of legitimacy? H2: individual legitimacy scor e. H3: Actors with a higher degree of centrality will correspond to their power or influence in the forum and, in effect, influence their perceptions of legitimacy. H4: Actors with a higher betweenness centrality will have a higher legitimacy score due to higher levels of brokering and access to, and control of, information. Research Question 3: How do the interactions between the meso (forum) level and the 3 The hypotheses and the methods to test th e hypotheses will as described below, be more fully developed in later chapters.
21 H5: Participants w ith more ties to the macro level will exhibit higher levels of perceived legitimacy. H6: micro level is associated with higher levels of legitimacy. To answer these ques tions, I designed five methodological components for data collection implemented throughout my research. The first component involved the development (pre testing and validation) of a scale to measure internal legitimacy. Thereafter, through archival resea rch and semi structured interviews with key informants, I gathered information related to the historical and organizational background of each forum. I also attended forum meetings and gathered observational data related to forum interactions and the decis ion making process. For the third component, I conducted the social network survey to examine both the strength of ties among forum members and their interactions with individuals outside of the forum at the micro and macro levels. This survey also include d a legitimacy scale used to with statements regarding the democratic quality of forums. The fourth component consisted of semi structured interviews with forum participants based on the data previously collected o n networks and forum deliberation. Within the forums themselves, I collected both quantitative and qualitative data at the meso level. I also conducted interviews with individuals in national government institutions to gauge the ir recognition of, and respo nsiveness to, forum demands regarding policy considerations at the macro level In Chapter 4, I further discuss the methodology that I employed in this research. Structure of Dissertation This dissertation is organized into ten chapters. The introduction p resents the background of the problem and relevant literature, research questions, and the methodology used to better understand the legitimacy of governance arrangements in the context of participatory
22 governance. Chapter two presents the relevant politic al and social context in which previously underrepresented rural and indigenous populations have gained a voice through political and social mechanisms and processes. This chapter helps contextualize the forums under study, as they all involve rural and in digenous people representing and voicing local concerns at different levels. Chapter three presents the background literature that structures this study, incorporating elements of collaborative and network governance as well as deliberative democracy. The fourth chapter provides the environmental and legal context of each forum and the background of the problem leading to the creation of each forum. Chapter five explains the methodological components of this research, which include social network analysis, a legitimacy scale, and interviews with forum members, supporting institutions, and government institutions as well as direct observation in forum meetings and public events. Chapter 6 provides a detailed description of the steps taken to develop the legit imacy scale. Chapter 7 presents the results of the social network analysis, in which I discuss the network dynamics of each forum. By examining the relationship between the network variables and internal legitimacy, Chapter 8 includes the quantitative and qualitative analyses used to answer Research Question 1 and 2. Chapter 9 presents ethnographic evidence to a ddress Research Question 3, in which I examine the cross level linkages between the meso and macro levels as well as between the meso and micro leve ls. Finally, Chapter 10 presents the discussion, conclusions, limitations, and contributions of this study.
23 CHAPTER 2 THE SHAPING OF RECENT POLITICAL HISTORY AND THE EMERGENCE OF THE INDIGENOUS VOICE In Bolivia, rural and indigenous peoples have been excluded from the political system for most of modern history. Significant political reforms in the last three decades have prompted the reconfiguration of social and political relations. Beginning with the 1952 Agrarian Reform Law indigenous peoples were first granted the right to vote marking an unprecedented recognition of this population from the dominant ruling elite class. Over the course of the following decades, changing governments and political instability gave way to the emergence of the indige nous voice. With the adoption of its structural reforms in the 1990s, the state began to recognize the multiethnic nature of the Bolivian population and granted indigenous people territorial rights. The increasing integration of the indigenous organization s into the political system climaxed in of the vote. In a country where the majority 1 of the population is made up of people with indigenous backgrounds, equated wit h poverty and exclusion, this was an historic moment for the indigenous movement. Many factors in the social and political context have inspired this remarkable change and 10; Fuentes 2007 ). In this chapter, I address the role of social organizations in the recent political history leading up to the process of change currently underway in Bolivia that has resulted in the profound recognition of indigenous rights. To illumina te the growth in indigenous political inclusion and influence, I review key political events and actions that have shaped the historical context. I also discuss how the structural reforms of the mid 1990s provided partial impetus for 1 According to the 2001 Boliv p u lation is made up of indigenous peoples (INE 2001).
24 this process for chang e, characterized by the emergence of previously underrepresented populations and social organizations in the political arena. This change was manifested in the victory of Evo Morales and the reconstruction of th e Bolivian constitution in 2009, which, along with other reforms, gave greater rights to indigenous peoples Furthermore, I also shed light on recent government actions that have restricted the voice of underrepresented populations and social organizations, and thereby adversely affecting the process for change. Revolution and Organization in the Andes In 1952, mine workers, campesinos (subsistence peasant farmers), and parts of the middle class led a revolt, known as the 1952 Revolution, to regain control of the government that was disregarding the r ights of these populations. Supported by the National Revolutionary Movement ( Movimiento Nacional Revolucionario or MNR), this revolution resulted in a series of populist reforms such as the nationalization of mines, the implementation of land reforms, and the extension of voting rights to campesinos and women, all of which represented a move to incorporate the people of the countryside into the political system. In particular, the land reforms granted campesinos small plots of land, which were often seized from large estate owners (hacendados) that previously controlled the countryside. 2 During this time, the campesinos started to form unions based on the ayllu, 3 a pre colonial form of rural organizat ion made up of families that had traditionally been used to defend territorial boundaries and resources, facilitate agricultural production, and administer community justice (Chaplin 2010 348). The MNR, 2 Prior to the land reform some e 2005, 158 ; see also Rivera 1987). 3 The ayllu is considered a geographical and territorial unity and a local f orm of self government identified by a shared ethnic origin. Unde r Spanish colonial rule, the leaders of the ayllus were responsible for law and order in the local areas and for the payment of tribute to the Spanish crown (Chaplin 2010, 348)
25 which enacted these reforms, institutionalized miner and campesino unions, 4 establishing the organizational foundation for popular protest coalitions that remain powerful in Bolivia today. 5 During the1960s and 1970s the succession of the military government and the creation of the Military Campesino Pact removed the MNR from power and initiated actions to dism antle the connection between MNR and the campesino movement that had emerged over the years since the 1952 Agrarian Revolution. The new military government began to benefit large landholders and agro business by directing government funds and other benefit s to these parties at the expense of campesino interests (Yashar 2005) The government showed increased disrespect for the sovereignty of indigenous communities and made efforts to block resources MNR previously promised to these communities, and thus, as a reaction to these threats a new generation of indigenous leaders emerged ( ibid ). Moreover, popular protests and the leadership of the Bolivian a union founded in 1952 that was opposed to the military regime energized the indigenous move ment in the Andes (Van Cott 2000). precipitated the formation of new social organizations, such as the National Council of Qollasuyo Ayllus and Markas ( Consejo Nacional de Ayllus y Markas del Qullasuyu (National Council of Qollasuyo Ayllus and Markas or CONAMAQ) and the Confederation of Campesio Worker Unions of Bolivia ( Confedracion Sindical Unica de Trabajadores Campesinos de Bolivia or CSUTCB), which became two of the mo st important social movement organizations in Bolivia (Yashar 2005) B 4 Referre 5 Although the MNR sought to transf orm the countryside and homogenize its organizational structures, indigenous authority structures persisted and often promoted ayllus. In many areas, unions served outward relations and 162; see also Albo 1997; Healy 1996; Urioste 1989).
26 government funds away from campesino interests), the CSUTCB assumed an anti government position and its leadership develop ed various campaigns that brought political awareness to the diversity of ethnic identities in Bolivia ( Albo 1991, 316 as cited in Yashar 2005 180 ). For example, in the early 1980s, CSUTCB developed an Agrarian Reform Proposal to revise the reforms of the 1950s, demanding the recognition of communal lands, cultural pluralism, and communal labor, therefore rejecting the homogenizing assumptions of existing state laws that dealt with residents i ibid ). The proposal was submitted to Bolvia in 1984, but was never brought to a vote. Yet, these reactions underscored the importance of communal autonomy for Bol ). As in many countries throughout Latin America, the 1980s in Bolivia were marked both b y an end to a long period of military dictatorships and a transition to democracy. In 1985, the new democratically elected government responded to pressure from a growing economic crisis and international financial institutions by implementing neoliberal economic reforms, which signaled the transformation of the economy and a fundamental change to the social and political arenas (Medeiros 2001). These reforms were designed to increase imports and international investment as well as privatize state enterpri ses. Moreover, these reforms decreased agricultural subsidies froze wages and reduced social programs and services relied on by campesino communities (Yashar 2005). During this time, the government significantly impaired the labor movement by ceasing oper ations at state tin mines, which previously employed thousands of workers (M edeiros 200 resistance, such as strikes, marches, road blocks and land occupations, involving a spectrum of social actors (Chaplin 2010).
27 The closing of mines forced more than 20,000 miners to relocate, initiating substantial migration at this time While a number of miners moved to cities, others returned to the rural communities of origin (primarily Aymara co mmunities). Upon return to these communities, the unemployed miners faced economic hardship, as price liberalization had made the production of many crops unprofitable (Yashar 2005). Many former miners decided to leave their communities of origin and move to the Chapare region to cultivate coca, a crop that promised good yields, international demand and high prices ( 185). 6 Th is migration brought these ex miners, with strong indigenous identities and trade union experience, into partnership with the cocaler o 7 unions in the area, which were already regulating land distribution among the community members and interacting with state institutions to promote the interests of the cocaleros (Fuentes 2007). Moreover, as a result of the 1980s, the unions began to form armed self defense committees to protect their livelihoods and Fuentes 2007, 100). 8 Emergence of Social Movements in Lowland Bolivia development, wherein indigenous communities remained relatively autonomo us from changes in state policy (Yashar 200 5). 9 However, state laws passed in t he 1960s and 1970s facilitated the exploitation of indigenous lands by providing opportunities to large landholders, loggers, and cattle ranchers to develop and use land that was not yet titled by the government (Yashar 2005). 6 Coca production skyrocketed with an estimated 250,000 300,000 people involved in the coca industry in the years following the mine closures (Yashar 2005, 185 ; see also Healy 1987; Sanabria 1997). 7 Term used in Spanish to refer to those who produce coca. 8 The cocalero movement continued to strengthen as it fought to gain respect for the coca leaf and to legalize its production and consumption by affiliating with the CSUTCB (Yashar 2005, 1 85) 9 Churches and NGOs provided social services in the absence of state penetration.
28 During this period of time, t he state became interested in exercising more power and control in the Amazon region to relieve land pressures in the Andes and to stimulate large scale development in the area. 10 Direct threats to local indigenous autonomy caused by the economic advanceme nt of protecting their land and territorial autonomy (Yashar 2005) In particular, the indigenous groups in these regions, which make up 30 of the 36 different in digenous identities in Bolivia, began to form regional organizations with the help of the e merging non governmental organizations (NGOs) and churches that encouraged the growth of the indigenous movements in the region at this time ( ibid ). In 1982, four i ndigenous organizations formed the Indigenous Federation of Eastern Bolivia ( Confederacin Indgena del Oriente Boliviano or CIDOB) to coordinate their activities in defense of the indigenous identity and rights to land and natural resources. The consolida tion of the indigenous movement first arose in 1990 with the first indigenous march, referred to as the Indigenous Regional Federation of Beni ( Central de Pueblos Indge nas del Beni or CPIB). Stimulating the coordination between the lowland and highland indigenous movements for the first time, thousands of indigenous marched 330 km during a 35 day period from the Amazonian lowlands to the capital city of La Paz, to demand state recognition of indigenous territorial rights and control over the use of their natural resources. This march inspired the political commitment required to initiate the process leading to the legal recognition of indigenous territori es in the 10 The 1966 Colonization Law particularly presented the greatest threat to these populations, encouraging Andean farmers and landholders to migrate east and colonize untitled areas, even distributing lands to colonizers and large
29 lowland s (Medeiros 200 1 as cited in Yashar 2005, 213) and demonstrated indigenous peoples to make territorial demands as well as the capacity of the state t o respond ( ibid ). The march also served to strengthen a collective identity that revolve d around self defense and territorial rights, which became the central political issue s for lowland indigenous movements (Molina 1997 as cited in Yashar 2005, 213 ). The territorial areas created by the demarcation of indigenous territories in the early 199 0s, required indigenous people to balance their local autonomy with greater national political integration (Yashar 2005). Additionally, CIDOB began advocating for a comprehensive indigenous agenda at this time that proposed policy reforms designed to inst itutionalize both the inclusion and autonomy of indigenous people by incorporating greater space in the political process for negotiation between the indigenous populations and the state (see Albo 1994). By the early 1990s, indigenous organizations demande d attention in most social and political debates as recognized actors and thus traditional political parties were pressured to win the ir vote (Van Cott 2000). Moreover, the highland and lowland indigenous movements represented the indigenous majority for decentralization reforms most notably, the Law of Popular Participation, which is discussed below. Decentralization Reforms Law of Popular Participation has been considered one of the most significant and innovative examples among Latin American countries that have undergone such processes (Mayorga 1997). The Law of Popular Participation ( Ley de Participacion Popular or LPP), enacted by the government of Sanchez de Lozada in 1994, was a response to pressure placed both on the state and on traditional political parties by indigenous and campesino organizations
30 (Kohl 2003) Embracing neoliberal structural reforms, wing government focused on th e privatization of natural resources and social and legal reforms that emphasized bilingual education, the decentralization of decision making authority over resource allocation, and the implementation of different international human rights norms, especia lly those involving the an 636 ). 11 The LPP set up a series of mechanisms designed to ensure grassroots participation and citizen oversight in local government by creating over 300 municipalities to whic h 20 percent of national ta xes were transferred. T hese reforms generated the redistribution of administrative autonomy to departmental and municipal governments, which resulted from efforts to decentralize the government and reallocate resources to long ne glected rural communities (Kohl 2003). With the partial aim of promoting an active citizenry, the LPP also created a system to encourage citizens to engage in local government by creating Territorial Base Organizations ( Organizaciones Territoriales de Base or OTBs ) based on recognized grassroots and civil society organizations such as allyus, rural campesino unions, and urban neighborhood councils (Medeiros 2001). Moreover, the LPP required communities to apply for legal status and obtain a legal personalit y. 12 Prior to the LPP, the territorial jurisdiction of a municipality was confined to the urban area, town or city. The LPP expands this jurisdiction to include the rural areas that include campesino and indigenous communities, thus creating a new layer o f autonomous local government and a significant extension of citizenship rights (Kohl 2003). At the c ore of the LPP 11 The prominence of the indigenous movement was reinforced in the international debate on the collectiv e rights of indigenous peoples inspired by the United Nations forums, such as Agenda 21 adopted at the United Nations Conference on Environment and Development in 1992 and later in the United Nations Declaration on the Rights of Indigenous Peoples adopted in 2007 (Mayorga 2007). 12 Documentation authorized by the state that provides legal status and recognition to organized groups
31 was the legal recognition of thousands of campesino and indigenous communities throughout the country, their traditional governing structure s, and their territorial rights (Urioste 2002). Furthermore, the LPP provided new political resources and legitimacy to OTB s ( Kohl 2003; Medeiros 2001; Van Cott 2000 ). The LPP was instrumental in opening new spaces for grassroots political participation a nd increased opportunities for indigenous and local organizations to participate in local decision making, while establishing an effective form of social control over local government actions (Medeiros 2001). As a consequence of these new spaces for parti social organizations began to grow through increased levels of organization and mobilization to make their demands more widely heard. 13 Moreover, the LPP and the political forces it encouraged locally and nationally marked a significant step in the process of democratization and the strengthening of civil society in Bolivia (Chaplin 2010; Thede 2011). INRA Law In response to the political spaces and opportunities for indigenous movements, CIDOB resurrected the first mobiliza tion to consolidate the existing indigenous territories and demand new ones in 1996. 14 This march led to the incorporation of a number of articles dealing with indigenous property rights in the National Agrarian Reform Law ( Ley Nacional de Reforma Agraria ) or the INRA Law passed in October 1996 which represented an unprecedented piece of legislation that amongst others, laid out a process for recognizing communal land rights for rural people living throughout the country. While the rest of Latin America was advocating the privatization of land markets, the INRA Law provided mechanisms for the state to legally 13 New alliances and coordinated forms of resistance emerged around the country in the late 1990s (Chaplin 2010). 14 CIDOB came to play an important role in social movement circles and policy debates by the 1990s (Yashar 2005).
32 recognize indigenous communal lands in the form of Indigenous Communal Lands ( tierras communitarias de origen or TCOs). The law essentially recognized spaces in which indigenous authorities could de facto institutionalize their authority over the TCOs ( Yashar 2005, 218 ) Despite these political advances in favor of indigenous rights, the implementation and political support of the discussed reforms sta lled with successive governments, which directed minimal attention to indigenous interests. 15 For example, the promise granted by the INRA Law was revoked. 16 In addition, when Lozada was elected again in 2002, he did not voice the same kind of popular commit ment to ethnic demands and thus severed his ties with the indigenous people. It is important to note that with the 1992 UNCED Conference in Rio de Janeiro, the theme of sustainable development became part of the economic development aid agreements. NGOs, e merging in the 1980s, played an important role in promoting the idea of sustainable development, as they received funds from donor agencies that provided them with technical support. In Bolivia, these NGOs played an important role in facilitating public po licy debate on land and indigenous rights and, in some cases, they helped to secure and administer economic support for retitling ( Arellano Lpez 2012). Even though social organizations supported the reforms in principle, in practice they tended to distrus t the reform implementation, referring to the reforms, actually constituted a victory for 132). 15 Such changes shift the reforms toward departmental political oversight and away from an emphasis on local participation to poverty alleviation and away fro m the municipalities (Van Cott 2000). 16 On the last day of the first Sanchez de Lozada administration, INRA issued a regulation for the Ley INRA that worked against the rights outlined in the law and created an extensive and costly bureaucratic procedure for processing claims for TCOs, thus making it more difficult for indigenous rural communities to complete the process (Almaraz 1998, 187).
33 Contemporary Political and Soci al Movement s The Rise of MAS By 1995, indigenous and campesino candidates running under the political party, United Left (Izquierda Unida), won 28.6 % of municipal government seats mainly in the Chapare region (Van Cott 2000). The Movement toward Socialism ( Movimiento al Socialismo or MAS), founded in 1997, grew out of the CSUTCB and the coca 17 of the Chapare. In order to advance the interests of the movement (the defense of coca production) and legitimize collective forms of decision makin g and political action, MAS sought to use the electoral process to gain political influence (Thede 2011, 217). In efforts to explain the rise of MAS, s ome scholars point to the profound crisis of the Bolivian state, characterized by t he lack of economic de velopment, the social exclusion of the indigenous people, and the lack of any real popular representation through the existing political party structures (Fuentes 2007; Thede 2011). T he local governance mechanisms introduced by the LPP that stimulated the integration of local unions as OTBs into the local political sphere and considered them legitimate expression of popular will and participation created the space for MAS and its growth in the political arena. MAS became a channel for popular voice in the context of the intense cycle of social protest starting in the year 2000 (Thede 2011 217 ). 18 Moreover, MAS rapidly gained support from rural indigenous communities both in the Amazon region and in the highlands, and by the 2002 national elections MAS bec ame the second most important political force, with 19% of the popular vote (ibid). 17 Headed by Evo Morales, the cocaleros became the center of natio nal resistance to imperialism. This expansion was aided when the predominately Quechua cocaleros gained control of the CSUTCB in the early 1990s (Fuentes 2007). 18 Left indigenous struggle was revitalized in the 2000 Cochabamba Water War against the World Bank driven privatization of water in that city fo llowing 15 years of right wing neoliberal reforms.
34 In 2003, the government proposed new taxes and tax increases, prompting uprisings among both rural and urban populations. These uprisings, which included the use of police and military intervention, almost completely destroyed the credibility of the political party system, forcing the president at the time, President Snchez de Lozada, to flee Bolivia. Social movements again witnessed a significant increase in their influen ce during May and June of oil and gas industry in Bolivia, which boasts the second largest deposits of natural gas in South Americ a after Venezuela (Webber 2011, 48). In addition to renationalization of the oil and gas industry, the protests also were associated with the nationalization of other natural resources, such as water, mining minerals, as well as the protection of land and indigenous territory. In March 2002, a coalition of the major campesino and indigenous organizations formed the Unity Pact, considered one of the most important efforts of political articulation in the history of the country's indigenous campesino movement ( ibid ) The Unity Pact set ou t to promote the establishment of a Constituent Assembly to restructure the 1967 C onstitution to incorporate the recognition of the pre existence of indigenous people on Bolivian lands, and thus provide for measures to protect indigenous and agrarian right s and land reform (ibid) In December 2005, Evo Morales was elected president, a victory signifying together of uption of a national revolution led for the first (Fuentes 2007 97 history also represented the defeat of the traditional neoliberal parties and the exhaustion of the dominant neoliberal ideolog y in the country. A New Constitution and With Morale rewriting the constitution to recognize the right of indigenous people to live according to their
35 traditional norms and customs and declare these rights ( Lupien 2011, 790 ) 19 The Assembly was made up of leaders from indigenous and campesino social organizations that incorporated their cosmovisio nes (worldviews) into the new constitution ( ibid ). In January 2009 Bolivia adopted which implements indigenous concepts of democracy and participation with respect to natural resources and autonomy, two important issues deeply rooted in the history of the indigenous peoples ( see Regalsky 2010). 20 rooted in a radical critique of the neo liberal development approach on which the land and forestry reforms had been based ( see Pacheco et al. 2010). A mong these reforms was a revised Land Law, known as the Law for Communal Reorientation of Agrarian Reform, issued in late 2006 that seeks to redistribute public lands to communities. In addition, under the new constitution, the TCOs became Indigenous Abori ginal Campesino Territories ( Territorios Indgenas Originarios Campesinos ). This reform eliminated the assumption that a territory could belong exclusively to a specific indigenous people and, through its changes, allowed anyone with indigenous heritage (i .e., settlers and highland campesinos ) to establish residence in such territories and claim certain land ownership rights and natural resource rights, as well as the right to involvement in the governance of the area. According to scholars, this and other land tenure 19 Plurinationality is founded on several principles, such as participatory democracy, local decision making, communal ownership of land and the capacity to exercise full citizenship without abandoning cult ural practices (Lupien 2011, 776). 20 Economic and social rights, land redistribution, and anti globalization have been promoted by social movement n of
36 reforms created strong incentives for the coca growing settlers in the Chapare province to legalize the land that they were illegally occupying (Arellano Lpez 2012). 21 The new constitution also paved the way for the expansion of local and indi genous autonomy in the Law of Autonomy and Decentralization ( Ley de Autonomias y Descentralizacion ), passed in 2010, which expands the LPP by enabling municipalities with a substantial indigenous presence to convert themselves into autonomous indigenous ar eas (Lupien 2011 ) This law adds an additional layer to the governance system, and some see the law as an attempt by the government to meet the demand for further decentralization without concentrating power in the hands of the departmental elites (Regalsk y 2010). 22 Moreover, these reforms evidence the political impact of social movements, which have pushed forward making on fundamental issues such as political decentralization (Ma yorga 2007, 76). Since MAS and Evo Morales assumed power, their political discourse and actions were based on denouncing the neoliberal extractivist policies imposed on Bolivia by developed countries. 23 Since 2005, the government has continually used a disc en defensa del proceso de cambio 1053 ). However, contradictions regarding this discourse have surfaced in recent years. 21 These reforms encouraged the expansion of the cocalero population in the Chapare region, which represents the ze of the TCO referred to as TIPNIS (Terriotiro Indigena del Parque Nacional Isiboro Secure) to allow for the expansion of the coca growers (Arellano Lpez 2012). 22 Autonomy is the most prominent claim of indigenous peoples around the world and is rooted in the call for greater self determination and self governance within indigenous territories. 23 mar revival of land and natural resources use (Vivir Bien), provided spaces for disseminating the official discourse (Arellano Lpez 2012, 11).
37 The Indigenous Rights Mo vement of Today Over the past two decades, indigenous rights groups have emerged as one of the strongest social movements in many countries of Latin America. The objective of many of these social accordin g to an indigenous worldview that promotes alternative concepts of democracy, citizenship, land ownership, and (Lupien 2011, 774). Given the nature and role of social organization in Bolivia that has been shaped by the exclusionary behavior o resorting to protest to make th 353). The adoption of the new constitution in January 2009 and the success of Evo Morales in the presidential and congressional elections of December 2009 have not only signified the recognition and political influence for the indigenous and campesino Bolivian people, but have also introduced indigenous cultural values into state policies (Lupien 2011). 24 Despite its overwhelming political c ommitment to indigenous rights, in the last several years, the government has revealed otherwise through its actions that have affected the voice of rural and indigenous populations as well as the voice of other civil society actors in Bolivia. In August 2011, the indigenous people from TIPNIS 25 were joined by both lowland and highland indigenous social organizations in an o rganized march from San Ignacio de Moxos in the Beni department to the city of La Paz to demand that the government stop the constructi on of 24 ; see also Alb 1991; Bengoa 2000; Brysk 2000; St avenhagen 2002; Wearne 1996 ). 25 TIPNIS was declared a national park in the Law 1401 in 1965. As a res ult of the first indigenous march in 1990, it was declared an indigenous territory (TCO) by the Executive Decree No 22610 in September 1990 because of the need to protect the watersheds and biodiversity from further colonization in addition to the socio ec onomic space
38 a highway from the Cochabamba department to the Beni department without consulting them 26 After approximately two months and 602 kilometers, the marchers arrived in La Paz to a crowd of t ens of thousands of Bolivian citizens lining the streets and c hanting in solidarity with the marchers to demand the cancellation of the government's planned highway. The event demonstrated the unpopularity of President Morales as a result of his failure to keep the promises international discourse regarding climate justice, anti neoliberalism, and indigenous rights (Laing 2012). As a result of the indigenous mobilization and the political pressure exercised by the indigenous organizations, the President announced the cancellation of the road two days later. The government also enacted a law requiring state consultation with populations inhabiting TIPNIS, including indigenous people, coca settlers, and cattle ranchers. 27 The purpose of the consultation process was to determine whether these populations agreed with the construction of the highway. In November 2012, the government completed this process and concluded that the state must complete further environmental impact analyses. The indigenous march represents an historic event in the co ntemporary history of Bolivia. According to scholars, TIPNIS has exposed the challenges of constructing a plurinational state, to the indigenous march further ex posed the tensions and contradictions inherent in the plurinationality and economic development, the latter of which is based on the extraction of non renewable resources, mostly located in indigenous territories (Urioste 201 26 The United Nations Declaration on the Rights of Indigenous Peoples adopted in 2007 requires state governments to obtain pre, prior, informed consent before beginning development projects that will affect citizens. 27 costs of transportation and open new markets for their production.
39 discourses and disregards th e (98). Analysts also contend that the to the redistribution of land in Bolivia ( Urioste 2011 ) Reported as a victory for the environment by the international press, this mobilization and indigenous 1051 ). 28 Although the march had the effect of weakening the popularity of the current government among indigenous populations, since 2012, the government has begun to coopt many of the indigenous organizations, which is interpreted as a move to regain political support. 29 In addition, both resentment towards NGOs and its political preferences were further revealed during the mobilization when it accused the indigenous people of acti ng as instruments of North American and European neo liberal environmentalist NGOs. 30 The government has also used its instruments of power other ways (Orias 2011 8 ). 28 Morales demo indigenous president, but the first trade Pagina Siete 2011 ). 29 The national elections will be held in 2014. 30 Despite its rhetoric of participatory democracy, the current government has publically announced its rejection and dismissal of many NGOs. The governme NGO sentiment is likely rooted in its distrust in international donors that supported the neoliberal reforms of the 1990s.
40 Conclusions In response to international political pressures and the intensifying social movement provided local access and use rights as well as land t enure to rural and indigenous communities (Contreras Hermosilla and Va rgas Ros 2002; Ferroukhi 2003). The progressive nature of principles of participatory d emocracy. As the implementation of the LPP in 1994 opened up spaces for the inclusion of previously underrepresented rural populations, it paved the way for the process that led to the reconstruction of the new state constitution, which has further restruc tured the governance system in Bolivia. S ocial movements have progressively gained more space, voice, and strength in the political sphere Moreover, t he movement of anti capitalism and indigenous liberation driving these reforms represented the necessity of overcoming the oppression of these populations embedded in Bolivian history. Despite considerable discourse regarding environmental protection and multi ethnic diversity, the current government has revealed serious tensions in its vision of development as it relates to these issues. Highlighted by the way it reacted to the TIPNIS conflict, the deteriorate. Moreover, the President, who has a personal affiliation with the cocaleros, has favored the cocaleros over other populations. Furthermore, in line with the anti imperialist platform embraced by MAS, the government has also placed restrictions on the freedom of political expression of other civil society actors, such a s NGOs. Accordingly, the indigenous voice and that of other civil society actors, such as NGOs, has been quieted in efforts to either gain popularity or dis miss their demands in general. Yet, despite these actions, the government
41 has, in the face of upcomi ng elections, made active attempts to regain political support from these populations.
42 CHAPTER 3 BACKGROUND LITERATURE Governance scholars note that shifts in favor of decentralization have multiplied stakeholders and opened spaces that enable processes for the interactions among various actors in natural resource govern ance (Brondizio et al. 2009; Andersson and Ostrom 2008 ). 1 Increasingly civil society actors have gained acces s to the policy process to influence policy at a number of political levels du e to expanding opportunities to access the institutional process and represent their particular interests. A growing number of scholars are see king to understand and opened up by the shift from government to governance (Connelly e t al. 2006, 268) and have raised challenging research questions, particularly in regards to the legitimacy and effectiveness of these new governance for ms as well as the applicability of democratic theory to contexts of governance beyond the state (Hajer and Wage naar 2003 ). The Emergence of Participatory Governance political spheres since the 1990s, principally through decentralization reforms, 2 states have been increasingly encouraged to introduce broader public participation into development processes ( Connelly 2009 ). Accompanying these reforms were the growing complexity of social issues, such as social exclusion and inequalities, which challenged traditional approaches to governing through hierarchical instruments of control. As a result the role of the state shifts from that of 1 Some authors contend that new spaces have been opened up for citizens to make demands because of the discourse of decent ralization itself has fostered expectations on the part of local communities and governments (Larson and Soto 2008, 218; see also Larson 2005; Ribot and Oyono 2006; Castro and Nielsen 2001). 2 arrangements among public institutions and social actors that emerge from a broader process with two principle dimensions: (a) top down measures aimed at transferring responsibilities [and powers] political, administrative and/or fiscal to lower levels o f government and (b) the gradual opening of spaces for participation from below, induced by the actions of social movements and local (Larson et a l. 2007, 216).
43 through direct forms of co in which the state collaborate with a wi de range of actors in networks that cut across the public, private, and voluntary sectors, and operate across different levels of decision making (Newman et al. 2004 204 ). Scholars argue that these arrangements must encompass not only collaboration betwe en and among organizations but also the role of the public and citizens in governance ( Nabatchi 2010; see also Bingham et al. 2008 ). Among other expectations, d ecentralization reforms encourage the state to become transparent and accountable as well as dev elop a civil society. In many developing countries, however, corruption and the absence of civil society, and a lack of efficiency and effectiveness of existing governance structures pose many governance problems. For example, research on decentralization reforms that have transformed the institutional arrangements for natural resource governance has underscored the substantial variation in the designs and outcomes of these programs in practice, many of which fail to meet its theoretical ideals ( see Edmunds and Wollenberg 2003; Larson and Ribot 2004; Ribot et al. 2006). According to decentralization scholars, these reforms are not delivering on their objectives because they do not transfer sufficient power to local institutions and the entities with the auth ority to represent and are not accountable to local communities (see Blaikie 2006 ; Ribot 2003; Ribot et al. 2006; Tacconi 2007 ; Turner 1999 ). Evolving in response to the shift from government to governance, participatory governance practices involve inter mediary spaces that readjust the boundaries between the state and its citizens, establishing new places in which the participants from both of these parties can engage each other in new ways (Cornwall and Coelho 2007). These participatory forms of
44 governan ce can be labeled in many ways, including participatory governance, deliberative democracy, and empowered participatory governance (Fung and Wright 2003, 5). The assumptions underlying the advancement of decentralization reforms and the concept of partici patory governance are grounded in the notion of participation, understood as a process whereby individuals, groups, and organizations choose to take an active role in making decisions that will affect them ( see Rowe and Frewer 2004 ) Whereas some scholars view public mechanically applied to social and political processes (Rowe and Frewer 2004), others and political processes that reflect socially co ( Abelson et al. 2007, 2117). Greater access to decision makers, higher levels of participation by various social groups in decision making, and the accountability of decision makers are often the claimed effects of participation engagement of citizens (individuals and groups) apart from the state (also referred to as civil society ) in the governance process (Blair 2008). Instrumental to decentralization, good an increasingly complex institutional landscape that in some cases has given rise to transfers of both resources and decision Fischer 2006, 21; see also Sto ker 2000 ). New mechanisms for participation and governance, which might generally be referred to lack of coordinated policy tools addressing intractable and cross cutting iss ues (Srensen and Torfing, 2005, 197). In this context, the
45 questions of how citizens express voice and how institutional r esponsiveness and accountability can be ensured have become central (Corwall and Gaventa 2001). Collaborative and Network Governance In the last decade, the shift from government to governance has attracted significant research interest in political scie nce, public administration, sociology, and law, shaping various approaches to governance. Considerable research has been devoted to establishing a workable definition of governance, a concept that integrates several features structures, arrangements, and p that enables the survival of the collaborative partnership or i 49) and is viewed as steering the process that influences decisions a nd actions within the private, system refer to the interaction and coordination of activities among state and non state actors through networks, partnerships, a nd deliberative forum s (Pierre 2000; Hirst 2000; Kooiman 2000 ). creating opportunities and the structural and procedural conditions a 2000, 73). T he governance approach used in this work is founded on the notion of collaborative and network governance. Collaborative governance is understood as T he processes and structures of public policy decision making and management that engage people constructi vely across the boundaries of public agencies, levels of government, and/or the public, private and civic spheres in order to carry out a public purpose that could not otherwise be ac 2). This approach is also highly relat ed to collaborative public management (see Bingham et al. 2008; Agranoff and McGuire 2003). Unlike other definitions that limit collaborative governance to only formal, state initiated arrangements and to engagement between government
46 and non governmental stakeh olders (see Ansell and Gash 2008 ), this definition encompasses civil society and communities as well as hybrid arrangements such as public private and pri vate social partnerships and co management regimes ( Agrawa l and Lemos 2007; see also Emerson et al. 2012 ). It also includes a range of community based collaborations involved in collective resource management and other types of collaborative arrangements i nitiated in the private or civil sectors (Emerson and Murchi 2010). Although the understanding of collaborative governance incorporates knowledge and concepts from a wide range of disciplines, it also involves concepts from public administration and democ racy. According to this perspective, deliberative democracy promises citizens opportunities to exercise voice and a more responsive, citizen centered government by account ability, 165). Across a variety of theory and research, collaborative governance has been applied to studies in several policy c ontexts (Emerson et al. 2012 4). The perspective that government, as the single decision maki ng authority, has also been expanded by the idea of multi level, polycentric governance in which many actors in different institutional settings contribute to policy development and implementation ( see Andersson and Ostrom 2008; Hooghe and Marks 2003; Mayn tz 2006; McGinnis 1999a; Young 2000 ). In particular, t he importance of the patterns of interaction between actors and the multi level institutional settings within which they interact has attracted substantial attention due to the complex interplay that th ese dynami c s create between structur e and agency (Bodin et al. 2006 ). level governance is particularly relevant as it takes place
47 in the public sphere i n which distinct activities occurs across a range of private, poli tical and civic associations and even networks ( Hendriks 2006 ). Given the emphasis on policy networks that are organized across policy areas and government levels the body of multi level governance as networ 3 especially where the issues involved are too complex for any single entity to manage. This approa ch focuses on collaborative networked arrangements or the governance of inter organizational networks. 4 Given that all networks comprise a range of interactions among participants, a focus on governance involves the use of institutions and structures of a uthority and collaboration to allocate resources and to coordinate and control joint action across the network as a whole (Provan and Kenis 2007, 231). 5 Several scholars discuss how such forms of governance contribute to the legitimacy of the decision mak ing processes by, for example, promoting a more effective use of knowledge and resources, improving problem solving capacity, and/or facilitating a more efficient service delivery (Kickert et al. 1997; Kooiman 1993 ). Furthermore, the collaborative governan ce literature highlights that such forms of governance depend on effective leadership, where leaders act with authority, vision, political skill, and relational qualities, as well as exhibit a long term commitment to collaboration and integrity ( Bryson et al. 2006; see also Bryson 2005a; Gray 1989; Waddock 3 This approach treats networks as the unit of analysis in which the network is viewed as a mechanism of coordination (Provan and Kenis 2007). 4 Such networks are made up of multiple organizations and are grounded on the expectation is that collaboration among organizations will lead to more effective ways of addressing community needs (Provan and Milward 2001). 5 Although much of the literature on organizational networks does not explicitly address governance, some scholars point to the impo rtance of mutually supportive (Provan and Kenis 2007, 231).
48 1986 ). These scholars discuss the importance of the role of leadership in cultivating trust, which they consider the essence of collaboration Used in a variety of different policy arenas, a common objec tive in collaborative networked arrangements is to establish structures and processes that facilitate collaborative dynamics among diverse participants that in turn can enhance the quality of decisions made and implemented ( Choi and Robertson 2013; see als o Huxha m and Vangen 2005; Mandell 1999 ). Considered in this way, interest in collaborative governance, in particular, has accompanied the rise of deliberative democracy theory, as the decision making process is considered a core component of collaborative governance (Ansell and Gash 2008 ). Such observations have inspired new forms of public involvement and civic engagement, referred to differently by scholars as participatory governance (Fung and Wright 2003) collaborative governance ( Ansell and Gash 2008 ) and the deliberative democracy movement (Gastil and Levine 2005) This movement emerged from civil society and sought more public deliberation, dialogue and shared decision making in governance to address conflict at the broader level of public policy ( B ingham 2009, 389). Scholarship has also begun to focus on citizen opportunities to exercise voice and a more responsive, citizen institutions with greater levels of transparency, accountability, and Henton et al. 2005, 5; see also Bryson et al. 2006; Nabatchi 2010). Furthermore, the framework for collaborative governance draws on and applies knowledge and concepts from a wide range of disciplines, such as public administration, conflict resolution, and environmental management, among others. The integrative nature of this framework makes it potentially relevant to scholars and practitioners working in several applications and settings, such as collaborative public management, multi partn er governance,
49 network governance, co management regimes, environmental governance, participatory governance, and civic engagement (Emerson et al. 2012 ). Moreover, this approach to governance clearly recognizes the cross scalar complexity of governance sys tems in general and are considered particularly appropriate given the complex nature of environmental issues that span jurisdictional and administrative borders as well as the social arrangements constructed around them (Bodin and Crona 2009). Furthermore, the need for a better understanding of cross scale interactions is put forth by many scholars ( see Berkes 2009; Cash et al. 2006; Meadowcroft 2004 ; Ostrom 2005; 2007). Deliberative Democracy Since the 1990s, there has been a growing interest in ways to en hance public involvement in governance as well as the quality and legitimacy of democratic decision making (Fung and Wrig ht 2001; Goetz and Gaventa 2001 ). Increased citizen engagement through deliberative processes is perceived as a direct response to publ ic discontent with past p ublic participation experiences and their loss of trust in public officials ( Abelson et al. 2003; see also Graham and Phillips 1998 ; Maxwel ). The practice of deliberative governance has been inspi r ed by deliberative democracy a growing body of democratic theory that emphasizes the importance of deliberation in collective decision making. Deliberation encompasses both analytic and social processes and offers a unifying conceptual and critical framewo rk for studying a full range of communication forms, including informal conversation, media and public opinion, elections, jury decision making, public meetings, and civic and community life (Gastil et al. 2008). Among the range of definitions of delibera tive democracy more liberal interpretations focused on the institutions of the constitutional state exist as well as and other definitions focusing more on the participatory nature of governance a s it relates to direct opportunities for
50 citizens to influe nce decisions (Bohman 2004). Similarly, the degree of formality in the design of deliberative fora also creates the distinct democratic implications. With reference to the latter, there are both structured fora and less structured fora. Whereas the former equal participants come together to decide on an agenda, reason and argue together and settle on movements, and activists that seek to influ 4). The former approach to deliberative democracy thus pays little attention to civil society, consequently ignoring its role fr om discussions on legitimacy or the procedural conditions for deliberation (Hendriks 2006). version of deliberative politics extending beyond the more formally organized p olitical system to 6 involving groups within civil society. According to Chambers (2002 ), the public sphere provides spaces where various discourses and ideas in civil society can be voiced and made politically efficacious ( 96 as cited in Hendriks 2006, 489 ) Some public spheres are characterized by public deliberation along with decision making authority, such as legislative and pa rliamentary institutions (Parkins and Mitchel 2005). Weak public spheres, on the other hand, are more amorphous networks and coalitions that form around issues of public concern, characterized by deliberation without decision making authority (Parkins an d Mitchell 2005, 532 ; see also Fraser 1990 ). Deliberative democrats value these weak forms of the 6 The concept and civil associations and networks (Young 2000, 160). Young contends that one of civil socie communicative interactions that support identities, expand participatory possibilities and create networks of
51 public sphere because they provide invaluable opportunities for public deliberation that can translate into important sources of influence on decision makers and the general spaces (Parkins and Mitchel 2005). Habermas and others consider the less structured form of deliberative democracy more legitimate than those forms with greater structure. However, as the democratic nature of these spaces is unclear, criti cs point the potential inequalities that may exist i n the public sphere. S conversation and formal arenas of decision ( Hendricks 2006, 497), an idea advanced by many authors ( ibid; see also Mansbridge 1999; Hendriks 2006; Parkinson 2003; Marques et al. 2007). What Meadowcroft (2004) refers to as an which typically depends upon a clear impulse from government that identifies an issue as i mportant, providing the general context within which it is to be addressed (31). According to this author, if there is unambiguous political commitment to accomplishing an outcome, more potential exists to incorporate actors into the policy process Accor dingly, without such political commitment, organizations are unlikely to engage seriously (ibid). While much of the discussion in sociology and political science over participation and civil society has been quite theoretical in nature (Fischer 2006), the re are lessons from deliberation in practice ( Beaumont and Nicholls 2008; see also Abers 1998; Avritzer 2002; Baiocchi 2005; Fung and Wright 2 001, 2003; Skelcher 2003 ). Due to the inadequate attention paid to equality in deliberative democracy, scholars ha ve turned to issues of representation, a fundamental aspect of participatory democratic processes. Nevertheless the process of including stakeholders and the circumstances enabling meaningful debate are in practice, difficult to achieve.
52 Empirical Inquir y The late 1980s and early 1990s witnessed a discernible shift away from top down policy making to the inclusion of citize ns in policy debates (Chambers 2003 ). As a result, increases in the quantity of public participation have been accompanied by interest in the quality of participatory approaches. Some deliberative theorists have responded to the mixed results evident in empirical studies of deliberation 7 (see Carpini et al. 2004; Rosenberg 2007b) with the recognition of the importance of social and polit ical contexts in determining both the deliberative and democratic qualities of deliberative processes ( see Ryfe 2005; Thompson 2008; Warren 2007). Scholars are increasingly turning to the degree of social and political equality that deliberative democracy demands such that discussion and reasoning, rathe r than domination and manipulation, will be determined in the procedures leading to outcomes ( Parkinson 2003; Bohman 2004 ). Recent research has begun to focus on h ow participatory processes can be construct ed and facilitated to ensure equal ( see Martin 2012 ). In his work exploring the potential for public participation, Gaventa (2004) turns to new forms of citizen state engagement to explore participatory appr oaches to ensuring the inclusion of citizen voices in democratic governance processes primarily in developing countries and at various scales. Fung and Wright (2001; 2003) analyze opportunities for citizens to meaningfully engage in shaping decisions in co njunction with state actors through practices that lead to more 7 A number of authors have used which to judge the adequacy of participator y initiatives and other forums.
53 demonstrates ho empowered militants in addition to well designed institutions facilitated democratic deliberations between diverse stakeholders (Baiocchi 2005). However, this work raises the challenges of the striking inequalities in deliberative settings and suggests that well designed participatory institutions enhance successful deliberations and fair consensuses only when all sta k e holders share equal amounts of power (Fung and Wright 2001). The literature presented has addressed several critiques related to the ide als of deliberative democracy, particularly that of equality and the inclusion of marginalized voices in decision making processes Many scholars contend that several changes can still be made in order to improve deliberation and thus minimize the inequali ty of resources, capacities, and 98 ). However, recognizing the possibility of strengthening citizen participation as a right in the governance process, this author indicates that far more needs to be learned about how such spaces work, for whom, and with what social justice outcomes. While Gaventa and other scholars acknowledge the advantages of the building outcomes of participatory governance, they also recognize that such outcomes exist only under certain enabling conditions. For example, constructing an effective set of rule s for deliberative processes remains a challenge, and scholars suggest that more research is needed on how different norms of delib erative practice act to include, exclude and impact the outcomes of the process (Martin 2012, 168; see also Gaventa 2004; He ndricks 2008 ; Young 2000 ). In addition, given the difficulty of determining the quality of the outcomes of any participatory exercise, many scholars suggest the need to consider which aspects of the process are desirable and to measure the presence or q ual ity of these process aspects.
54 A great deal of recent work addresses the difficult relationship between the increasing diversity of public values and the appropriate design of institutions for public deliberation ( Ansell and Gash 2008 ; see also Bohman 1996; Phillips 1995; Williams 1998; Young 2000 ). In general, empirical work on deliberative democracy may be distinguished by its focus on the procedural components of deliberation, often called the inputs and the substantive outcomes of deliberation, often re ferred to as the outputs (Scharpf 1999; see also Papadoupolos and Warin 2007). Thompson (2008) provides a review of the different analytical approaches adopted by researchers. Although a number of studies debate whether d iscussion meets the standards to be considered deliberative, others empirically evaluate whether such discussion produces any of its theoretically claimed benefits. R esearch on collaborative governance has recently developed useful frameworks for examining the contextual conditions, design features, and process issues that some scholars argue must be considered in the development and maintenance of effective collaborative governance systems ( see Ansell and Gash 2008 ; Bryson et al. 2006; Emerson et al. 2012 ). Based on the literature on coll aborative g overnance, Ansell and Gash (2008 ) identify several fundamental consider critical for the procedural legitimacy of the collaborative governance pr ocess (555). These design principles i nclude (Ansell and Gash 2008 process itself that emphasizes that the process must be open and inclusive; (2) the opportunities for each stakeholder to contribute to decision making ; and (3) easily accessible and transparent information as well as the clear and consistently applied ground rules that assure stakeholders that the process is fair, equitable, and open (555 556) According to these authors, the process
55 must be open and in clusive because only groups that feel they have had a legitimate opportunity to participate are likely to develop a ). Literature also suggests that the legitimacy of the process depends, in part, upon tions that t (Ansell and Gash 2008, 557). Clear and consistently applied ground rules reassure stakeholders that the proces s is fair, equitable, and open (ibid) In this way, process transparency means that stakeholders can feel confident that deliberation is genuine. This feature is particularly important considering the power involved in such spaces and the possibility of manipulation ( ibid ) Collaborative governance scholars and deliberative democrats both find that devel oping decisions through deliberation increases the chances that such decisions might be regarded as legitimate by citizens, and thus enhances the possibility of successful implementation and the perceived legitimacy of the system (Mendelberg 2002). As Youn g articulates, t he normative legitimacy of a democratic decision depends on the degree to which those affected by it have been included in the decision making process and have had the opportunity to influence the (Young 2000, 5 6 as cited in Parkins and Mitchell 2005, 534 ). Moreover, the main assumption behind deliberative democracy is that democratic political systems cannot survive without mechanisms securing their legitimacy (Abels 2007 105 ). Legitimacy, however, has several different i nterpretations. Institutional theorists argue that legitimacy building is the driving force behind decisions as they pertain to organizational strategies and structures (Meyer and Rowan 1977; DiMaggio and Powell 1983; Zucker 1987) and that societal accept ance of the organization and its subsequent survival depends on its attain ment of support from relevant entities with which the
56 organization interacts ( Human and Provan 2000; see also Baum and Oliver 1992; Ruef and Scott, 1998 ). 8 Considering the various s ources of legitimacy, some scholars consider legitimacy not as a taken for granted quality, but as a property constructed in and through specific processes of gove rnance (Connelly et al. 2006, 270; Beetham 1991 ). Beetham (1991 ) proposes that shared beliefs are both about acceptable process and about whether a process delivers adequate and adequately distributed benefits. Scharpf (1999) defines these as the of legitimacy, which in a purportedly democratic process essentially re evaluation of how a process allows them to influence the process and if it delivers ac ceptable results (Connelly 2006, 270). Some models of democracy clearly favor input legitimacy (e.g. participatory or pluralist models) and can be la ocess orie nted Schmidt 1997, 26 as cited in Abels 2007, 106) Although the two elements a re to some extent separable (Sch arpf 1999), there is a continuous, dialectical relat ionship between them, and in practice they cannot be disengaged in democratic systems (Papadopoulos 2003, 484 as cited in Connelly 2006, 270 ). Recently, network scholars have also drawn attention to the concepts of internal and external legitimacy, where legitimacy is understood as critical for maintaining the status and viability of an organization (Suchman 1995). However, there has been little empirical research ) work on legitimacy, Human and Provan (2000) first introduced the concept of legitimacy into the network governance literature with a focus on internal and external legitimacy, defining legitimacy as a generalized perception that the actions, activities, and structure of a network are 8 My research does not focus on the legitimacy of the political system, in which legitimacy may mean the acceptanc e of the political system, the outcome of policy processes, and the quality of policy making (Abels 2007).
57 desirable and appropriate ( see Suchman 1995 574 ). These authors propose three necessary and distinct dimensions as critical for networks interpreted as (1) the legitimacy of the network as a form that can attract internal a nd external support and resources, (2) the legitimacy of the network as an entity that is recognizable to both insiders and outsiders; and (3) the legitimacy of the network as an interaction that builds trust among members to freely communicate within t he network (Bryson et al. 2006, 47). Based on this interpretation of legitimacy, a key concern for any governance structure is to develop internal legitimacy among participants. According to Provan and Kenis (2007), if participants do not see interactions an d coordinated efforts as being a legitimate way of conducting business, with potential benefits from these interaction, then the network is likely to exist in name only with little real commitment by participants to n etwork level goals and outcomes (243). In addition, network participants need to believe that collaboration with one another is beneficial. On the other hand, any form of governance must be responsive to external expectations, such that outside actors must recognize the network as an entity i n its own right rather than a group of organizations that occasionally get together to discuss common concerns (ibid). Some network authors have considered the way in which the social forces and economic pressures surrounding a network or what they refer effective ( see Mandell and Keast 2008 720 ). These authors suggest that the external environment may have an impact on how the network is perceived by outside, relevant stakeholders. Moreover, havi ng external legitimacy may also reinforce the commitment of network participants who are more likely to see themselves as part of a viable network (Provan and Kenis 2007 243 ). Along these lines, in their work on network performance, Mandell and Keast ( 2008) also stress the importance of the continuing support given to the network by
58 relevant stakeholders outside the network that provide the framework and legitimacy to function in this way ( Mandell and Keast 2008, 720; see also Keast et a l. 2004 ). 9 Furt hermore, the work discussed illustrates that the inherent tension in constructing legitimacy requires building internal network interactions while at the same time, building the credibility of the network to outsiders (Provan and Kenis 2007). According to these authors, effective network governance means building structures that are responsive to both internal and external legitimacy needs. As follows legitimacy management is considered to be a continuous process that involves gaining, maintaining, and i n some cases regaining legitimacy for the organization ( Massey 2001 156 ). Legitimacy is also considered to be conditional and subject to challenge, particularly when social norms and institutional structure s are changing (Connelly 2011 932 ; see also Bee tham 1991 ). Moreover, as Huxham and Vangen (2005) found, collaboration is frustrating and collaborative inertia is a major challenge to overcome. Collaborative engagement processes have been studied from various disciplinary lenses, and scholars have iden tified numerous positive outcomes from successful engagement such as improved clarity on key issues ; en hanced trust and mutual respect; inc reased decision making capacity; and greater perceived legitimacy within and outside of the collaboration (Agranoff and McGuire 2003; Bryson et al. 2006; Emerson et al. 2009; Fung 2006; Leach and Sabatier 2005; Milward and Provan 2000). Some researchers have also identified conditions or factors that they consider to be contributions to good performance, such as the inc lusion of stakeholders; partner selection; mutual trust; honesty and reliability; shared vision; mutual interdependence, open communication; appropriate distribution of power, and political influence ( Huxham and Vangen 2006; see also Gray 1989; Mattesich e t al. 2011 ; Sherer 2003 ). 9 According to Suchman (1995), legitimacy leads to the persistence of organizations since they receive such support resources from other actors.
59 Based on the theoretical propositions of deliberative democracy and collaborative governance, empirical research on governance processes expect that decisions or actions produced through strong engagement processes will be fairer and more durable, robust, and efficacious ( Emerson et al. 2012 ). However, there is limited research on the quality of collaborative processes ( Emerson et al. 2012, 12; see also Bingham and O ). Furthermore, while the literature on collaborative g overnance considers legitimacy to be a condition for effective collaborations, it does not clearly define legitimacy. In this research, I adopt an approach that includes both internal and external legitimacy discussed by network governance scholars, where the former is grounded on the importance placed on the procedural components of deliberation as suggested by deliberative democracy theory It is expected that the creation of legitimate spaces for the involvement of multiple stakeholders offers greater op portunities for citizen participation in the policy process by engaging cit izens with democratic practice and increasing the quality of information available to government on ci preferences. This Research In this research, I consider the role of several forums in which civil society actors resolve natural resource governance issues emerging at the local level and those introduced through policy frameworks at the national level. Considered weak public spheres, these forums were initiated by and composed of representatives of different rural communities and social organizations. I set out to understand the internal and external legitimacy of these forums in the context of participatory governance from which they were inspired. Internal legiti macy is what
60 10 and is examined based on the following principles: (1) stakeholder representation; (2) equal opportunity or inclusiveness; and (3) process transp arency (see Ansell and Gash 2008 ) 11 External l egitimacy is defined as the recognition of the forum by actors outside of the forum (Human and Provan 2000). Considering the importance placed on the involvement of different actors in the governance process, I combine the network governance and network a nalytical approach (Provan and Kenis 2007). Whereas the former views the network as the unit of analysis and as forms of social organization, the latter views the network as a set of actor or nodes, with relationships between these nodes as being either pr esent or absent (ibid) Thus, networks are considered to vary with regard to their structural patterns of relations. Consistent with this logic, I employ concepts associated with social network analysis (SNA) to examine several network structural features of the forums related to the governance arrangements of each forum. Social network structure has implications for the ability of groups to solve problems and act collectively in a natural resource governance context ( Bodin et al. 2006). Furthermore, t he ne twork characteristics that I examine provide context, establish social structure, and enable and constrain actor behaviors within the networks All of these elements influence the role and development of network leaders (Provan and Kenis 2007). In this wa y, I consider the role of individual actors and their agency that is affected by both the nature of th e links within a social network and the overall network structure. 10 process including agenda setting establishing procedural rules, selecting the information and expertise to inform the process and assessing 11 These principles are not meant to capture all dimensions of democratic governance, yet are relevant to guide the empirical analysis of assessing the democratic quality of the public fo rums for partici patory decision making processes.
61 Based on previous research on various structural network variables, 12 I focus on the mes o (forum) level and include an analysis of (1) the whole network structure of the forums (network cohesion); (2) the structural network position of each participant in the forum (degree centrality, betweenness centrality); and (3) the cross level exchange between forum members at the meso level and actors at the macro (national) and micro (local) levels. Moreover, I examine how the (Figure 3 1) Figure 3 1. Conceptual map displaying the cross level linkages involved in constructing internal and external legitimacy within a multi level governance system. In the next chapter, I present a descrip tion of the three forums selected to conduct this research. I discuss the geography and legal and environmental context of each forum in addition to the background of the problem motivating its creation. 12 I discuss the network terminology in more detail in Chapter 4.
62 CHAPTER 4 FIELD SITES The field sites are defined by the central location of the forums studied. Recall from Chapter 1 that these forums represent a space in which grassroots actors resolve natural resource governance issues emerging at the local level and those introduced through policy frameworks at the n ational level. After explaining the process involved in the selection of cases, I present a background description of each case selected for this research. Inventory of Cases and Case Selection I began an inventory of cases (or forums) during my preliminar y research from June to August 2010 and completed this in t he first two months of my field work in 2011. Of the cases identified in the inventory of forums (Table 4 1), I chose those that share similar characteristics, such as: (1) level: takes place at the meso level; (2) age: at least five years old; (3) size: composed of at least 15 members; and (4) impetus: created by local communities and organizations themselves to represent local environmental concerns 1 The forums selected for my study were all initi ated by grassroots organizations and are composed of representatives of social organizations or communities unified by a common environmental concern. Based on the theoretical model developed and explained in Chapter 3, it was first important to identify those cases that occur at the meso level or in between the micro and macro levels. I set a five year minimum age, as I was interested in those with enough experience to examine the study variables as defined in Chapter 3. I also determined that forums with at least15 members were eligible in order to meet the recommended sample size to conduct a whole network analysis. Table 4 2 presents additional characteristics of the forums that are important to consider. 1 These characteristics were determined once I completed the inventory.
63 Geography Bolivia makes up an area of 1,098,581 km (Figure 4 1). Preliminary results from the 2012 census indicate that Bolivia has 10,389,913 inhabitants which indicate a 26% increase in its population from 2001 with 8,274,325 (INE 2013). Bolivia is made up of nine departments that are integrated in the state government system t hrough departmental governments. Composed of highlands, valleys, lowlands and Amazonian forest, Bolivia is considered to be one the most biodiverse countries in the world. Coordinator of Affected Communities of River Desaguadero, Ur Ur y Poop (CORIDUP) Site Location and Context The Department of Oruro has a population of 490,612, which increased by 25% since 392,769 in 2001 (INE 2013). Oruro is located in the An dean central highlands of Bolivia, between the Andes and the West where the altitude ranges between 3,600 and 4,000 meters. C haracterized by a cold and dry climate, the soil is usually difficult for farming bu t rather suitable for cattle ranching. Oruro w as considered the second largest producer of silver during the mining boom in the early seventeenth century (Marquez 2006). After the nationalization of mines in the 1952 Agrarian Reform, small scale mining emerged and by 1967 and was responsible for 25% o f the total value of minerals produced and exported ( Marquez 2006; see also Capriles 1977). While mining has been a staple contribution to the Bolivian economy since colonial times, it has been managed in an environmentally irresponsible way. In the depart ment of Oruro, approximately 300 mines are located along the two mountain ranges. The Desaguadero River and Lakes Ur Ur and Poop have traditionally served as the most important sources of freshwater in the central highlands and are surrounded by these m ines. After using the water, most of the mines
64 discharge all effluents containing acids, heavy metals, and dissolved solids directly into the water without any treatment, which produces considerable consequences for the environment and the inhabitants of d ifferent communities (Mollo 2009). Given the amount of water used in the mining sector, one of the most serious consequences from mining practices for the near future is the scarcity of fresh water (ibid). Legal and Environmental Context Both the 1992 Envi ronment Law and Mining Code include regulations that require mining companies to have an environmental permit to operate legally. The Mining Code requires all companies whether public or private to mitigate the environmental damage arising in concessions a nd mining. All mining operations must present documents purporting the activities, works or projects that serve as an assessment of their environmental impact and a plan for preventative measures to mitigate the generation of further environmental impacts (Mollo 2009). While the Environment Law regulates the activities of mining activities for the purposes of protecting the environment, most mining companies have not complied with the regulations. 2 The minimal efforts made by the mining companies to comply with the environmental legislation as well as the porosity of the legislation and weak institutional infrastructure to control such operations explain the increasing environmental conflicts that have arisen between the many affected communities from contam ination and the mining operations. Environmental problems caused by the irrational and excessive exploitation of mineral resources without considering environmental control measures, has come to characterize the area surrounding mining operations in Oruro The r esulting threat and risk of this contamination to 2 Mining operations are classified into three groups large, medium, and small. All o f the large mining companies belong to the state and must obtain their licenses to operate; however, few medium and small mining operations have their licenses.
65 the health of the residents has caused intense conflicts within the communities surrounding the river, as they depend on the water not only for consumption, but for their land and their livestock. Si nce the early 1990s, communities surrounding Oruro began to manifest changes in their water, soil, and biodiversity. 3 As all of these problems have a negative impact on the social, economic, and cultural development within the communities, they have starte d a series of complaints to authorities, demanding a solution to this complex problem through CORIDUP. Creation of CORIDUP In response to the ineffectiveness and perhaps indifference of the government authorities addressing the environmental consequences o f mining activities, communities affected by the society organization to stop mining pollution and fight for their environmental rights. The communities decided to call the organization CORIDUP (Coordinator of Affected Communities of River Desaguadero, Ur Ur y Poop) and in January 2007, CORIDUP elected its first board of officers, which represents the affected communities in four river sub basins: Desaguadero, Huanuni, Poop, and Antequera Canyon Among the greatest achievements of CORIDUP was the approval of Executive Decree 4 No. 0335 on October 21, 2009, which declares the zone as applies to the municipalities of Huanuni, Machacamarca, El Choro and Poop in the department of Oruro. 5 3 Among the consequences are high salinity in surface and ground water, salinity and sedimentat ion of soils, water and soil pollution by heavy metals (Pb, As, Cd, Zn, Cu), loss of crop and livestock production (diseases and genetic deformities), loss of vegetative cover (desertification and genetic deformities), loss of quality and quantity of irrig ation water and consumption progressive disappearance of biodiversity, as well as the migration to cities of the community members, particularly the youngest generation ( CEPA/LIDEMA/UTO 2010 ). 4 A legislative act approved by the President. 5 Refer to Perra ult (2012, 8) for a map of the Huanuni River valley.
66 c meetings with different actors as well as CORIDUP meetings, they often clarify that they are not ant summarized the effects of mining on communities and the mission of CORIDUP: There has been contamination since the mining operations started using D UP wants to encourage solutions contamination exists. Since 2006, we have been fighting. The companies have ignored us, put us down, and even hired [or paid off] communal authorities so they eate factions in our environmental justice. CORIDUP is composed of 20 representatives of the affected communities, including a board of officers or leaders (president, vice preside nt and secretary) and four representatives from each of the four affected municipalities who serve as environmental representatives of their communities. The officers are democratically elected every two years and forum members are elected by their communi ties depending on their respective election cycles. One important reside in the communities they represent, but rather maintain primary residence in the city of Oruro 6 Although this reflects a common pattern of migration due to mining contamination, it may also affect the representative nature of the forum. The first board of officers was elected in 2007 and was re elected for another term in 2009. As of January 2013, however, the same individuals continue to serve by de facto as elections have not yet been held. Meetings are held 6 Migration is common given the contamination that has made agricultural production an unreliable source of income.
67 every few months depending on when the leaders decide they have enough concrete information to share with the rest of the members. Wit h the technical assistance of the NGO, CEPA ( Center for Ecology and Andean Peoples) the board of officers developed a proposal for The Executive Decree 0335 (ED 0335) leaders and members made 56 trips over a 10 month period in 2009 from Oruro to La Paz (four hours) to work on and present the proposal to the Ministry of the Environm ent and the Ministry of Mining. In response to the escalating levels of contamination in the zone and the encouragement from the Vice Ministry of the Environment to take advantage of the cabinet meeting in which the Executive D ecree would be discussed, CO RIDUP, along with other organizations of Oruro began a mobilization to the city La Paz on September 26, 2009 to demand the approval of ED 0335. They were stopped on their way into the city by a government authority who advised them that President Evo Moral es would enact the Executive Decree that same day. According to the Vice Minister of the Environment at the time, who worked close ly with CORIDUP to develop the Executive D Executive D ecree because they though effort to convince them [ cabinet members ] 7 placed on the government to pass the Executive D ther Executive D ecree has been passed with such significance for th e environmental sector as the ED 0335, as it was first Executive D ecree that included a plan with concrete actions for the government. 7 Includes the president and the Ministers of the government ministries.
68 e of the most successful experiences because of how they achieved the Executive On December 16, 2010, the national government and departmental government of Oruro made an im proposed "Program for Sustainable Management of Natural Resources of the Lak was part of the ED 0335. The project involves an investment of 14 million Euros to be di stributed to four municipalities within the Poop watershed (Huanuni, Machacamarca, El Choro, Poop, Paza, and Antequera). Considered in line with the E D 0335, this funding went ed by the state government for any projects geared towards environmental remediation of this zone and manage these funds. The implementation plan outlined in the ED 0335 includes six working groups, each led by a different Ministry. The Ministry of the Env ironment is responsible for all six working groups and leads one specifically focusing on environmental remediation in the four sub basins. The other Ministries involved include: the Ministry of Rural Development and Land, Mining, Education and Health. Eve ry six months, they are supposed to provide an update on their progress to civil society actors (community members, CORIDUP, miners, mining companies) in releva nt questions and express concerns. Besides monitoring the progress of different actors involved in the ED 0335, CORIDUP is also conducting a monitoring process to the Environmental Audit Mining Operation of EMIRSA Kor i Kollo (Inti Raymi) (Mollo 2009 ).
69 CORI DUP has maintained a very close working relationship with the NGO CEPA since its creation. In addition to providing the legal and technical assistance needed to sustain its actions, CEPA also provides a central meeting location for forum officers and membe rs. Environmental Justice O ffice who assist them with their legal and technical actions. Of particular lf of the affected these issues. The secretary of CORIDUP is also an employee of CEPA. T hese leaders have gained experience organizing meetings with different actor s and building their negotiation skills in dealing with state and private actors, particularly the mining companies. In this way, CEPA provides an enriching learning environment that serves to train leaders in their efforts to affect change on the environm ental regulations controlling mining operations. Block of Campesino and Indigenous Organizations in the Northern Bolivian Amazon (BOCINAB) Site Location and Context The region of the Northern Bolivian Amazon includes the Department of Pando and the provin ces of Vaca Diez in Beni Department and Iturralde Province in La Paz Department 8 and has an area of approximately 100,000 km 2 or an extension of 11,300,000 ha. Moist tropical forest characterizes the region with an annual precipitation of about 1700 mm dur ing the wet season from December to April and a dry season from May to September (Zuidema and Boot 2002 ). According to Killeen et al. (2007), tropical rain forests cover around 40 million ha of the Bolivian lowlands, correspondin g to 90% of its original ex tent yet d eforestation rates exceed 0.5% and tend to increase over time 8 Refer to de Jong et al. (2006, 449 ) for a map of the Northern Amazon region.
70 Bordering Brazil and Peru, the Northern Bolivian Amazon is considered the most geographically isolated and least populated region in the country. Cross border movements of borderland resident populations have been common between these countries According to the 2012 census Pando has a population of 109,173 and 425,780 in Beni (INE 2013). Since the early 19 th century, international market demand for forest products from the Amazon re gion has stimulated various waves of migration from the highlands as well as periods of intense production of forest products in the region. In particular, the Northern Amazon region is known for Brazil nut production, and although this region consists of two separate departments, it is considered economically integrated since the majority of the Brazil nuts harvested in the region are produced in one of the departments, Pando, and processed for international export in the other department, Beni. Brazil nu t production is the most important economic driver in the Northern Amazon region today, but this was not the case until the 1980s, when Brazil nuts first outpaced rubber in economic importance. 9 The demand for Brazil nuts skyrocketed internationally in the process the nuts for international export. Additionally, due to Brazil nut harvesting, the Northern Amazon forest experiences substantial yearly migration from regional municipalities ( Llanque 2006 ). called barracas, 10 either moved to other areas or s tayed on the estates, which were transformed for Brazil nut production (Assies 2003; Henkemans 2001; Llanque 2004). In association with the campesino 9 10 These estates were transformed to B razil nut estates in the 1980s where land owners or patrons enslaved indigenous peoples to collect Brazil nuts
71 communities; however, these com munities had no tradition of self governance (Llanque 2006). In 1994, an Executive Decree imposed a size limit on land held as a barraca and allowed barracas As a result, the total area held by b arracas in Bolivia declined by half (from 3.5 to 1.8 million hectares or ha ), while communal lands increased from 60,000 ha in 1984 to an estimated 3 million h a after titling (de Jong et al. 2006, 451). During this time, campesino and indigenous groups mig rated to the region from other departments because of the profitability of the Brazil nut. Legal and Environmental Context Land tenure issues have played a prevalent role in Amazon region, the political framework of land ownership continues to transform the landscape. As discussed in Chapter 2, the implementation of decentralization policies, in particular the Law of Popular Participation, in the mid 1990s, allowed rural settlements to obtain legal status and receive funds made available to municipal governments under the new decentralization legislation. Many of these newly reco gnized settlements called OTBs also obtained new legal status as campesino communities under the INRA Law of 1996. In addition to these refor ms, the Forest Law of 1996 11 fundamentally changed access to state forestlands, by allowing new social groups (indigenous and campesinos ) to engage in commercial forest exploitation in addition to conventional forest entrepreneurs ( de Jong and Ruiz 2012; se e also Pacheco 2007 ). Two Executive D ecrees, 25848 from July 2000 and 27572 from June 2004, gave indigenous and campesino communities in northern Bolivia the right to 500 ha of communal forestland (de Jong et al. 2006 451 ). As a result, the area of forest controlled by the corporate sector was reduced significantly. 11 This law introduced new measures including long term forest concessions, area based forest fees, and the allocation of forest ownership rights to private landholders including indigenous grou ps (Pacheco et al. 2010).
72 The government of Evo Morales has criticized the forest sector reforms enacted sin ce 2006. Morales points to the market oriented approach, inequitable distribution of benefits, and uncontrolled illegal logging that results from these reforms (Pacheco et al. 2010). Parties in the Northern Amazon region find that the greatest weakness of the current Forest Law is its failure to recognize the unique diversity of forest resources in the region, where numerous products other than timber are produced. Purportedly, the development of the Forest Law was dominated by stakeholders from the Departm ent of Santa Cruz, a department known for its timber production As one key informant stated, t forest t hat is not Amazon forest. In the Santa Cruz forest, timber is the most important resource because of its importance in the world market and there was no other, more important resource. Here in the Amazon forest, the most important resource is the Brazil nu t. In response to the lack of progress made on the 1996 INRA Law by 2003, campesino and indigenous organizations of the Northern Amazon began a mobilization in defense of the untitled territories and land in the Northern Amazon. This march led to the gover approval of the Law No. 3545 or the Renewal of the Agrarian Reform in 2003, which promoted the economic financing of the implementation of land reclamation. In September of 2006, INRA signed an institutional cooperation agreement for the conclusio n of land titling in Pando, extending the completion of the land titling process from seven years to 2013 (INRA 2010b). However in August of 2008, after five years of intensive work from communities, organizations, NGOs, INRA, and others, the government o f Evo Morales announced the 100% completion of land titling in the department of Pando (6.3 million ha (ibid), 12 the first department where this has been accomplished. 12 The process is still not complete in the province of Vaca Diez in the department of Beni.
73 of the urgency for land security in the Northern Amazon was for many what illuminated BOCINAB an important player in the region. An important milestone for the Amazon region was established in 2008 at the Constituent Assembly which began revising the state constitution in 2007. With approximately 70% of the Amazon territory in Bolivia (including Beni, Pando, Santa Cruz, and parts of Sucre and Tarija), there was significant pressure on the state to recognize the Amazon region. 13 Chapter 8 of the Plurinational State Constitution includes two articles that recognize the Amazon region, previously ignored in prior legislation. Creation of BOCINAB Responding to the insecurity of land and territory under threat from loggers and other actors, campesino and indigenous organizations as well as other actors in the Northern Amazo n region joined together to form BOCINAB in 2002. According to one informant from an NGO that worked closely with BOCINAB at the time There came a time when the tension escalated and the unity between campesinos t campesinos and indigenous decided to create BOCINAB with the help of NGOs that supported the creation of an organization that would represent the interests of both sectors with demands rooted in community land rights. For the first time, these two secto rs came together with the same mission to promote the conclusion of the land titling process in the Northern Amazon. In the words of one member, "it was a wake up call ... before we were truly blind because never before had these two sectors (the most vuln BOCINAB represents campesino and indigenous groups organized in social organizations in the departments of Beni and Pando. It serves to voice the 13 The Constituent Assembly involved a variety of social movement actors including the Amazon Commission, made up of 11 constituents repr esenting the Amazon.
74 demands of rural populations in the region that have traditionally had very little rep resentation at the national level. BOCINAB was instrumental in organizing the march in 2003. BOCINAB member organizations, along with several NGOs, INRA, and the Church formed what was referred to as the Inter agency Committee which set out to train and p by helping establish community documentation and define their boundaries. Local indigenous and campesino advocates were trained as legal advocates that accompanied INRA in the process of titling the communities in Pand o and Beni. The land titling process allowed campesinos to become citizens and process their land rights recognized by the State. Until then barraqueros 14 and other sectors doubted the existence of the campesino communities. The consolidation of land right s in the Northern Amazon was considered the principal motivation for the formation of BOCINAB, and the completion of land titling in Pando in 2008 represents its major achievement. Through in depth exploratory interviews with former members, I found a stro ng sense of n yet they were the same rmants emphasized that the vision of 14 Owners of barracas.
75 BOCINAB is composed of 10 member organiza tions and 15 members representing these organizations. Its board of officers (First, Second and Third Coordinators) are elected on a rotational basis every two years so that each member organization has the opportunity to lead the forum. The official repre sentatives or members of BOCINAB are the presidents of the respective campesino and indigenous organizations that are elected democratically at the organizational level with varying election cycles. 15 Of the nine member organizations, five are located in Ri beralta, three in Cobija, and one in Guayaramern. In addition, seven are campesino organizations and two are indigenous. Two member organizations have additional representatives. 16 The frequency of forum meetings ranges from monthly to several times a yea r yet varies in response to the urgency of particular issues. Meetings are to be held on a rotational basis in the organizational offices of each of the member organizations between the three cities of Riberalta, Guayaramern, and Cobija. Given the locatio n of the majority of member organizations in Riberalta as well as the location of the current president, however, meetings are most frequently held in Riberalta. The 12 hour travel time by bus between Riberalta and Cobija makes travel between the two citie s both time consuming and costly. The NGOs that played an instrumental role in the development of BOCINAB continue to provide financial and technical expertise to the member organizations and the forum itself. 17 15 The campesino is every five years. In addition, most organizations allow for the re election of their presidents 16 represent different indigenous groups in the organization) that allows them more representatives and provides them a voice considering the majori ty of organizations in the region are campesino The regional indigenous organization located in Riberalta has three official representatives. In addition, the campesino organization located in Guayaramern has two representatives given that this is the on ly organization in the area. 17 They are considered advocacy NGOs that focus both on extension projects (i.e., community agroforestry projects) and judicial processes involving land rights.
76 Each of the four NGOs supporting BOCINAB has one to two individuals who attend meetings and provide assistance to the member organizations at any time. 18 travel to meetings, these NGOs assume the role of interlocutors between the involved member organizations and the governm environmental policy at the national level. Esperanza Dam 19 BOCINAB demanded the government accurate informat ion on the impacts that the dam would have on the populations surrounding the proposed construction site. With the aid of an environmental NGO, BOCINAB wrote letters to the appropriate government ministries to solicit information on the environmental impac ts of the project, but according to former members, the government never responded to these letters. Instead it accused BOCINAB members of being environmentalists and part of a right winged opposition to the government both accusations derived from its wor k with NGOs. 20 BOCINAB has the potential to play an important role in voicing the demands of the campesino and indigenous sectors through developing proposals for the emerging national environmental laws. These laws include the Amazon Law, Forest Law, and L and Law. Its 18 The NGOs coordinate among themselves and share sponsoring meetin gs so that no more than two NGO representatives are present at one time. 19 Cachuela Esperanza hydroelectric project would be located in the Beni River, near the town of Cachuela Esperanza, in the department of Beni. The project is expected to produce 990 M W, equivalent to 80% of the energy currently produced by Bolivia. The aim, however, would be to export most to Brazil. Its construction is expected to take 7 to 8 years and would require an investment of 2,000 million dollars. The project is also part of t he controversial IIRSA Initiative (Integration of Regional Infrastructure in South America). The feasibility report, ordered by the Bolivian government, shows that the environmental and social damages would involve the flooding of more than 900 km2 of land which is expected to affect around 100,000 people. Despite these potential impacts, the government considers the project beneficial and has moved forward with it (Conservation Strategy Fund 2012). 20 perspective of NGOs.
77 current agenda includes several proposals including their demands for the legal security of land in the face of natural resource extraction from the state and other actors. 21 Integrated Management Area of Apolobamba (ANMIN A) Protected Areas a nd C o management in Bolivia One fundamental strategy for the conservation of biodiversity in Bolivia from 1939 to 2004 was the establishment of protected areas, 22 of which are considered national ter ritory covering 182,716 km or approximately 15 % of the national territory ( SERNAP 2006 ). communities, of which 70% are campesino indigenous and intercultural communities (SERNAP 2011). Since its creation in 1998, the state institution, the National Service of Protected Areas (SERNAP), has been responsible for the management of all protected areas in Bolivia. In contrast to other countries, the management of protected areas in Bolivia involved the populations living within t he areas cultural diversity of these areas. The LPP promoted the increasing demand of the populations living within these areas organized in social organizations for greater participation in th e management of protected areas. In 2006, in a time of institutional instability and growing from the protected areas led a march to reclaim their rights. They present ed a proposal entitled participated in decision making concerning protected areas. In 2007, SERNAP and the communities worked on developing a proposal for co manage ment that would officially involve 21 According to Pacheco et al. (2010), as time passes, the uncertainties of the forestry regulatory system the continuation of forest concessions, and the slow progress made on land tenure regularization in addition to the increasing c ompetition for access to forests in indigenous territories and remaining public lands have all increased illegal logging (275).
78 the participation of different actors the state and communities in strategic administrative decisions within protected areas. They worked on developing a proposal for a Executive D ecree that would legalize co management i n protected areas. Despite efforts made to advance this proposal between 2007 and 2010, in 2011, SERNAP changed its agenda, bringing the attempts to legalize co management to a screeching halt, further discussed below 22 Site Location and Context In Januar y 1972, the Bolivian state created the National Wildlife Reserve Ulla Ulla, an area of 240,000 ha, with the objective to protect and conserve the vicua ( Vicugna Vicugna ), considered in danger of extinction, as well as the flora, fauna, and native ecosyste ms considered endemic. In 1977, the National Reserve was recognized by the Organization of United Nations for Education, Science and Culture (UNESCO) as a World Biosphere Reserve. In January 2000, the surface area of the Reserve was expanded to 483,743,800 ha and reclassified as the National Natural Area for Integrated Management of Apolobmaba (AMMIN A ) (Plan de Manejo 2006, 2). 23 According to the ED N 25652 or the Regulation of Protected Areas, ANMIN A tural and renewable resources in line with its management plan that sets out to balance sustainable development and the conservation of ANMIN A is located in the western most part of the Depa rtment of La Paz in the Provinces of Bautista Saavedra, Franz Tamayo, and Larecaja, encompassing the municipalities of Charazani, Curva, Pelechuco, and Mapiri. The area has distinct ecosystems: highlands, 22 According to key informants, the proposal developed by the communities in the protected areas was never forwarded from SERNAP to higher levels of government for approval. 23 Over 1500 plant species and more than 300 animal species are endemic to the area. ANMIN A is also located in the Conservation Corridor Vilcabamba Ambor as part of the international effort by Conservation Inte rnational to protect the most biologically diverse regions of the world (SERNAP 2006).
79 va lleys, sub tropics, and tropics. It is comprised of two distinct areas characterized by disparate climactic conditions, primarily due to the altitudinal range from 1250 m in the eastern most part (Mapiri) to 5600 m in the mountain range in the western most part. A variety of production systems exist in e ach of the ecological zones. In general, the highlands are home to vicua ( Figure 4 2 ) and camlido or alpaca (Figure 4 3 ), 24 fishing, agriculture and mining; 25 whereas in the valleys and the lowlands mining is the principal source of income with livestock a nd agriculture as secondary sources of income. Communities that make a living from the management of vicua and alpaca benefit most from the integrated management of resources within the area. 26 The organizational structure of the population is characteriz ed as with different organizational structures. CONAMAQ, CSTUCB, and the Intercultural organization is the most predominant in the area. While the union form of organization has expanded, the ayllu remains present as a sociocultural unit in the area. 27 The leaders of these organizations are referred to as the mallku, or authority that serves a political function and fulfills a political and administr ative function. The 1952 Agrarian Revolution generated structural change in this area and the agrarian unions were strengthened as a result. 24 The ideal pasture for vicuas and the domesticat ed species of alpaca (Figure 4 2 ) are located at an altitude higher than 4.000 m with a cold and humid climate (Meyers 2002). 25 Mining activities began in the 16th Century in ANMIN A, particularly in the lowlands (Conservation International 2003). Mining activities have recently expanded in the latter half of the 1980s in Pelechuco and Charazani (Araucaria 2004). 26 Vicua fur is sold once a year at a price of approximately $480/kilo whereas camlido fur is sold several times a year at approximately $30/kilo. As an endemic species, vicu a meat is illegal to sell in Bolivia 27 See Chapter 2 for a description of this organizat ional unit.
80 Projects promoting the management and use of vicua were a main component of the 00 to 2007. Prior to these protection initiatives, the species was in danger of extinction due to pressure on this resource for the high quality of its fiber and its high market value at the national and particularly international level. Approximately 27 communities or 30% of the communities within the area have participated in this activity which has provided a positive option for the development of sustainable activities. According to SERNAP (2012, 12), from 1999 to 2008, the population of vicua and the amount of fur sold has increased every year, which indicates that the conservation measures taken to increase the population of vicua have been effective. Since the creation of ANMIN A, local communities were very skeptical of the protected area, which l ed to the suspici on of communal authorities of A NMIN As a result, communities first rejected projects introduced by international instit utions within the involved in these projects. With the expansion of the concept of social participation into co management in 2001, the communities gained more con began to participate in programs promoting vicua management. Legal and Environmental Context Mining activities in ANMIN A represent one of the principal economic activities for the population and at the same time i s considered the principal threat to the environment (SERNAP: 9 10). With the escalating price of gold since 2008, mining practices have rapidly multiplied in the area. As of 2011, 48 of the 88 (or 56%) mining cooperatives did not have rights however 26 (3 6%) have been controlled (SERNAP 2012). The intensity of mining activities is expected to have a negative impact on ANMIN A. According to SERNAP (2012), the mining cooperatives
81 operate illegally, and in the majority of the cases, they have not complied wit h adaptation or mitigation measures, resulting in a greater environmental impact. The need to take actions to avoid further environmental degradation from mining A lacks the necessary economi c resources or institutional support to make these operations comply with the legal norms presented in the Environment Law. While the Regulations of Protected Areas, passed in 1998 does not allow mining activities in protected areas, it also does not prohi bit such activities. Rather, protected areas with mining operations must develop zoning plans to regulate the expansion of such activities into conservation areas. The way in which mining activities within ANMIN A have escalated is characterized as both i ntense and chaotic. The weak enforcement of regulations via sanctions has contributed to the expansion and continued operations of illegal operations. A lack of economic and human resources to enforce these norms and the time intensive process of legalizin g mining operations To illustrate the effects of mining in the area, the population of vicua is migrating to the southern end of the area because of the contamination from the increasing mining operations. Creation of the ANMIN A The management committees or forums coordination, deliberation and voluntary support for protected area management (SERNAP 2006 ). Its role in assuring t he management of natural resources in protected areas is considered the most important form of public participation in protected areas Co management committees Created in 1994 and officially established in May 1995, ANMIN A was the first management committee of all protected areas. As opposed to the earlier application of this concept in protected areas where simply the different parties
82 involved just had to sign contracts, communities that live within th e area are now involved in decision making (with a voice and a vote) through co management. 28 SERNAP is thus considered one actor or participant in the co management committees and represents the state along with local state actors, such as mayors. One key informant who works closely these management committees explains the vision of co management: T new vision of management opens the space to decision making regarding the protec making authority, thus greater participation was always a critical part of our demands. The management committee of ANMIN A is composed of 20 members, including the director of the area (an employee of SERNAP), three mayors from the municipalities of Charazani, Curva, and Pelechuco, two sub mayors, two provinc ial authorities (Franz Tamayo and Bautista Saavedra), and several representative of the mining cooperatives. Elections for president, vice president and secretary are held every two years and members change at various times based on their respective electi on cycles. In August 2012, the mayor of Curva was elected President and the mayor of Pelechuco was elected vice p resident. Prior to this election, only representatives from social organizations filled these positions. The frequency of committee meetings depends on the leadership of these elected individuals. Meetings are to be held on a rotating basis between La Paz and the major municipalities in the protected area, yet are most frequently held in La Paz, as the size of the area makes getting from one p lace to the other within the area further and more expensive than 28 Although co norms that guarantee their right to participation.
83 traveling from the area to La Paz. 29 Several of the forum members travel to La Paz several times a month for work, thus facilitating their participation in meetings held in La Paz. 30 With th e reclassification of ANMIN A as one of integrated management, it was expanded to include an additional province, Larecaja. However, members from only two provinces Franz Tamayo and Bautista Saavedra actually attend. Despite its inclusion with the expansi on of the protected area since 2000, the municipality of Mapiri in Larecaja had not been invited to meetings until 2011 and consequently has not had official representatives that serve on the forum. As of 2013, Mapiri is becoming integrated into the forum because of the initiative taken by the current protected area staff members. ANMIN A was long considered one of the most active management forums in Bolivia, in part because of its success in increasing the population of vicua that interested representat ives from distinct social organizations in integrated management activities. Members were especially motivated to participate because of the available funding from NGOs for management from 2001 2007. As of 2008, however, an alarming increase in mining oper ations in the area has introduced interests that are incongruent with protected area management. Considering the profitability of mining activities, an increasing segment of the population is becoming affiliated with mining cooperatives, including elected officials who are also forum members. Despite the involvement in mining is not considered illegal, and is thus unrestricted. 29 Public transportation is available daily from the area to La Paz (8 11 hours of travel time by bus) whereas transportation is unreliable from one town to the other within the area. 30 For example, the mayors have offices in La Paz and the indigenous authorities often attend meetings with their national social organizations in La Paz.
84 Table 4 1. Inventory of forums Forum Purpo se Level of Operation Year created Actors #Members Frequency of meetings BOCINAB To influence environmental public policy (pertinent to forest management in Northern Amazon) Regional 2001 Social orgs: 11 member organizations 4 NGOs 15 1/3month CODERI P (Consejo de la Cuenca del Rio Pilcomayo) To bring together communities affected by mining contamination along the Pilco Mayo River (the most contaminated river in Bolivia from mining) River basin 2002 Reps from 4 zones representing 50 affected communiti es (invited but 14 2x/month CORIDUP To bring together communities affected by mining contamination and stop the contamination River sub basins 2002 Reps from 4 sub basins 22 1/month Comits de Agua (Fed de Juntas Vecinales) To promote p rojects focused on the improvement of the river basin and water quality River basin 2010 Neighbor committees 70 varies Comits de Suelo To promote decision making regarding agricultural production River basin 2010 Neighbor committees 25 1/3 months SIM ACO (Suyo Ingavi de Markas, Ayllus y Comunidades Originarias) To promote decision making among different indigenous groups regarding multi resource management Indigenous territories 2005 Indigenous orgsanizations; municipalities; 1 NGO 25 1/month Co m anag ement Committee Apolobamba To promote the participation of social organizations (inhabitants in area) in the construction of management norms in the protected area; to expand the space whe re decision making takes place and involve both the people living in prot ected area and the state Protected area 1995 Campesinos, colonizers, Bartolina Sisa, 3 municipalities 22 1/month
85 Table 4 2. Characteristics of forums selected for this study Forum Populations represented Groups represented Forum members Meeting loc ation Frequency of meetings CORIDUP 4 municipalities of the Sub Cuenca Huanuni All affected and affiliated communities in the 4 municipalities 20 total; 4 environmental representative from each of the 4 municipalities Depends on ur gency (not consistent) BOCINAB The Northern Amazon (Department of Pando, Province of Vaca Diez in Beni Department and Iturralde in La Paz Department) 9 social organizations (3 in Pando; 5 in Vaca Diez, 1 in Iturralde); 6 campesino organizations and 3 15 total; the president of each of the 10 social organization (1 organization has 3 representatives and another has 2) Rotates among member headquarters in Riberalta, Guayaramerin, and Cobija Every 3 months ANMIN A 3 Provincia s: Bautista Saavedra, Franz Tamayo, Larecaja, encompassing municipalities of Charazani, Curva, and Achiri Mapiri 3 Social organizations, Ayllus, and mining sector in all 3 municipalities Mayor of each municipality (3) indigenous authority from each prov ince representative of mining sector Rotates among municipalities and La Paz (SERNAP) Every 2 months
86 Figure 4 1. Map of Bolivia.
87 Figure 4 2 Vicua, an endemic species found in the protected area of Apolobamba. Fig ure 4 3 Alpaca, the domesticated s pecies of vicua found in the protected area of Apolobamba.
88 CHAPTER 5 METHODOLOGY This chapter provides information on the methodology that I used to design the study, prepare the research in struments, collect and analyze the data, and interpret the results to answer my research questions. I employed a comparative case study and network analytic approach, which included the triangulation of data through a social network survey, a series of interviews, and the review of documentation The empirical data were collected during two periods of time from July 2011 to April 2012 and from November 2012 to January 2013. Research Design This is a case study design in which three different forums in Bolivia are analyzed (see Chapter 4). The case st udy can be defined as an empirical inquiry that investigates a phenomenon within its real life context (Yin 2003). Case study research design is most commonly applied where the phenomenon of interest is complex and highly contextualized with multiple varia bles unsuitable for control (ibid). As such, the use of mixed methods and multiple sources of data are used to capture this complexity, promoting rigor in case study research (ibid). In reference to work on collaborative processes, Ansell and Gash (2008), consider case studies to The first step of my study involved the collection of quantitative data using a social network survey designed to examine the relationships am ong forum members at the meso level. The forums themselves set up the boundaries of the whole network, and each of the forums is composed of approximately 15 to 20 members serving as representatives of various rural populations and social organizations. Th ese forums also coordinate with NGOs, municipal governments, departmental governments, and the national government. In addition to collecting network data to investigate network interactions, I collected qualitative data through follow up
89 interviews with s elected forum participants for the purpose of gaining additional insights regarding the network dynamics and network interactions with the macro and micro levels. I also interviewed external actors at the macro level to examine external legitimacy. Sampli ng Frame For two of the forums (CORIDUP, ANMIN A), I obtained a list of the forum members from the forum president and, for the third forum (BOCINAB), I worked with several key informants to create the list of forum members. I verified this list with sever al other key informants before using it. As I collected whole network data where the network includes all the forum members, this list was very important. Thus, it was imperative that I made sure no one was left out. All of the forum members were willing to participate in my study. About 50% of the forum members were present at each forum meeting. I made consistent efforts to contact and meet those participants who did not attend meetings. However, despite these efforts, there were a total of nine forum me mbers that I was unable to locate. These individuals no longer attend meetings and were either considered inactive members or were no longer considered members at all. Therefore, since the majority of active forum mem b ers participated in my study, my sampl e represents active forum members with a response rate of about 95%. 1 I completed data collection for the survey questionnaire by making several visits to each field site during my field work. Addressing Validity and Reliability Zeller and Carmines (1980) propose that the process of identifying theoretically meaningful relationships bridges the gap between theoretically deduced measures and empirical 1 The social network analysis, discussed below, provided information on the the inactive members.
90 evidence. From the beginning, I used the published research literature to identify the constructs, variable s, and indicators for my research. I incorporated both previously validated measures and some variables not yet explored to create a theoretical model. Developing a model rooted in my theoretical perspective guided my decisions regarding what data to colle ct and what techniques of data collection to use (Marsh 1982). Validity is the most important consideration in developing, evaluating, and measuring measure s what it 242). The traditional psycho social definition of validity includes content, face, construct, and criterion validity (Rubin and Babbie 1997). A more contemporary perspective views validity (or lack thereof) as the result of a process of validation (Adcock and Collier 2006). In this view, the validation process begins with a thorough exploration of the research literature to both develop a clear, theory based definition of constructs and explore the variables and indicators pre viously used by other researchers. Unlike validity, reliability is determined by how consistently an instrument measures the phenomenon under study. In the next section, I explain the steps taken to ensure reliable and valid research results from the opera tionalization of variables to the development of instruments and the data collection techniques employed during my fieldwork. Instrumentation I collected data at the individual level for all three forums (CORIDUP: n=18; BOCINAB: n= 12; ANMIN A: n=17). Alth ough I started with 48 respondents in total, I ended up with a total of 47 cases because I eliminated one case with missing data post hoc. A combination of qualitative methods, including in depth interviews, direct observa tion of meetings and documentary analysis in addition to quantitative methods was used to measure the social network variables and legitimacy.
91 I developed a series of instruments with the target population in mind and pre tested each instrument during my 2010 and 2011 field work. The inst ruments I applied at the meso level included (1) a key informant interview to familiarize myself with the context of each forum, (2) a perceptions of legitimacy, (4 ) a semi structured follow up interview with forum members to explore all variables, (5) and a semi structured interview with key informants at the macro level (Appendix B). I also developed a direct observation guide to use in the forum meetings that I at tended during my field work. My sample of social network participants consisted of 47 respondents for the social network questionnaire and legitimacy scale and, based on their accessi bility, I selected 30 of these respondents for the follow up interviews. M easurement of the social network variables and legitimacy produced interval data. There were a total of nine forum members who did not show up at any of the forum meetings and who m I was otherwise unable to locate in their communities during my fieldwork According to forum leaders, they no longer consider some of these individuals to be forum members because they infrequently or never participate in meetings. However, since their names were included on the list of forum members used as my sampling frame, I made every effort to locate these individuals. Direct and participant observation 2 was also used as a primary source of evidence, while archival records (e.g., organizational documents, government reports, newspaper articles) were used in a complementary manner (Yin 1994). Moreover, in addition to the follow up interviews with current forum members, I conducted a total of 28 interviews with former forum members 2 P articipant observation is favored by many evaluators because it provides the richest data on both process and context characteristics and permits in depth analysis of the relationships linking process variables to (Conley and Moote 2003, 381).
92 and supporting institutions (i e. NGOs) in order to familiarize myself with the forums. I als o conducted a total of 20 in depth interviews with representatives of national government institutions to explore the meso macro linkages. Inductive and in depth evaluation methods, particularly ethnographic approaches, have gained credibility over time be cause they allow for consideration of complex interactions between variables (Conley and Moote 2003). I use qualit ative data and models that are part of such ethnographic approaches in order to inform my interpretation of the quantitative models. Pre testi ng Instruments The first step in pre testing the interview guide was to conduct cognitive testing with informants during the exploratory phase of this research, which I conducted with the assistance of colleagues and practitioners in Bolivia during the sum mer of 2010. This step resulted in the elimination and alteration of interview questions, based on both the feedback provided by the test subjects and direct observation. I used the information gained through the cognitive testing to detect unclear stateme nts and to alter the language in order to make the questions more understandable. I also used the information gained from the questionnaires to identify relevant questions and topics to explore in the interviews and to eliminate some interview questions. Pre testing with informants from local groups similar to the sample population ensured appropriate use of language for all instruments. A critical aspect of instrument construction is the use of effective language, such that formulating items in a way that their meaning is clear and unambiguous. Pre testing instruments provided valuable feedback as to particular terms, technical language, or euphemisms familiar to the sample population but not to the population as a whole (Colton and Covert 2007).
93 Intervie ws with Key Informants For each forum, I conducted at least five key informant interviews with individuals who have worked closely with the forums, such as former forum leaders and supporting NGOs. Interviews lasted up to one hour depending on the availabi lity of informants and the detail provided by them The topics explored in the key informant interview included the history behind the creation of the forum, the purpose of the forum, important actors in its history and how the roles of these actors have c hanged, organizational changes over time, rules of decision making processes, the relationship with communities, articulation with the national level, strengths, weaknesses, and future directions. All interviews were audio recorded, and I took notes throug hout the interview. I was more interested in focusing my attention on the responses of the informants to generate further questions if necessary. Social Network Survey I developed a social network survey to collect whole network data for each of the thre e forums and the respective forum members in order to measure the network variables (Table 5 1). This survey was paper based and conducted orally. It took take place in a comfortable place for the participants involved. The survey was composed of three par ts designed to (1) obtain background information of the respondents (age, sex, level of education, group affiliation, leadership experience); (2) measure structural features of the networks (network position and brokerage); and (3) elicit personal network micro (community) levels. To address network structural features, I asked each respondent to rank, on a scale of 1 to 5 (1=never to 5=all the time), to indicate their frequency of interaction with the other forum members over forum issues outside of forum meetings.
94 of their level of interaction with the other forum members, I used five paper circles that increased in size (from small to large) to represent an i ncreasing level of connectivity between individuals. Additionally, to capture the macro level influences, I asked key informants of each forum to identify national level government institutions with which the forum has interacted in the last two years. I then asked each respondent to indicate on a scale of 1 to 5 (1=never to 5=all the time) the frequency of their interaction over the past two years with each of the institutions they identified. I also asked respondents to indicate their level of satisfacti responsiveness to forum concerns on a scale of 1 5 (1=not at all satisfied to 5=very satisfied). forum member using the following name gen provide some information about each of the 25 network alters, such as their perception as to whether the alter thinks thei r interests are represented by their person in the forum on a scale of 1 3 (1=not well represented to 3=very well represented). Throughout the data collection process, I entered all of the responses from the forum members into a matrix for analysis at the meso, macro, and micro levels. Although I originally hoped to explore local level knowledge of forums by interviewing individuals in randomly selected communities represented by forums, accessing communities proved to be very difficult. 3 Therefore, I depen ded on several questions as a proxy in the follow up interview with forum members. I first asked the forum members to indicate how effective the forum is at communicating with the communities it represents and then to elaborate on how and when they share i social organizations. 3 Due to the distance and extensive travel involved.
95 Network level measures At the meso level, I operationalize the network structure with the degree of network cohesion or the extent to which actors are interconnected via some kind of social tie (Prell 2011, 39; see also Wasserman and Faust 1994). Two measures density and centralization are typically used together to examine cohesion. Density refers to the proportion of people that are connected in the network. As density i ncreases (and the number of ties between network members grow), communication across the network is considered to become more efficient (Rowley 1997). However, density is known to be somewhat problematic, such that large networks tend to have lower density levels because the potential number of ties is so large (Prell 2011). 4 Therefore, I also employ degree centralization which is a measure of the extent to which one actor in a network is holding all the ties in that network. Degree centralization ranges from 0 (low) to 1 (high) If a network is dominated by a small number of nodes (or ac tors), it is highly centralized at 1. A score of 0 reflects a network where all actors have the same number of ties. A high level of centralization could indicate differen t levels of engagement in the network and/or the existence of power inequities. Low centralization, on the other hand, could suggest a more equal distribution of power and influence within the network ( ibid ). A network with high density and high centraliza tion would be less cohesive than one with the same density score but a lower centralization score (ibid). Some scholars find that an uneven distribution of ties leads to asymmetric relations of influence and power (Ernst son et al. 2008 ; Diani 2003b), mea ning that both the issues of legitimacy and accurate representation of peripheral actors also need to be incorporated into the analysis (Bodin and Crona 2009). In light of these considerations, these measures are used to 4 has a high density score may be the result of the majority of ties flowing through a
96 identify the marginalization of som e actors within the network, suggesting that SNA may be used to assess whether the most relevant actors are being invited to, and engaged in, participatory processes. Node level measures At the individual level, I examine network centrality or the positio n of individual actors within a network using measures such as degree centrality and betweenness centrality ( Brass and Burkhardt 1993 ). C entrality refers to an actor's position in the network relative to others and is considered a structural source of powe r that results from holding a central position within a network ( ibid ), exerting influence in decision making (Friedkin 1993), and possessing innovative qualities (Ibarra 1993). D egree centrality refers to how many others an actor is directly connected to and is measured by the number of ties an actor has with other actors in the network. such actors are considered to have access to many alternative sources of inf ormation, resources, and so forth (Rowley 1997). 5 Betweenness centrality refers to the degree to which an actor indirectly connects other actors in the network (Freeman 1979). Actors with high betweenness centrality are brokers or gatekeepers as they facil itate exchanges between others that would otherwise be disconnected sets of actors (Burt 1992) 6 H igh betweenness centrality also provides an actor the ability to influence the flow of resources between others as well as a diversity of resources provided b y the others (Burt 2004 ; Granovetter 1973 ). Moreover, betweenness centrality reveals differences in the potential of withholding or distorting the flow of information (Freeman 1979). 5 Individuals with high centrality may have a high level of control and power in the network. 6 In social networks, brokerage positions often emerge that link otherwise disconnected actors, which in effect means they mediate social rela s on et al. 2010 ; see also B urt 1992, 2002, 2005 ).
97 I consider the impact of network centrality in terms of who occupies the se positions and how they utilize their position with respect to the governance process (Bodin and Crona 2009). The structural network position is considered a key entry point for analysts to understand how individuals in any natural resource governance s etting shape the development of the system in which they operate ( Bodin and Crona 2009 ). Furthermore, given the association between the stratification of the social system into different network positions and unequal distribution of status and access to so cial power (Baron and Pfeffer 1994; see also Fiol et al. 2001; Thye 2000), the interpretation of power here is that these different social status and power conditions lead to different perceptions of the social system (Lamertz and Aquino 2004). In this re search I also explore the degree of cross level exchange (the proportion of ties connecting actors with different organizations), between forum members and national government institutions as well as between forum members and their constituents. Many schol ars point out the need for better understanding about such cross scale interactions in natural resource governance ( see Berkes 2009 Cash et al. 2006, Ostrom 2005). This analysis enables the crossin see Burt 1992, 2004; Granovetter 1973, 1983 ). While social network analysis can help untangle the pattern ing of ties that can gi ve rise to network level mechanisms that enable or restrain individual and collective action, it is relatively on cultural and meaning m 280). Many authors emphasize the value in using both quantitative and qualitati ve analyses to better understand network structure ( see Emirbayer and Goodwin 1994; Mische 2003). According to several scholars, any general framework for thinking about the process of participatory deliberation
98 using a multi criteria analysis will itself depend fundamen tally on context (Stirling 2006 ). For this reason, I complement social network methods with ethnographic methods to enhance the understanding of network structural features within a highly complex context. In this research, ethnographic dat a was collected at the micro and macro levels to further analyze the cross scalar relations of forum participants at both of these levels. Legitimacy Scale I developed a scale to measure the internal legitimacy of forums from the perspective of the indivi democratic quality of deliberation founded on models of democracy that favor input legitimacy (e.g. participatory or pluralist models) or procedural legitimacy, which can be label ed as rocess 26). This understanding of legitimacy holds that the process must be open and inclusive because only groups that feel they have had a legitimate opportunity to participate will develop a commitment to t he process ( Ansell and Gash 2008 ). In order to measure the conceptual dimensions, I borrow from theoretical concepts suggested in the literature, such as (1) stakeholder representation; (2) equal opportunity or inclusiveness; and (3) process transp arency ( Ansell and G ash 2008 ; see also Abelson et al 2003; Caron Flinterman et al. 2006 ; Rowe and Frewer 2000; Webler and Tuler 2000 ) (Figure 5 2). I pre tested the legitimacy scale prior to data collection and employed several analyses post hoc to ensure the reliability and validity of the scale. The individual items in the scale are statements, originally rated on a scale from strongly disagree to agree (from one to five). The legitimacy score for each I explain the steps involved in developing this scale in Chapter 6.
99 Follow up Interview I conducted in depth interviews to further examine the conceptual dimensions of procedural legitimacy and the social network features explored. Questions regarding t he quality of deliberation focused on the mechanism used in the decision making processes, with a particular emphasis on inclusion, and on the functioning of information diffusion both within and outside of the forum. To explore the social network features I used the visuals generated from the network data in NetDraw and asked selected respondents why they think the network is structured the way it is and to give their explanation for their position as it relates to others in the network. The follow up int erview generally raised questions about the implications of the social respondents to discuss their capacity or ability to contribute to deliberation based on their position within the network (Duran 1999). Moreover, the follow up interview allowed for the discussion of the explored relationships, which either helped support or negate the proposed model generated from the network data and legitimacy scores. The dura tion of the follow up interview varied, ranging from 20 to 40 minutes. In some cases, time was limited, particularly when I had to conduct the follow up interview immediately following the questionnaire due to the difficulty of locating the respondent at a future date I found it particularly challenging to conduct both the questionnaire and follow up interview during my first one on one interaction with a respondent, as there was insufficient time for trust building. These circumstances may have affected i nterview responses. Nonetheless, where I subsequently ran across or interacted with these individ uals, I made efforts to engage in further discussion.
100 Macro Level Interviews I conducted 23 in depth interviews with key contacts at several national governme nt institutions. I identified these individuals either at events where the particular forums were present or through the actual government offices. Interview questions pertained to the key rience working with the forums themselves, and coordination with individual forum members, among others. These interviews contributed to my understanding of forums and forum demands which facilitated my analys is of government responsiveness. The governed [peoples] by executing policies in a way that is consistent with [their] (Morlino 2010, 214). Although this is a complex concept to measure, for the purpose of my study, I interpret government responsiveness as its reception to forum demands. Participant Observation data on both p rocess and context characteristics and permits in depth analysis of the relationships linking process variables to o 391). Although it can be quite time consuming, participant observation is favored in inductive research and is well suited to theory building (ibid). Participant observation was particularly appropriate for looking at the participatory process, especially since the forums I studied provided an interesting context in which to pose certain questions about deliber ation in practice. Through my personal connections with forum members and the supporting institutions, I obtained permission to observe forum meetings. Both within and outside of forum meetings, I gathered information regarding the communication dynamics among actors, such as the types of interactions among members and the way the actors participated in discussions Taken together,
10 1 these questions provide a schema for analyzing how perceptions are constructed within and beyond a process. When observing m eetings, I used a recording device and depended on my notes; therefore, I was able to focus my attention on the substantive issues discussed and the process of interaction among participants as they deliberated over issues. Where possible, I noted verbatim the language used; otherwise I sought to captu re the sentiment expressed by verbal and non verbal communication with each other. In the event of a day long meeting, I also chatted informally with participants over lunch about their invol vement in the forum. Data Processing and Analysis I selected data analysis methods that correspond to the chosen data collection methods. In this study, questionnaire data were analyzed statistically, and data from interview, participant observation, and documentation were analyzed thematically. The data analysis was iterative in this study and influenced the refinement of ongoing data collection. In this way, the procedural step overlapped with the data collection step (Rosenberg and Yates 2007). Social Network Analysis I used UCINET version 7.20 ( Borgatti et al. 2002 ), a social network analysis program, to calculate the social network variable scores for the numerical values that the 47 respondents chose for the responses. I also calculated the m ean scor e of the 47 respondents on the legitimacy scale I then examined the correlations among the variables in this study and the significance of the proposed relationships among the social network variables and perceptions of legitimacy using linear regression analysis. I used MAXQDA (Kuckartz 2001), a computer assisted qualitative data analysis package, to analyze the qualitative data that is examined in relation to the study variables.
102 Linear Regression Analysis I use linear regression to statistically examin e the relationship between the independent variables or social network variables and legitimacy, the dependent variable. This analysis is a statistical technique that attempts to explore and model the relationshi p between two or more variables. The resulti ng model is expected to explain the variance in legitimacy scores as a function of the network variables. Qualitative Analysis I used several techniques to analyze the qualitative data from in depth interviews with current and former forum members, NGOs t hat work closely with the forums, and representatives from government institutions that have worked with the forums. During data collection, I brought all the data sources together using Excel and a MAXQDA database to prepare for analysis. 7 I used audio re cordings of the interviews and had the recordings transcribed at the local case study sites. I read through the interview transcripts and began coding to develop theoretically based themes and emerging themes for each forum during and after data collection Coded interview data were placed in a matrix to obtain more overview and insight. I then considered the data by established and emergent themes, allowing me to move away from individual respondents and towards a more comprehensive analysis of the issues covered by the dataset. Fieldw ork Essentials Gaining Support For each of the forums, I met with key contacts that work very closely with the forums. For example, before meeting forum members, I established good relationships with NGOs in 7 I used Excel to record data from interviews as they were conducted whereas I used MAXQDA to further code interviews once I received interview transcriptio ns from local transcribers.
103 the cases of CORID UP and BOCINAB and the protected areas institution in the case of ANMIN A. Coordinating with these individuals facilitated my work with the forum s, as they introduced me to the forum leaders and other people that I continue d to work with throughout my fiel dwork. In my meetings with these individuals, I made sure to communicate my expectations and objectives, such as the deliverables they should expect from my research findings. One of my most important personal priorities as a researcher is to give back hel pful insights from what I learned for the purpose of benefiting the future work of these forums. I also believe that the meetings I initiated with members prior forum meetings, as well as my presence in the various field sites also facilitated my work, as the forum members realized that I was serious about my commitment to this work. Research Team Identifying several field assistants who were able to commit several months to data collect presented one of the mos t challenging parts of my field work during dat a collection. I worked with quite a few field assistants during the data collection process who were all anthropology students at the public university la Universidad Mayor de San Andres (UMSA) in La Paz. The fact that these assistants were students may ha ve cause the difficulty because they were busy preparing to begin data collection for their own theses. In most Bolivian universities, many of the academic departments require a thesis and in the field of anthropology, students spend a significant period of time collecting data and writing their thesis. Despite their busy academic lives, three students were able to commit about two months to data collection. I paid each field assistant a fixed monthly salary, so as to avoid paying one more than any other. This monthly salary was based on a fair market price and was agreed to by my field assistants before they made the commitment. Any food and transportation costs incurred by the field assistants were refunded. I conducted a training session that lasted for several hours for the
104 field assistants involved in data collection. I also explained the proper way to obtain oral and informed consent on the research protocol (Appendix A). In the case of overlapping schedules of forum meetings, each field assistant was assigned to a different forum. Participant Accessibility One of the major challenges presented during data collection was contacting and scheduling a time for the questionnaire and interview with each forum member. Between idents of their organizations as well geographic location ( far from major towns or cities), they were difficult to locate outside of forum meetings. As a result, I generally conducted the questionnaires several days prior to or following forum mee hours to their respective communities. In addition to the geographic factor, all of these individuals maintain a busy schedule, as they are all the presidents or primar y leaders of their social organization. Apart from their leadership responsibilities (work without pay), many depended on different forms of livelihood. Informed Consent Before beginning the social network survey and interviews, I provided key informants and forum members with a copy of the informed consent (Appendix B) and explained the purpose of my study I also informed participants of the approximate time it would take to complete the surveys and interviews. In order to participate in the study, part icipa n ts were required to verbally state their agreement to participate. Conclusions The empirical data were collected during two periods of time from July 2011 to April 2012 and in a follow up visit from November 2012 to January 2013. The case study meth odology included triangulation of data through questionnaires (including the social network
105 survey and legitimacy scale), follow up interviews with forum members, key informant interviews, participant observation, and documentation. In depth interviews wi th key informants throughout the research process were critical to examining the relationship between the research variables and the contextual features of each forum. I made a concerted effort to ensure reliable and valid research results from the operati onalization of variables to the development of instruments and the data collection techniques employed in my field work.
106 Table 5 1. Operationalization of network variables Network Variables Definition Operationalized measures Measure significance Ne twork level variables Network cohesion The extent to which actors are interconnected in some kind of social tie (Bodin and Prell 2011) Density: the proportion of connected to non connected nodes in network (# of connected nodes/total # of nodes) Deg ree centralization: the extent to which an actor is holding all ties in network where all actors are directly connected to each other (Ranges from 0 to 1) As density increases, the # of ties between network members grow The more ties each member has wit h other actors the greater the density of the network If a network is dominated by few nodes, it is highly centralized at 1; A network where all actors have the same # of ties has a score of 0 Node level variables Network centrality The position of individual actors within a network relative to others (Brass and Burkhardt 1993) Degree centrality: the connectedness of an actor with others in the network (The # of direct ties to and from an actor referred to as Betweenness central ity: how many times an actor rests on a short path connecting two others who are themselves disconnected; Indicates an between other actors in the network ( The # of connections that link unconnected actors) Actors with high de gree centrality are considered to have more access to alternative sources of info, resources, etc. Actors with high betweenness can facilitate information flow or exchanges between less central actors
107 T able 5 2. Key literature used for evaluating procedural dimensions of internal legitimacy. Conceptual Dimension Definition Stakeholder representation process must be open and inclusive (Ansell and Gash 2008; see also Murd ock et al. 2005; Plummer and Fitzgibbon 2004; Power et al. 2000; Reilly 1998, 2001) Equal opportunity Transparency Equal opportunity for different stakeholders in contributing to decision making and equal access to information and other resour ces (see Caron Flinterman et al. 2006; Abelson et al 2003; see also Fiorino 1990; Laird, 1993; Rowe and Frewer 2000; Webler and Tuler 2000). Clear and consistently applied ground rules that assure stakeholders that the process is fair, equitable, and op en (Ansell and Gash 2008; see also Caron and Flinterman et al. 2006; Murdock et al. 2005), clear definition of roles (Ansell and Gash 2008; see also Alexander et al.1998), and the information used in the process that must be easily accessible and transpare nt (Abelson et al. 2007)
108 CHAPTER 6 DEVELOPMENT OF LEGITIMACY SCALE I developed a scale to measure the internal legitimacy of forums from the perspective of the individual forum members. In this chapter, I first present the steps taken to develop the scale and review the different pre and post hoc analyses conducted to ensure the validity of scale items as well as the reliability of the scale used in data collection. Although I set out to measure internal legitimacy as a multi dimensional construct, p ost hoc factor analysis indicated otherwise. Scale Development readily observable 8). A scale requires a number of items to assess the variety of characteristics associated with a particular variable of interest. According to Colton and Covert (2007), because scales produce a numerical value, they involve additional steps in the instrument construction process and considerably more pre t esting to ensure that the scores they produce are valid and reliable measures. As such, I performed several statistical operations to test the association between the items and the corresponding conceptual dimensions or sub components. I explain the steps taken to ensure the validity and reliability of the final legitimacy scale. Generating an Item Pool I developed a set of statements (items) to measure the different dimensions of the construct of internal legitimacy (DeVaus 1996). I spent several weeks br ainstorming items based on the literature, personal experience, and talking with colleagues. I focused on both face and construct validity in generating items (see Netemeyer et al. 2003). I began by drawing concept models including the three different conc eptual dimensions on a piece of 30x30 drawing paper.
109 This exercise helped to consolidate the different dimensions that I would include in the scale and begin brainstorming possibilities for the corresponding items. Ultimately, I came up with a list of appr oximately 70 statements to measure the three dimensions of legitimacy. I made several considerations during the brainstorming process that helped to ensure a valid measure. First, I considered the tone or wording of the statements and created positively a nd negatively worded (i.e. reverse coded) statements (see Colton and Covert 2007). For example, two items used in the pre he discussions are the item influences the quality of the response ( see Gendall et al. 1996). Researchers have found that negatively worded items either do not exh ibit as high reliability as positively worded items do or can be confusing to respondents (Netemeyer et al. 2003). These authors suggest weighing the potential advantages and disadvantages of using negatively worded items in the item pool. Based on the res ults from pre testing and the literature, I ultimately used only positively worded statements. Next, I avoided creating double barreled statements or items that contain two separate ideas. The problem with such statements is that agreement or disagreement with the item implies agreement or disagreement with both parts of it. In addition, I made sure the language used was appropriate to the sample population. As such, I checked statements for readability, unnecessarily technical language, and sensitive wordi ng. Individuals may not answer questions when they do not understand the content or are uncomfortable with the terminology. Finally, I considered ways of ordering the response set from strongly agree to strongly disagree or the reverse. Some researchers s uggest that the direction of the scalar response may
110 cause confusion and thus may invalidate the responses. I used a scalar response ranging from 1 (strongly disagree) to 5 (strongly agree). To avoid causing confusion, I clearly explained the scalar respo nse to research participants and u sed five paper circles that ranged in size from small to large to assist respondents in indicating the degree of agreement with the statements read. Pre Testing Pre testing addresses aspects of both reliability and validi ty and ensures that the indicators and items in an instrument are contextually meaningful. The validation process continues through the data collection and analysis phases of the research. I pre tested the legitimacy scale to ensure the validity, reliabili ty, and precision of the statements. I first tested the scale in Spanish for clarity and simplicity with 15 colleagues at the University of Florida whose native language is Spanish. They rated each of the 53 items based on its strength with respect to the different dimensions using a 5 point scale (1=very weak to 5=very strong). I then shared the scale with several contacts in Bolivia who are more familiar with the sample population. These individuals also suggested changes to the items to improve the relev ance of the statements or to reflect the regional language. I used these results and feedback to alter the language and eliminated 12 statements considered weak and unclear. After several weeks in Bolivia, I performed a reliability test on the 41 item sca le based on the responses of 201 respondents at la Universidad Mayor de San Andres, the public university in La Paz, Bolivia as well as at several workers unions in La Paz. Participants were asked to respond with their level of agreement using a 5 point sc ale (1=strongly disagree to 5=strongly agree) with respect to their membership and involvement in their student organizations organized by academic departments or union.
111 Reliability Test s alpha, a statistical test to see how well a set of items measures a single construct, the item total correlation measures the degree to which the items are consistent with each other (DeVellis 2003). As such, I measured the strength of the relationships between an item and the three sub components that I set up to measure. The stronger the statistical relationship (or correlation), the more confident one can be that the item is an actual measure of the construct (ibid). The idea is that if respondents are The reliability test on the pre test responses produced a Cronbach's alpha of 0.86. The components were as follows: Representat ion (0.86), Equal opportunity (0.78), and for Transparency (.84). Based on the item total correlations, I removed 18 items with an item total cutoff at .497 0.92. Many of the negative items were eliminated due to low total item correlations, all of which corresponded with equal opportunity. The remaining 23 items were included in the legitimacy scale used in data collection. Data Collection and Analysis I collected data for 47 respondents across the three di fferent forum s (CORIDUP: n=18; BOCINAB: n=12; ANMIN A: n=17). The legitimacy scale was included along with the social network survey in the first visit with participants. I conducted the surveys orally and they took, on average, 35 minutes to complete. It em Analysis Following data collection, I performed several statistical procedures discriminatory power analysis and confirmatory factor analysis using principal component analysis. After
112 deciding to collapse all of the theoretical dimensions, I ran a relia bility test on all items. I used a SPSS to complete the various statistical tests explained step by step below. Discriminatory Power Analysis Discriminatory power is an underlying property of reliability, validity, and responsiveness in the ability of sta true different levels of the outcome variable and to discriminate among people at a single poin (Marinac and Antonio 2010, 123; see also Janssen et al. 2007 ). In other words, this test is used to ens ure that items discriminate well among the respondents. I first calculated the total legitimacy score for each respondent based on the 23 remaining items. Respondents were then sorted based on their final scores from least to greatest. I divided the respo ndents into three groups or quartiles 25% with the lowest scores, 50% in the middle, and 25% with the highest scores. I then compared the quartiles to determine which items do not discriminate amo ng respondents. A total of eight items were eliminated that were not discriminating well among respondents. A total of 15 items remained. Factor Analysis I conducted principal component analysis (PCA) to see how the remaining 15 items components. The items are used to identify t he conceptual meaning of the factors. Considered a type of confirmatory component analysis, PCA has become a useful technique to test the internal consistency and validity of a measure at the later stages in scale development (Netemeyer et al. 2003 ). I fi rst used the varimax rotation method to extract the factors with an eigenvalue of at least 1.0. The varimax rotation is a method of orthogonal rotation, which means that it results in uncorrelated factors. Although the component analysis presented four com ponents with eigenvalues greater than 1.0 (Table 6 2) the component p lot in rotated space (Figure 6 1 ) did not
113 support the multi dimensionality of the construct. Based on the results of these tests and the understanding that the number of factors extracte d is equal to the number of observed dimensions being a nalyzed in PCA (Hatcher 1994), I decided to exclude the weak fourth component. Combined, all three components accounted for 67% of the total variance. In interpreting the rota ted component pattern (T able 6 2 ), an item was loaded on a given component if the loading was .40 or greater for that component and was less than .40 for the others. T he items that loaded on the components, however, did not align well with the intended component and therefore I t urned to the literature on scale development to determine the next step. According to Hatcher (1994), interpreting the results from the PCA is a subjective and thus, a challenging task. This author offers interpretability criterion to assist in solving th e ber of Hatcher 1994, 21). Hatcher suggests that by satisfying one (or more) of the criteria, one maximizes chances of retaining the correct number of components. Even though there were at least three items with significant loadin gs on each retained component, not all of the items that loaded onto the components aligned with what I had expecte d. In addition, several items did not have relatively high component loadings for some variables and near zero zero loadings for the others. In addition, I found that not all items that loaded onto the components shared the same conceptual meaning. After performing these post hoc analyses, I decided to use all items to measure the construct of internal legitimacy. Reliability Test I ran a reli alpha and the item total correlation (Table 6 4). Reliability analysis was carried out to ensure the factors were reliable (Bearden et al. 1993). The results are based on 47 ca ses. alpha for all items was .857 indicating reliability I used these 15 items to obtain the mean legitimacy score for each respondent.
114 Conclusions I developed a scale to measure the dependent variable, internal legitimacy, as a multi dime nsional construct. I took several steps to develop an extensive list of items pertaining to the three dimensions of stakeholder representation, equal opportunity, and transparency used to assess this construct. I pre tested the scale in Bolivia and elimina ted items with low item total correlations. After collecting data for 47 respondents across the three forums, I conducted post hoc analyses and a total of 15 items remained from the 23 used in data collection. I used principal component analysis to determi ne which of the remaining items loaded onto the intended three components. As a result of the analysis, in which not all of the items that loaded onto the components aligned with what I had expected, I decided to use all 15 items to measure the construct o f internal legitimacy instead of the intended multi dimensional measure.
115 Table 6 1. Item total correlations f or the 15 items included in the final legitimacy score. Items Corrected Item Total Correlation This forum is accessible to a wide range of groups. .429 This forum includes all those who should be represented. .297 I feel that my participation in meetings is valued by others. .533 If feel comfortable expressing my points of view in meetings. .672 I am given sufficient time and spa ce to present my opinions. .485 I understand how decisions are made in meetings. .604 There is sufficient time to discuss all relevant topics in meetings.. .415 If I have a question, I feel comfortable asking it. .706 All participants have the same opp ortunity to influence decisions. .527 The information that I receive helps me make decisions. .544 All participants show mutual respect during meetings. .299 If I do not agree with a decision, I feel comfortable expressing my opinion. .450 The debates allow sharing different opinions. .653 My role in the organization is clear to me. .475 I receive the necessary information to effectively contribute to decision making. .659 =.857, N=47
116 Table 6 2. Eigenvalues for items used to determin e number of salient scale components. Component Eigenvalues 1 5.637 37.578 37.578 2 1.782 11.877 49.455 3 1.567 10.446 59.901 4 1.123 7.487 67.388 5 .998 6.650 74.038 6 .765 5.097 79.135 7 .640 4.264 83.399 8 .537 3.583 86.982 9 .481 3.207 90.189 10 .363 2.419 92.607 11 .294 1.961 94.569 12 .246 1.642 96.210 13 .226 1.509 97.719 14 .190 1.265 98.984 15 .152 1.016 100.000
117 Figure 6 1 Scale items plotted on to rotated space used to determine distinct components.
118 CHAPTER 7 SOCIAL NETWORK ANALYSIS Recall from Chapter 4, I have meso, macro and micro data. I use whole network analysis to operationalize the interactions at the meso level, personal network analysis to measure the influences from the micro level, and a two mode network techniqu e and ethnographic data to capture scalar influences to the macro level. This analysis focuses on the meso level and includes an examination of (1) the whole network structure of the forums (network cohesion); (2) the structural network position of each pa rticipant in the forum (degree centrality and betweenness centrality); and (3) the cross boundary exchange between forum participants at the meso level and actors at the macro level. In SNA there are network level and node level (at the individual level) m easures (Table 7 1). While I first look at density and network centralization to measure network cohesion at the network level, the remaining analyses focuses on node level measures to explore different network structural features. Structural data and attr ibute data are analyzed systematically to generate information about the internal dynamics and processes of each forum. I collected data on the frequency of interactions between forum members (meso level) to measure network structural features, such as net work centrality and betweenness centrality. I used UCINET t o analyze my social network data. I first present the visualizations of the data from the three forums at the meso, macro, and micro levels and then discuss the significance of these visualizations Before I began analysis, I recoded the original data set, which was based on a scale from 1 to 5 (1=never to 5=all the time), to a scale from 0 4 (0=never to 4=all the time). I recoded the data, in part, because UCINET interprets a response of 1 as pert aining to some kind of relationship or tie, whereas 1 in the scale used for data collection refers to the absence of a relationship. Although there were nine non respondents in the meso data sets, I have data from
119 other respondents about them, and thus the lack of responses provides no reason to eliminate this Network Level Analysis At the network level of analysis, I used the raw data sets for each forum at the meso level network cohesion. While there is some variation, all of the forums appear to have relatively dense networks CORIDUP has a density score of .642, BOCINAB has a de nsity score of .588, and ANMIN A has a density score of .502. CORIDUP appears to have the densest network, indicating that there are fewer individuals in the network who are isolated than in the others. What explains the differences in density scores acros s groups? There may be several explanations for why CORIDUP, compared to the other two forums, is the densest network. A characteristic of CORIDUP that distinguishes it from the other forums is it has use of an office space where its principal leaders can be found on a daily basis. While CORIDUP does not own this office, CEPA provides a dependable place for members to nd with other members. In addition, all forum meetings are located in CEPA and the members that must travel to Oruro from their communities are reimbursed by CEPA for travel costs incurred. The majority of CORIDUP members live within a two hour distance fr om Oruro and therefore reimbursement for the cost of travel, along with a free lunch, justifies the trip for many of the members. BOCINAB, on the other hand, does not have its own office space and meetings are held on a rotating basis in the offices of th e member organizations in Riberalta, Guayaramern, and Cobija. The NGOs that support BOCINAB, however, do provide a space for the leaders and members to use for informal meetings apart from the forum meetings.
120 Since BOCINAB tends to hold informal meetings in or around Riberalta, where 1 and the majority of the NGOs are located, those who most frequently interact outside of meetings were campesinos living in and around Riberalta. The relatively far distance between Riberalta and Cobija makes geographic location an important factor affecting the frequency of interactions. 2 Similar to the geographic circumstances relevan A s members must come from up to 12 hours away to at tend meetings when held in the city of La Paz. 3 However, since La Paz is the capital of the department where the protected area is located, many of the members frequently travel to La Paz as part of their work as indigenous authorities and local government officials. Since the measure of density is known to be somewhat riddled with problems, I also measure degree centralization, or the extent to which one actor in a network is holding all the ties in that network. BOCINAB is the most centralized network ( 48 .72%). ANMIN A also has a high level of centralization (40.36%), but not quite as high as BOCINAB. In contrast, CORIDUP has a much less centralized network ( 28.95%). What are possible explanations for why these networks, particularly BOCINAB and ANMIN A, appear highly centralized? There are few centralized nodes in BOCINAB (Figure 7 3), most of which are presidents of the organizations located in Riberalta who must often communicate with each other to coordinate forum meetings and activities. The members located in Cobija are the least central actors, explained by their infrequent coordination with those who are located in Riberalta. As mentioned, informal meetings are accessible to those living in Riberalta, which is relatively close 1 Five member organizations are located in Riberalta, one in Guayaramern and three in Cobija. 2 Travel time between Riberalta and Cobija is 12 hours on the bus and travel time from Riberalta to Guayaramern is one and a half h ours. 3 headquarters), meetings are most often held in La Paz.
121 to the location of BO the geographic location of the organizations in Riberalta may explain the highly centralized network, as it makes interaction among members in Riberalta more feasible. Similarly, th e president of ANMIN A worked in La Paz during his presidency, allowing him to coordinate frequently with the director of the area and several others who worked together closely to make decisions outside of forum meetings. In contrast, CORIDUP, the least c entralized network, uses a specified office space in CEPA and forum members know where to find the president and other leaders that work in CEPA on a daily basis. Considering the geographic proximity of CEPA located in Oruro to the communities, the accessi bility of leaders and information in CEPA is relatively greater than in BOCINAB and ANMIN A. There are several issues of concern presented by highly centralized networks. Networks cess to multiple sources of information and does not allow for the inclusion of a diversity of knowledge in the decision ma king process (Crona et al. 2011, 60). For example, the most highly centralized actors are those that live and operate in Riberalta an d not in Cobija. Those who participate most in forum meetings and guide the actions implemented by BOCINAB are the president of BOCINAB as well as several other presidents of the member organizations in Riberalta. According to Prell (2011), a network with high density and high centralization is less cohesive than one with the same density score, but a lower centralization score. B ased on this notion of cohesion, BOCINAB and ANMIN A are not as cohesive as CORIDUP. The high density and low centralization cha racterizing CORIDUP is expected to promote trust and increase the possibility for social control (Granovetter 1985; Pretty and Ward 2001) as well as encourage the spread of information through increased accessibility to information (Abrahamson
122 and Rosenkop f 1997). The accessibility of CORIDUP leaders and accumulation/collection of among members Node L evel Analysis Thus far, I have compared the cohesiveness of the forums at the network level; however, another way to look at network structural features is at the node level. In this section, I examine the node level measurements of centrality which are measurements based on the number of people to which an indivi dual is connected. There are four measures of centrality used widely in network analysis: degree centrality, betweenness, closeness, and eigenvector centrality. To identify the individuals most likely to be important in terms of influence and control withi n the network, I use a combination of degree centrality and betweenness centrality. In UCINET, I these network structural positions. Meso level: Degree and Betwe enness Centrality Degree centrality refers to the number of immediate contacts an actor has in a network. I examine the direction of the ties in each forum, focusing on how many incoming ties a focal actor receives (in degree) and how many incoming ties a focal actor gives (out degree). An actor with high levels of in actor with high out degree centrality can be considered more as an indicator of influence or power within the network. In this section, I assess the significance of the actors with the
123 maximum values of in degree and out degree with respect to the graph centralization of the whole network presented in Tables 7 1, 7 2, and 7 3. In addition to examining the in degree and o ut degree values for each forum, I performed the Freeman Centrality function for node position and produced visuals in NetDraw that show the degree centrality of the forum members to see who plays a more central role (and less central role) in the network based on their frequency of interaction with the other actors. As indicated by the size of the nodes, Figures 7 1, 7 3, and 7 5 display the degree centrality of members within the network. The larger nodes indicate the members who are most connected to the other members based on their frequency of interaction with others, whereas the smaller nodes represent the members who are less connected to other members. I symmetrized the data on the maximum value reported by the two respondents. Symmetry, also known a s reciprocity, refers to the extent to which a relationship is bi directional or, in other words, the extent to which a particular actor's evaluation of his or her relationship with another actor is reciprocated by the other actor and thus mutually acknow ledged. The visuals from NetDraw display the strongest ties, which represent the most frequent interactions. Centrality can also be considered as the degree to which an individual actor connects other actors who would otherwise not be linked to that actor (Burt 1992). In SNA, the degree to which an actor indirectly connects other actors is often quantified using the betweenness centrality metric. An actor situated between many other actors in the network is said to have a high betweenness centrality. This s uggests that the actor is in a position to potentially act as a bridge between these other actors that would otherwise be disconnected sets of actors. High betweenness centrality provides an actor with the ability to influence the flow of resources
124 between other actors and also provides this actor with a diversity of resources from the other actors ( Granovetter 1973 ; Burt 2004 ). I specifically utilize the betweenness centrality measure because it indicates brokerage, and I am interested in determining the m embers best positioned to be brokers of interactions within the forums. To make this determination, I used the Freeman Betweenness function in UCINET to examine node brokerage for each forum member. I obtained the betweenness score for each forum member an d produced the respective visuals for each forum. Figures 7 2, 7 4 and 7 6 display node betweenness, as indicated by the size of the nodes. The larger nodes indicate the members who play a greater brokering role in the network, whereas the smaller nodes re present the members that play less of a brokering role in the network. The social network results demonstrate that the individuals who are the most central actors are either the formal leaders of the forums or active forum members that interact frequently with the other members. Results also indicate that many individuals overestimate their frequency of interaction with the other members, which in some cases inaccurately depicts their positions within the network. One potential reason for such overestimati on may be that all the members who overestimated their frequency of interaction are important leaders within their respective organizations and perceive of themselves as interacting frequently with everyone. Forum 1: CORIDUP In reference to Table 7 1, Pac o and Richard have the greatest out degree, while their in degree shows that they exceedingly overestimate their ties with other forum members. Julio, Martin, and Miguel have the greatest in degrees, meaning that the other members share information with t secretary, vice president) who have the most control over the information flow in CORIDUP.
125 S ome of the most central nodes and leaders (Julio, Martin, and Miguel) have lower out d egree than in degree, meaning that they underestimate their interactions with the other members. Members in CORIDUP with the greatest number of ties within the network are Paco, Fernando, Julio, and Martin (indicated as the largest nodes in Figure 7 1). P centrality in the network partially reflects their overestimation of the connections they have with the other members. On the other hand, Julio and Martin actually do occupy more central network positions. Julio, the current president of CORIDUP, and Martin, the current secretary of office. Julio, the founder of CORIDUP, was elected president upon its creation in 2007 and was re elected for anothe r two years via referendum in 2009. Prior to the creation of CORIDUP, Julio was instrumental in organizing the affected communities in order to facilitate discussion of the problems that these communities faced as a result of the practices of the mining co mpanies, years. heard members refer to CORIDUP and Julio in the same breath. E ven those outside of the organization, such as institutions in Oruro, associated CORIDUP with Julio. However, it is important to note that Julio has a different financial situation than most people, which allows him to sacrifice his time without pay 4 He i s from the community of Santo Tomas and comes from a cattle raising family. Julio owns about 2,000 cattle, which gives him substantial financial security and permits him to work without pay from the city of Oruro. 5 Not only has Julio fulfilled 4 Similar to the other forums, leaders of CORIDUP receive no compensation for their work 5 The poor quality of the water available to his cattle augments his determination to pressure mining companies to change their practices.
126 an important leadership position in CORIDUP, he is also responsible for establishing and important entities. also from Santo Tomas. He was an acquaintance of Julio prior to his work with CORIDUP and was one of the first people to join Julio in his quest to develop an organization to defend the affected Martin is an employee of CEPA and works in its Environmental Justice Office. Julio and Martin have backgrounds very different from the other members of CORIDUP. member of CORI DUP), prepare letters, make visits to government institutions, and organize events in order to pressure the national government and/or the mining companies into changing the practices of the mining companies. 6 This daily presence of these two leaders at CE PA presence at CEPA has made it a central meeting place for the affected communities, allowing the leaders to be accessible to these populations. support has encouraged the persistence of in influencing the relevant government institutions and mining companies. Paco, on the other hand, does not occupy one of the major leadership positions in CORIDUP (president or secretary), and cannot be found in Oruro on a daily basis. Paco is retired from his work with a local mining company located close to his town of Poop, which is about a 45 minute drive from Oruro. Paco is an involved member of CORIDUP he attends meetings 6 d egree in law, an uncommon achievement for Bolivians from rural communities.
127 and contributes centralized members in CORIDUP is mostly attributable to the fact that he overestimates his ties with the other forum members. Members in CORIDUP that exhibit the capacity to act as brokers in the network are Paco, Julio, Martin, and Fernando (indicated as the largest nodes in Figure 7 2). As discussed above, the first of these three actors are also the most central actors in the network. According to this visual, Paco is displayed as the most important broker in the network. Paco serves as an important connector between the municipality of Poop and Oruro; however, this importance may partially reflect his overestimation of his ties with the other forum members. While Julio and Mar tin spend the majority of their time in Oruro, Paco more frequently contacts other members that live closer to the sub cuenca of Poop. He also frequently communicates with several members who have not attended a CORIDUP meeting in more than a year, such a s Rosela and Santos. In addition their year long hiatus from meetings, Rosela and Santos rarely visit CEPA or interact with Julio or Martin. While Julio and Martin play an important role in coordinating CORIDUP meetings and guiding its daily conduct, they are not the brokers of the network. Paco, on the other hand, can more properly be considered the most strategic member in strengthening the organizational nexus with some communities. Fernando, also from the town of Poop, began working with a local mini ng company in no longer participates in the forum. In an interview with him, Fernando spoke about his distrust the mining company, Fernando has not attended any meetings and has not interacted with the
128 members actively involved in CORIDUP. He claims, however, that he has not received notification of any meetings in the last year. Fernando plays an important role in connecting those members who do not frequently attend meetings, such as a fellow miner and neighbor, Forino. Forum 2: BOCINAB In reference to Table 7 2, Lucio has the highest out degree, while Rosemary, Felipe, and Sandra have the highest in degrees. Assessing these values as well as the difference in the out degree graph centralization of 64.05% and the in degree graph centralization of 17.42%, it is Members of BOCINAB, to a greater extent than CORIDUP, overestimated their ties with other forum members. As a consequence, in many c ases, fewer ties are reciprocated by other actors as compared to the amount of ties estimated by the actors themselves. Such overestimation challenges the accuracy of the results regarding the network positions of forum members. Figure 7 3 shows that memb ers in BOCINAB with the greatest number of ties in the network are Lucio, Calixto, Pepe, and Felipe. Partially due to their overestimation of ties, Lucio and Calixto are portrayed in the network as more central than they actually are in BOCINAB. Lucio is t he president of one of the two indigenous organizations in BOCINAB that are located in Riberalta, and he has held this position for the last eight years. As such, his membership in BOCINAB dates back to its creation in 2003. Table 7 al position in the network may reflect his overestimation of his interactions with other members. This is demonstrated by his out degree of 52 and in degree of 19. Based on information from key informants and personal observation, Lucio does not actually a ttend BOCINAB meetings, but instead relies on two individuals Calixto and Pepe from his organization to act as representatives. When the forum discusses issues affecting the indigenous sector during its
129 meetings, these two individuals are more vocal than other members and, as a result, the other members of BOCINAB look to them for the indigenous perspective on issues. Felipe, on the other hand, appears less central than Lucio, Calixto, and Pepe, even though During my fieldwork, I found that Felipe was one of the most difficult individuals to locate, especially because he moved around frequently and was not often accessible via cellular phone. As discussed in Chapter 4, the First Coordinator (or president) of BOCINAB rotates by member organization every two years. In organization, automatically became the president of BOCINAB, despite his unfamiliarity with BOCINAB This initial unfamiliarity and his other obligation may explain why, over time, he has demonstrated less than complete commitment to BOCINAB. As displayed in Figure 7 interactions within the ne twork. It is likely that frequent communication with the rest of the potential as a broker would be greater if he participated more in the meetings and in other BOCINAB events. Instead, Lucio relies on the representation of Calixto and Pepe, which explains why Calixto and Pepe are much more informed than Alberto when it comes to BOCINAB matters. Considering that BOCINAB is dominated by campesino organizations, Cal ixto and Pepe serve as important connectors between the campesino and indigenous organizations. However, while they occupy an important role in the network, Calixto and Pepe do not represent the entire indigenous sector, especially considering that they ar e not associated with the other indigenous organization, located in Cobija, in the same way that they are
130 Forum 3: ANMIN A In reference to Table 7 3, Andres and Tomas have the highest out degree, while Roger, Alfredo Fernando, and Andres have the greatest in degree in the network. As committee president, Andres actually does occupy a highly central position in this network; however, Tomas greatly overestimates his ties with other actors. As for those with the greates t in degree, Fernando is the director of the protected area, Alfredo is the mayor of a district within the area, and Roger is both a long time member of ANMIN A and a former mayor of a district in the area. Again, several actors overestimated their popular ity in the network (Andres, Tomas, and Alfredo). Figure 7 5 displays the very dense network of ANMIN A, where there is little variation in the centrality of the nodes. The members with the highest number of ties in the network are Andres, Fernando, and Ro ger. Andres, the president of ANM IN A from 2008 2012 (two terms), also participated as a member of the forum for a number years prior to his leadership position. In his first term as president, Andres assumed a very strong leadership role, which potentiall y explains why he was re elected to serve on the board of officers in the next election. Fernando director of the protected area since late 2010. With a universi ty degree in Administration and Management, Fernando worked as a professional in the field of natural resource management over the last several years. He is an important actor in this network, as an integral part of his job is to coordinate with the differ ent provinces inside of the protected area. Along with Andres, Fernando presides over the meetings of ANMIN A. Roger also has one of the greatest numbers been a member of the management committee since 2003.
131 Figure 7 6 shows that Tomas acts as a broker in the network; however, this figure does not accurately represent the interactions among forum members. Table 7 3 suggests that Tomas overestimated his interactio ns with the other members, as he has an out degree of 63 and an in degree of 28. This apparent discrepancy explains why he is represented as a broker within the network despite the fact that he does not actually hold such an important position as compared to other members. connection between the organizations and SERNAP because he worked closely with SERNAP during his presidency. Macro level: Degree Centrality I collected data on the frequency of interaction of each forum participant and generated a list of up to 15 institutions and organizations with which the forum interacts. Apart from national government institutions, this list includes NGOs and other supporting institutions that serve the role of interlocutors between the forums and the national level (Appendix C) The list of institutions was used to ask all forum members about their frequency of interaction with these institutions. The resulting data set produces a two mod e matrix, actor to organization, with which members interact. The original scale used in data collection was from 1 to 5 and was recoded from 0 to 4. I eliminated the non respondents for which I do not have data. I then transformed this two mode matrix (ac tor to organization) to a one mode matrix (actor to actor), as I am analyzing the latter. In the one mode matrix, the connections between the actors represent the number of ties they have in common with respect to their level of interaction with each insti tution or organization. These matrices were then added as attributes for individual nodes and used to create network structures based on affiliation. To transform the data from a one mode matrix to a two mode to 1 mod
132 minimum value (for valued data) since the data set is not symmetric. I also dichotomized the data, which converts valued data into binary data. Since I measured the strength of ties among actors from 0 = no tie to 4 =stro ng tie, dichotomizing this data transforms it into the absence or presence of a tie (zero or one). Although converting an ordinal or interval measure of relation strength into simple presence or absence may eliminate a substantial amount of information, m any social network analysis tools were developed for use with binary data only, and may even give misleading results when applied to valued data (Hanneman and Riddle 2005). Moreover, many of the tools in UCINET that are designed for binary data will arbit rarily dichotomize interval or ordinal data in ways that may not be appropriate. Therefore, I dichotomized the macro data based on a cut off value determined by examining the distribution of responses. I chose to dichotomize the data at a cut off value of 15 based on the distribution of data (see Figures 7 7, 7 8, 7 9). Thus, responses Once I dichotomized the data, I produced the visualizations in NetDraw that display the number of ti es forum members have in common with other members as it relates to their level of interaction with the macro level institutions and organizations (Figure 7 10). The lines represent the number of ties forum members have in common. This figure illustrates t hat many members have a high level of interaction with these institutions and organizations. Ethnographic data, on collaborate most with NGOs ( particularly CEPA) and make continuo us efforts to contact and coordinate with government institutions unlike the other members, who are not at all involved in these actions. Figure 7 11 also displays the frequent zations. The
133 proximity of most forum members (in the case of CORIDUP and BOCINAB) to the supporting institutions and other member organizations facilitate s frequent interaction QAP Correlation I compared the meso matrix with the macro one mode matrix to determine whether the interaction among members within forums is explained by their affiliation with macro institutions. For this analysis, I also include NGOs and other supporting institutions as macro institutions since they represent interlocutors between the forums and the national level. I utilized a quadratic assignment procedure (QAP) in UCINET, which develops standard errors and tests associations between two network structures that have the same actors ( Borgatti et al. 2002 ) The QAP correlation calcula tes measures of nominal, ordinal, and interval association between the relations in the two matrices. The result produces a Pearson coefficient, which is a number between 1 and +1 that indicates the degree of association between two variables. A higher co efficient indicates that the matrices are strongly associated. A p value of less than 0.05 indicates that the result is significant, meaning that it is unlikely that the observation is due to chance. The results from the QAP correlation show t hat the network structure at the meso level is correlated and associated with the netwo rk structure at the ma cro level with a slight variation across cases The results indicate that for BOCINAB there is a strong correlation and association between its meso network structure, based on who interacts with whom within the forum, and the network structure based on co affiliation with macro institutions (r =0.491 and p=0.011). This affiliation wi th macro institutions. The results for CORIDUP show that there is less of a correlation compared to BOCINAB but a significant association of network structures (r =0.300 and
134 p=0.035). In the case of ANMIN A, the results indicate that there is a significant ly weak correlation and association between network structures (r=0.261 and p=0.028). Despite the similarities in the degree of association and the significance between the meso and macro structures across forums, the correlation between the network struc tures at the meso and macro levels of BOCINAB is the strongest (Figures 7.12 and 7.13). One explanation for this association is that BOCINAB works very closely with several NGOs that are located in one of the major cities, Riberalta, which serves as one of the geographic bases of the forum. Thus, the most central actors of BOCINAB, who are located in Riberalta, coordinate often with these NGOs. As the NGOs have specific projects that aim to facilitate the articulation of forum demands at the national level, these central actors may easily participate in this work in Riberalta. The most central actors in CORIDUP work very closely with just one local NGO that greatly contributes to the work of the forum. These actors coordinate meetings and activities with sev eral organizations and institutions with which they formally collaborate, such as the network of neighborhood associations representing urban areas and the local university in Oruro. The most central actors are also those who coordinate most often with gov ernment institutions, such as the Vice Ministry of the Environment and the Ministry of Mining, located in the capital city of La Paz, in efforts to influence environmental regulations. The other central actors in CORIDUP are not as involved as the principa l leaders in coordinating with these government institutions. Several of the less central actors claim that only the president and a few others have the opportunity to trave l to La Paz, and therefore the leaders are the only forum members who can kee p abreast of the issues on a continuous basis. Nevertheless, several of the less central members do interact frequently with the forum leaders and members.
135 The actors with the most central positions in ANMIN director, who works for SERNAP. The weak correlation between the meso and macro network structures may be explained by the minimal interaction between forum members and the macro these institutions. Conclusions The network level analysis indicates that, although the forums are all relatively dense networks, greater variation is present in their network centralization. SNA demonstrates that BOCINAB and ANMIN A have fewer central nodes than CORIDUP, meaning that they have fewer individuals controlling information flow within the forum. In this way, CORIDUP represents the most cohesive network. Differences in network cohesiveness across forums may be explained by the geographic dist ribution of members, the accessibility of leaders, and the accessibility of information to forum members. Node level analyses indicate that the actors with the highest degree centrality and betweenness centrality are the forum leaders who are the most impo rtant actors in sharing information with the other forum members. These leaders are also the most connected at the macro level, meaning that they fulfill a brokering role between the forum and the macro institutions. The QAP Correlation shows that the netw ork structure at the meso level is correlated and associated with the network structure at the macro level, with a slight variation across cases. The results indicate a strong correlation and association between work structure based on co affiliation with macro institutions, which is attributed to its work with several supporting institutions (i.e., NGOs).
136 Table 7.1 Out degree and In degree for CORIDUP members Node Out d egree In d egree Paco 58 35 Richard 54 1 3 Jorge 48 20 Julio 47 54 Martin 47 53 Augustina 42 42 Miguel 41 51 Jose 41 44 Marliz 39 41 Roberto 38 39 Adriana 35 12 Juan 32 40 Fernando 28 37 Tilo 23 33 Leonis 18 41 Forino 14 15 Minton 8 15 Nina 7 13 Rosela 0 7 Santos 0 13 Figur e 7 1 Degree centrality of CORIDUP members.
137 Figure 7 2 Node b etween n ess of CORIDUP members Table 7 2. Out degree and In degree for BOCINAB members Node Out degree In degree Lucio 52 19 Felipe 41 27 Sandra 38 27 Calixto 31 18 Cesar 31 22 Pepe 30 27 Marco 28 23 Rosemary 18 29 Maria 10 9 Luis 9 17 Alberto 7 18 Lorena 1 18 Rosa 0 22 Vincente 0 21
138 Figure 7 3 Degree centrality of BOCINAB members. Fi gure 7 4 Node between n ess of BOCINAB members.
139 Table 7 3. Out degree and In de gree for ANMIN A members Node Out degree In degree Andres 74 47 Tomas 63 28 Felix 54 31 Nelson 54 32 Fernando 47 47 Dante 45 23 Guillermo 45 33 Roger 44 50 Eddy 42 37 Irene 37 37 Tito 37 31 Alfredo 35 47 Santino 34 22 Pedro 23 37 Oscar 17 2 7 Hubert 12 21 Liliana 7 40 Orton 0 17 Marcelo 0 13 Celino 0 31 Augusto 0 15 Gregoria 0 5 Figure 7 5 Degree centrality of A NMIN A members
140 Figure 7 6 Node b etwee n ness of A NMIN A members Figure 7 7 Histogram of CORIDUP macro data us ed to determine cut off point for dichotomizing
141 Figure 7 8 Histogram of BOCINAB macro data used to determine c ut off point for dichotomizing Figure 7 9 Histogram of ANMIN A macro data used to determine cut off point for dichotomizing
142 Figure 7 10 Macro degree centrality for CORIDUP members Figure 7 11 Macro degree centrality for BOCINAB members
143 Figure 7.12 Degree centrality of BOCINAB members. Figure 7 13 Macro degree centrality of BOCINAB members.
144 CHAPTER 8 INTERNAL LEGITIMACY AS A FUNCTION OF NETWORK VARIABLES In this chapter, I use quantitative and qualitative analyses to examine the following question: W hypothesized that a highly cohesive for um will be associated with a higher average legitimacy Linear regression analysis reveals that there is a significant difference in legitimacy across cases and that meso degree centrality explains a fair amount of variance in the relationship between the independent and dependent variables explored in this research, producing a final model with an r squared of .317 and a p value of .000. Legitimacy scores were high, with low variability in the responses among individuals in each group. E thnographic data however, atic quality of deliberation. My ethnographic findings shed light on the importance of the leadership roles assumed by the most central network actors, but they also suggest that this relationship is explained by additional factors not included in the mode l, such as the relationship of forum leaders with supporting institutions (i.e., NGOs). Before performing the statistical analyses, I calculated the meso and macro social network variables using a variety of symmetrization and dichotomization methods (disc ussed in Chapter 7) in UCINET. In addition to dichotomizing the macro data, I also symmetrized the data at the meso level based on the minimum and maximum response values of ties reported by two respondents of the frequency of interaction. Symmetrizing on the minimum captures the minimum of the two values reported, while symmetrizing on the maximum takes the maximum value reported by two respondents. I also chose to dichotomize the meso data with a cut off
145 with a cut 1 The scores for the meso and macro variables that vary by symmetrization and dichotomizati on methods served as potential predictors w hen fitting the regression model. Transformation of Legitimacy Scores I transformed the dependent variable using the square root of the inverse of the original variable scores. The scores associated with this tra nsformation were then used in the final model to run the linear regression models (discussed in Chapter 7). I chose this particular transformation because it was based on analysis of the residuals that produced a more normal distribution of the residuals i n the final model than with the untransformed scores. The on the high side, implying that the legitimacy scores tend to be on the high side. There are several e xplanations for the high legitimacy scores and the low variability in the responses among individuals. During data collection, I noted inconsistencies in the responses on scale items and follow up interview questions. For example, respondents would express a high level of agreement with statements on the scale, but would indicate otherwise in responses to follow up questions regarding their experiences. This tendency to respond positively to scale items may also explain the elimination of the negative stat ements in the post hoc analyses due to low item total correlations. I noticed that respondents did not consistently respond to the negative and, in some cases, respondents questioned the meaning of the statement. The decision to eliminate the negative stat ements during data collection was also based on this observation. 1 The response scale used to gauge frequency of interactions ranged from 1=not at all to 5=all the time.
146 Regression Model Quantitative data were analyzed using the statistical package for the social sciences software (SPSS) version 20.0. Descriptive statistics were run for both site specific an d aggregate data (i.e., all sites combined). I conducted linear regression with the social network variables along with the micro variable, the control variables, and the transformed dependent variable. Age was correlated with many of the variables and was thus excluded from several models due to issues of col l inearity. Figures 8 1, 8 2, and 8 3 display several socio demographic differences of the groups. The 47 research participants were predominantly male (77%) and tended toward middle age (mean of 46.7) Figure 8 2 shows that CORIDUP has the highest average age of 51, which is defined as older age in Bolivia (50 years or older). Thirty eight percent of the participants held at least a high school education and 28% obtained a university education. CORIDUP displays a somewhat higher level of education as compared to the other groups. Table 8 1 displays the four regression models, where the best predictor variables have a p value of approximately < .05. Model 1 included all network and socio demogra phic vari ables. Model 2 included the same variables, but did not include age in case the collinearity masked imp ortant effects. Model 3 dropped the non significant network variables and consisted of the most significant predictors (the group variable and meso degre e centrality) with the socio demographic variables not including age. Model 4 included only the most significant variables. This analysis provides several findings presented in all four models. Model 4 or the final model shows that the g roup variable (p=0 .002) is the most significant predictor of legitimacy and the meso degree centrality (p=0.029) variable is the most significant network related predictor variable The r squared of this model is .317 with a p level of .000, which means that both the
147 group variable 2 and the degree centrality explain a fair amount of variance in the relationship between the independent and dependent variables. The residuals of the model were normally distributed. In this particular model, the one significant social network va riable has been symmetrized and dichotomized. The variable of meso degree centrality is symmetrized on the maximum and dichotomized with a cut centrality is symmetrized on the maximum and dichotomized at gr indicate that the network position of degree centrality is the only significant network redictor of individual legitimacy scores and thus explains the variation in the individual legitimacy scores. This means that legitimacy score. As such, forum members who perceive the forum to be less legitimate hold a less central position in the network as compared to those who perceive the forum to be v ery legitimate and hold central positions in the forum. These findings only partially corroborate my predictions based on the theoretical model, as I expected greater correlations between the other social network variables and the dependent variable. In th e final model, I included the group variable, as there is a difference in the means of the dependent variable across groups (Table 8 2). By including the group variable, the model adjusts for the differences in the means or any other variable that I did no t explore in this study that may explain this difference. Table 8 2 presents the comparison of mean variable scores across groups and includes the untransformed legitimacy score. One important observation with respect to these mean scores is that responden ts generally responded more similar ly within 2
148 groups than across groups. This is an indication that respondents in each group have perceptions that are similar to each other but that are different from respondents in other groups. What explains the differ ence (or variation) in legitimacy scores across groups? Results from the linear regression model indicate that the group variable (p=0.002) is the most legitimacy of each group differs when compared to the other groups. The final model also indicates that the meso degree centrality variable is the most significant network related predictor variable centrality in the network is highly correlated with their perceptions of forum legitimacy. ANMIN A has the lowest mean legitimacy score of the three groups, whereas CORIDUP and BOCINAB have similar scores that are significantly higher. Legitimacy scores were high with low variability in the responses among individuals; however, the ethnographic data capture greater variability, especially with respect to acy score is significantly different across cases, and forum legitimacy is largely reflected in network positions, particularly network centrality, my ethnographic data also point to the importance of leadership roles in forums and their influence on forum legitimacy. Compared to all forum leaders, CORIDUP leaders assume a highly influential role within the forum, directly influencing its legitimacy. My ethnographic findings also suggest that this relationship may be clarified in all cases by additional fac tors not included in the model. In the remainder of this chapter, I intend to expand our understanding of the predictors of legitimacy by examining the patterns uncovered in my ethnographic research.
149 Qualitative Model Results In this section, I use ethnog raphic findings to discuss how the leadership roles assumed internal legitimacy. Based on my ethnographic findings, forum leadership greatly affects the transparency of information and the opportunity of forum members to participate in the forum. I found that the relationship between forum leaders and supporting institutions (i.e., NGOs and ernal legitimacy Relationship with Supporting Institutions The theme of platform leadership emerged as an important feature affecting the leaders generally contro l the lines of communication among members and are most influential in decision making. One important factor that I found to affect the ability of all members to voice their concerns in meetings, shape the agenda, and influence decisions is the forum leade relationship with, and dependency on, supporting institutions. This relationship seems to on making. dependen cy seems to represent a consequence of the limited technical capacity characteristic of all the forums. In the case of CORIDUP and BOCINAB, the supporting NGOs are instrumental istence, operation, and progress as they provide technical a nd logistic al support to forums. In contrast, ANMIN A works most closely with SERNAP in a more institutionalized collaboration in addition to several NGOs that have funded projects in the area. CORIDUP has maintained a very close working relationship wit h the NGO, CEPA, also provides
150 spend a considerable amount of time at CEPA, where they meet informally and frequently Environmental Justice Office, who assist them with legal and technical issues. I observed that these leaders routinely follow the recomm endations legal advisor who is an environmental lawyer at CEPA. This particular individual plays an important role in CORIDUP, as he provides significant legal support to the forum, such as drafting the letters of inquiry and formulating the d emands to the relevant government institutions p rimarily the Vice Ministry of government institutions to ensure that actions taken and words spoken are legally sound. As su the information relevant to CORIDUP and assume the responsibility to act on behalf of the affected communities on a daily basis. the financial resources to be able to, for example, cover the 6 6 trips that we made in done (get the ED 0335 passed). I also found that CEPA highly values the relationship it has with CORIDUP, especially with its president admiring him for his strength and dignity as a leader. At the same time, however, the decision making have also placed limits on the involvement of, and leadership oppo rtunities provided to, other members. 3 Some members, both active and inactive, feel excluded from the 3 As explained in Chapter 4
151 they once had. In addition, the rich leadership and technical e xperience of CORIDUP leaders, credited to their collaboration with CEPA, has also created an atmosphere in which some members feel that they are not as skilled as these leaders and do not have the opportunity to particip ate in efforts to advance the ED 0 335. Several members have also grown frustrated with and have lost faith in CORIDUP ing the forum, yet they infrequently hold member meetings to inform the members of these issues, even when s agenda and its strategies guiding the a culture of misinformation, where those who feel excluded and uniformed suspect that EPA. Of the four NGOs that provide BOCINAB financial and technical support, three work primarily with the campesino organizations whereas only one works strictly with the indigenous organizations. Through key informant interviews and direct observation, I found that the agendas the opportunities of the actual members t o shape the agenda, but it also affects the relationship among the member organizations. Various employees of these NGOs acknowledge that the NGOs are funded for specific projects and must actually implement the projects in order to receive funding, they are often inclined to promote certain project specific objectives over other objectives.
152 making process through formal agenda se veral days prior to meetings. 4 Prior to discussing the concerns arising from their NGOs to work on meeting agendas, and the leaders often succumb to the suggestions of the NGOs. This tendency is largely reflected in the resulting agenda. Since 2010, BOCINAB has been deliberating over how to prepare, and the ultimate form of, proposals for several national environmental laws. Considering that proposal development require s technical expertise and knowledge of the legal system, skills un commonly found in these social organizations, the proposal development process requires support from NGOs. Explaining the need for such assistance as well as the lack of expertise of the lea ders, one I also found that weak leadership skills and indifference to inspiring collaboration among leaders believe facilitate the collaboration for them. One former member reflected on this situation: The leaders have gotten accustomed that the and leaders] have demanded that the NGOs stop working in this way and that its leaders [take on the responsibility]. Despite the critical technical assistance they provide to BOCINAB, I observed that the campesino 4 The NGO offices also provide a central location for leaders to meet and work given that BOCINAB does not have its own office.
153 and indigenous member organiza tions. NGOs working either exclusively with campesino or exclusively with indigenous organizations promoted specific proposals in line with their own agendas. Accordingly, I found that rather than encouraging the two sectors to come together to collaborate collaboration. This division is reflected in the unequal balance of interests within the forum, as the demands of the campesino sector increasingly dominate those of the i ndigenous sector. 5 In a space specifically designed to unify the campesino and indigenous sectors of the Northern Amazon region, campesino demands prevail and drive the forum. The dominance of the campesino demands may, in part, be attributable to BOCINA dependency on NGOs. Although those who work for the supporting NGOs expressed their awareness of the fact that they may have f orced these spaces and encouraged the persistence of power themselves from this space Moreover, informants from the NGOs suggested that it will be very difficult for the forum to sustain itself if the leaders fail to take such initiative. In contrast to the close relationship between the leaders of CORIDUP and BOCINAB with NGOs, ANMIN working relationship. This relationship is accompanied by an inherent reliance on state funding. As a state institution with state resources, SERNAP is expected to disperse the necessary financial and technical resources to assist in protected area management, demonstrating its primary role of the director of the area is to assure the viability of the nexus between the state 5 The majority of member organizations are campesino organizations t hat receive assistance from three of the four NGOs.
154 SERNAP and A N MIN making as well as t heir opportunity to exercise meaningful control over management activities. Considering that the objective of the co management model is to promote shared decision president of th e forum generally have exclusive control over the decisions made regarding area management. With respect to this relationship, I found that the amount of control the forum exerts over projects and funding in the area largely depends on the initiative taken leadership. For example, during my field work, the president worked in a particularly proactive manner and closely coordinated with the director to seek out funding to finance and manage projects in the area. 6 The president also held meetin gs once a month in order to both utilize the state funding allotted to meetings 7 and work more closely with the other members to improve area management, thus facilitating member involvement in decision making. However, in contrast to previous terms when l eaders were solely representatives of different social organizations, the newly elected leaders are now all mayors. 8 I learned that interest in capturing funding from SERNAP and funding allocated to NGO projects in the area pursuit of leadership roles. Assessing these new dynamics, key informants suggested that these new leaders, who are all in their first terms as mayors, seem 6 This individual also worked for an organization in La Paz during his presidency. As SERNAP is also located in La Paz, this close physical proximity facilitated the coordination between the president an d director of the area. 7 SERNAP provided funding for committee meetings; if these funds were not used then less funds would be distributed during the next funding cycle. 8 Elections were held in August 2012.
155 overly concerned with serving the interests of their particular municipalities in order to gain po litical support for future reelection, rather than serving the entire protected area. To illustrate, the current president failed to share important information with the rest of the committee on a decision that would affect the area, such as the constructi on of a road in his municipality for which he did not apply for an environmental license. 9 Moreover, key informants suggested that these leaders are interested in expanding their power by taking advantage of the supporting institutions at the expense of fa cilitating more involvement of the forum members. Leadership Experience The cases explored present variations in leadership experience and skills, which also help explain the leadership roles assumed by forum leaders. I highlight the importance of the te rm length of forum leaders in order to shed light on the significance of the consistency of forum leadership on the forum. The term length of leaders and the manner in which leaders are elected differs across groups, representing important considerations i n examining leadership roles across cases. In the case of CORIDUP, its president and board of officers were elected in January 2007, re elected by unanimous vote in August 2009, and remain unchanged as of March 2013. The work of the current president was so significant in a chieving the ED 0335 that the affected communities requested his re election in 2009. However, president, along with the others who remain on the board of officers, have overstayed their two year term by two and a half years. Moreover, as discussed, 9 All projects within the area must obtain an environmental license to begin any such project. This individual did not communicate this project with the committee or SERNAP which is required by law.
156 ANMIN A also holds elections for its board of officers every two years. Its president served his first term from 2008 to 2010 and was elected vice president in 2010 for the next term. The president elect for the 2010 term, however, was unable to serve due to familial issues and, as a result, the forum asked the former president (and newly elected vice president) to assume the position o individual as a result of his respected work ethic during his first term as president. BOCINAB, in contrast, has a different way of conducting its elections, as its board of officers (First, Second, and Third Coordinators) systematically rotates by member organization every two years. As opposed to CORIDUP and ANMIN dictated by this systematic rotation, whereby the pr esident of the indicated organization becomes the leader of the forum. 10 While this rule maintains a democratic foundation, it also assumes that the president of the delegated organization for First Coordinator is willing to dedicate time to this position. In the most recent election in late November 2012, I observed that the delegated While holdin g elections every two years and allowing anyone to run for elections are generally considered fair and just practices under democratic norms, evidence indicates that CORIDUP and ANMIN change of forum leadership may inhibit the progress made by the forum. leadership has resulted in a loss of organizational memory and the deterio 10 As explained in Chapter 4, every election cycle, or every two years, the leading organization that w ill preside over BOCINAB as First, Second, and Third Coordinators, rotates. The members of BOCINAB change based on the election cycles of their respective organizations, and the president of each social organization automatically becomes a member of BOCINA B.
157 mission, causing its leaders to struggle to maintain the progress of the forum as it relates to the developmen t of various policy proposals. The systematic rotation of member organizations, which mandates members to assume leadership positions without taking personal initiative, may and thus the viability of the forum. Moreover, leadership experience and the corresponding commitment of lea ders to forums may explain the variations in the internal dynamics of forums, and may have important implications for the differences in forum legitimacy across cases. Conclusions Results from the final regression model suggest that forum legitimacy is re flected in network centrality, indicating that the most central actors in the networks, or forum leaders, are present different levels of legitimacy, with CORIDU P and BOCINAB showing the highest levels of legitimacy. My ethnographic findings suggest that the relationship of leaders with supporting institutions seems to provide the leaders with important resources and leadership capacity, which enhances the leaders making processes, and thus shapes the democratic quality of the forums. Also, t he dependency of the forums on supporting institutions seems to represent a consequence of their limited technical competenc e. However, as the relationship of dependency presented in the particular forums varies, the nature of internal legitimacy also varies. Additionally, evidence indicates that the type of leadership roles assumed by forum leaders at least partially is influe nced by their leadership experience in the forum, which may be represented by the length of time served as forum leaders. Although the theme of consistent leadership exceeds the scope of the initial discussion on internal legitimacy here, its influence on the legitimacy of forums over time represents an important consideration.
158 Figure 8 1. Percentage of males and females by group. (Group A=CORIDUP; Group B=BOCINAB; Group C=ANMIN A) Figure 8 2. A verage a ge of forum members by group. (Group A=CORIDUP; Group B=BOCINAB; Group C=ANMIN A)
159 Figure 8 3. Highest level of education completed by forum members by group. (Group A=CORIDUP; Group B=BOCINAB; Group C=ANMIN A)
160 Table 8 1. Linear regression m odel s of network variables and socio demographic variables on forum legitimacy perceptio ns. Variables Model 1 Model 2 Model 3 Model 4 (Constant) 1.603 (0.417) 0.001 1 .191 (0.300) 0.000 1.406 (0.202) <.000 1.366 (0.132) <.000 GROUP 0.213 (0.075) 0.008 0.223 (0.076) 0.001 0.236 (0.072) 0.002 Sex 0.111 (0.171) 0.520 0.079 (0.171) 0.650 0.046 (0.174) 0.793 Age 0.009 (0.007) 0.171 Education 0.016 (0.085) 0.848 0.005 (0.084) 0.950 0.013 (0.082) 0.874 MICRO 0.059 (0.076) 0.447 0.0643 (0.077) 0.409 MESO Degree Centrality 0.006 (0.003) 0.056 0.007 (0.003) 0.026 0.006 (0.003) 0.039 0.006 (0.003) 0.029 MESO Betweenness Centrality 0.075 (0.05) 0.138 0.074 (0.050) 0.149 MACRO Degree Centrality 0.004 (0.003) 0.099 0.003 (0.002) 0.175 R squared 0 .419 0.390 0.319 0.317 p value 0 .005 0.005 0.002 0.000 Values shown in each cell Beta and standard errors in parentheses with p value below these values
161 Table 8 2. Comparison of mean and standard deviation of legiti macy, socio demographic variables, and network variables by case. GROUP Legit Sex Age Edu MESO Deg ree Cent rality MESO B etweenness Cent rality MACRO Deg ree Cent rality CORIDUP Mean 9.0536 .2778 50.0556 2.1111 58.8235 .8170 55.5556 SD 1.00511 .46089 11.42995 .90025 24.79306 .67459 30.48039 BOCINAB Mean 8.9301 .3333 44.7500 1.6667 27.2727 2.2727 51.5152 SD 1.11254 .49237 7.25039 .65134 20.51186 2.35041 29.60632 ANMIN A Mean 7.96 43 .1176 44.7059 1.8235 33.8235 1.0294 46.3235 SD .91582 .33211 13.17082 .95101 18.49552 .62325 28.81606 Total Mean 8.6280 .2340 46.7660 1.8936 41.7255 1.2655 51.1847 N 47 47 47 47 47 47 47 SD 1.10387 .42798 11.31604 .86562 25.27309 1.41008 29.29184 (CORIDUP n=18; BOCINAB n=12; ANMIN A n=17)
162 CHAPTER 9 CROSS LEVEL LINKAGES AND EXTERNAL LEGITIMACY eir relationship with the national government and the communities they purportedly represent to capture the recognition of the forum by these external actors. Founded on the argument that societal acceptance of an organization or network and its subsequent survival depends on its attainment of support from relevant entities in its environment (see Human and Provan 2000), g overnment responsiveness and information sharing with communities are considered important indicators of Accordingly, it is expected that a legitimate entity enjoys recognition from both insiders and outsiders and can attract internal and external support and resources (Human and Provan 2000). In assessing the meso macro linkages, I use ethnographic evidence from interviews with to forum demands. I examine the meso micro l inkages using ethnographic evidence from interviews with forum members, which suggests that there is very little information dissemination and articulation between forums and the populations they represent. In assessing the meso macro linkages, it is impo rtant to note the nature of the relationship between each forum and the government. CORIDUP has the most formal relationship with the government, as it has worked closely with government actors to monitor the advancement of the ED 0335. BOCINAB, on the oth er hand, has a less direct working relationship with the government, although it is beginning to insert its demands into national discussions regarding the various environmental laws included in the national legislative agenda. Similar to the other
163 protect ed area forums, ANMIN A can and has influenced policy decisions in the past; 1 however, it has a less politically influential role at the national level than CORIDUP 2 Despite the very different nature of these relationships erests remain a constant interests, the government generally displays aversion to organizations that question its actions, disfavors NGOs, and dismisses environmen tal demands. Meso Macro Relations: Government Responsiveness to Forum Concerns I use ethnographic evidence from interviews with both forum members and representatives from government institutions (public officials and staff members) to assess the national I also use in depth interviews with strategic actors in national government institu tions and direct observation to assess government responsiveness. CORIDUP Before investigating the ethnographic data in detail, it is useful to recall survey responses How satisfied are you and in CORIDUP, the mean score for government responsiveness to forum demands was 2.38 on a 5 point response scale. The majority perceived the government to be considerably unre sponsive to its demands 1 Refer to Chapter 4 for an explanation of how ANMIN A along with the other protected area forums have come as national policy. 2 This lack of influence results primarily from of its close working relationship with the decentralized state institution, SERNAP.
164 with about 60% who were either not at all satisfied or hardly satisfied. In contrast, only16.7% was either somewhat satisfied or very satisfied with the government institutions. Some of the most common perspectives held by CORIDUP members regarding the is to make us tired and to make us and our In interviews with public officials in the Vice Ministry of the Environment, the Ministry of Mining, and the Ministry of Rural Development and Land, I found that these officials know CORIDUP very well because of its persistence in pursuing Executive Decree 0335. 3 The officials interviewed, particularly those from the Vice Ministry of the Environment (VME), all expressed similar sentiments of resentment and/or frustration with the way CORIDUP conducts its work. Several common themes emerged from the responses and depict the tense relationship between the representativeness of the local populations, and (3) the authoritative position it assumes. I found capturing a more complete All In particular, they express that they (government institutions) have done. St ated succinctly by one official, 3 The Vice Ministry of the Environment has had the most frequent contact with CORIDUP because it coordinates the six strategic wor king groups implementing projects for the ED 0335.
165 contributed so much but the truth is that these people [CORIDUP] will never be content [with ou ement that the leaders of CORIDUP have used in public events the contamination from mining into historical perspective: that the state [this administration] solve the se problems [in just one term]! us It became clear that, in addition to taking these accusations personally, these officials have reached the threshold of their tolerance levels. As one informant said, ations, but we also have to These public officials also expressed their concern that a relatively small number of people speak for CORIDUP, suggesting that the voice of the forum may not be representative of the whole forum. Recalling C invite a sm
166 personal interests, and intentions. operation run by one of the lar gest private mining companies in Bolivia, Inti Rymi, has fomented the tension between the VME and CORIDUP over the past year. Although independent auditors conducted this environmental audit, CORIDUP argued that the audit did not include important observat CORIDUP leaders (with the help of their legal advisor) have written several letters to the VME stating dissatisfaction with this audit and demanding a reconsideration of its re sults. CORIDUP lacks the legitimate authority to serve a supervisory role and hold the state accountable. These officials suggested that CORIDUP leave this role to the Controler a, the state institution with the legitimate authority to hold the government accountable. Furthermore, they their work and make efforts to coordinate with the municipalities and departmental government, which, as they explained, are the real legitimate political actors, democratically elected by the affected populati ons. 4 Along these lines, the officials argued that the state requires that these 4 Although recognized that the municipalities neither participate in the strategic working groups nor pri oritize environmental mitigation in the projects they implement. Informants indicated that the municipalities, particularly the mayors, have other priorities (most are miners and thus their interests do not tend to coincide with environmental remediation), which also makes coordinating with them very difficult for all of the institutions involved in the strategic working collaboration with the mu nicipalities since all federal funding is channeled through them (such as the Program Cuenca Poop which brought in 14 million Euros dedicated to environmental remediation in municipalities included in ED 0335).
167 CORIDUP, as a social organization or a representative of civil society (not an elected p olitical authority), lacks this authority. I now turn to the underlying theme of political interests. One of the first reactions of the officials expressed regarding CORIDUP is that it works with an NGO. This was one of the major criticisms expressed by the informants/government officials, as they mentioned letters sent from mining cooperatives in Oruro to government ministries, informants maintained the belief that C are most frustrated by the support given to CORIDUP by the NGO, which they claim has 5 They also attributed the degree of preparation e it s work with this NGO 6 especially those t hat critique be questioned. It looks for things so that [CORIDUP and other such dissenting organizations] 5 On several occasions, CEPA, the NGO workin g with CORIDUP, has published articles in several newspapers to voice the demands of the affected communities and note the unresponsiveness of the government. 6 ced its dismissal of many NGOs. NGO sentiment is likely rooted in its distrust of international donors that supported the neoliberal reforms of the 1990s. Even though social actors supported the reforms in principle, in practice they t 132).
168 An event held during my return visit to Bolivia in December 2012 revealed a ll of the previously discussed themes and sentiments related to the relationship between CORIDUP and the government. The VME, along with the other involved ministries, held a public evaluation meeting 7 in Oruro related to ED 0335 in order to present their progress to relevant civil society actors. I uncovered some important findings with regards to the relationship between CORIDUP and the government. To further analyze this relationship, I focus on both the role of the VME in producing the meeting and the a ctions taken by the VME at the meeting. The VME was the coordinator of the six working groups attending the meeting and the organizer of the event. Before attending this particular meeting, I reflected on a prior meeting that I attended in October 2011 wi th the same purpose. For several reasons, I found this most recent meeting organized much differently than the former meeting. First, I was surprised that the organizers did not request a microphone for such a large space, especially since the space was f illed with at least 120 people. I was also surprised to find that, unlike the prior meeting I attended, nobody moderated this meeting. Considering that this event consisted of actors with conflicting interests, such as miners and community members who are not miners (those affiliated with CORIDUP), the absence of a moderator seemed to signal 8 At the beginning of the four hour meeting, the government authorities, sitting at a long table in front of the audience, explicitly sta ted their purpose to give a report on the progress made in each working group, and nothing more. However, anyone with experience with the mining sector knows that serious tension often exists between the miners and other community 7 every six months. They involve different civil society actors (community members, CORIDUP, miners, mining companies) as well as relevant governme nt actors. 8 According to informants, the event in December 2012 marked the first one in which the miners were officially invited. Prior to this meeting, miners did not have much knowledge of the ED 0335.
169 members present and that, as such, considerable debate would ensue apart from, and perhaps in response to, the reports made. As it turned out, there was substantial and hostile debate between repeated accusation that CORIDUP is an NGO. I observed that, by remaining silent and choosing not to direct or facilitate the discussion, the public officials from the VME reinforced the aggression that developed as a result of this accusation. Two hours before the close of the event, secretary, left the room with fists held high in anger and frustration. At that point, I asked myself: I let two weeks pass until I visited the public officials and staff members at the VME in La Paz to get their perspective on what occurred during this event. After speaking with the three officials who led the event, I learned that their intention to solely present their work, as well as their lack of facilitation, was part of their strategy. These governmen t representatives continued to express complete exhaustion and disappointment with CORIDUP and its constant demands. disappear so that we could just deal with the indi genous authorities [it would be] better. We T his sentiment, along with the way the event unraveled, reveal s the tense relationship between the government and CORIDUP, demonstrating t BOCINAB point response scale. The majority, or about 61%, of forum members were either somewhat
170 satisfied or very satisfied with the government institutions, while about 25% were either not at all satisfied or hardly satisfied with these institutions. The majority expressed satisfaction with the gov conclude the land tenure process in the Northern Amazon from 2006 2008. 9 However, through in s political Minist Cachuela Esperanza dam 10 in Beni, the government has considered BOCINAB as an opponent to the government. With the aid of an environmental NGO, BOCINAB sent letters to the appropriate government ministries to solicit information on the environmental and so cial impacts of the project but, according to former members, the government never responded to these letters. Instead it accused BOCINAB members of being environmentalists and part of a right winged opposition to the government both accusations derived fr om its work with NGOs. As a result, functioning of BOCINAB, especially since the government h as developed a relationship with several member organizations of BOCINAB and provides special privileges to these 9 As discussed in Chapter 4, BOCINAB assumed a s trategic role in these efforts by both representing and involving rural communities that gained their citizenship in the process. 10 Refer to Chapter 4 for an explanation of this project.
171 organizations. 11 Some feel that these organizations, which no longer participate in BOCINAB, have been co opted by the government. A founder an d former member of BOCINAB, who optation, these shameless opted by the government they isolate themselves The absence of these organizations in BOCINAB has affected its strength and unity, which is exactly what several forum members indicated as the reason for why BOCINAB is not recognized as a legitimate entity at the national level. Former members and other key informants suggested that these strategic relationships between the government and the particular organizations in BOCINAB are meant to stifle the through co opting organizations, which creates divisions betwe en previously collaborating organizations. One former member expressed his frustration by what he referred to as BOCINAB] a ton of benefits which is not legal. An organi zation has to fulfill certain requirements [such as constitutionally establishing its legal personality] so that it can be legally In this way, informants suggested that BOCINAB has been stigmatized by the government for the purposes of undermining its work and silencing its voice. 12 11 Over the last two years, the member organizations from Cobija have been coopted and have become less interested in participating in BOCINAB. 12 Several key informants suggested that if BOCINAB was stronger internally, it could withstand these challenges.
172 Despite this internal division, which the organization believes is exacerbated by the environmental law s on the national legislative agenda. As such, it is beginning to engage in discussions related to relevant policies and make connections with strategic government actors to insert its demands into the emerging environmental regime. In November 2012, BOC INAB was invited by the Chamber of Deputies 13 to present a proposal for the Amazon Law in La Paz. A representative from BOCINAB was included on the agenda in addition to a representative from the indigenous sector. The legal advisor 14 of the indigenous organ izations in BOCINAB presented the proposal for the indigenous sector, while the campesino sector presented a proposal that they developed several days prior to the presentation. 15 The campesino proposal closely resembled the indigenous proposal since it use d as a reference. Following the presentation to the Chamber of Deputies, I interviewed several public officials to get a feel for the way the relevant government institutions intend to respond to as well as in other environmental laws. When asked how the government plans to include interests of different stakeholders in the Amazon region, the individual working on developing a proposal for the Amazon Law and Forest Law indicated that BOCINAB woul d be included in this process, along with the other relevant social actors affected by the legislation. He also indicated that, while 13 A government body that is closely analogous to the House of Represe ntatives in the United States of America. 14 15 As mentioned, the indigenous sector had worked on a proposal for the Amazon Law that was developed wit h one of the supporting NGOs. Several days prior to this event in La Paz, BOCINAB held a meeting in Cobija to, in part, work on a proposal to bring together both the indigenous and campesino demands. The campesino organizations, however, had not developed a proposal prior to this meeting. Nevertheless, BOCINAB never reached this agenda item in the meeting. Instead, three days later, the campesino organizations decided to work together on their own proposal in order to prepare for the event.
173 informant also suggested that BOCINAB incorporate more specific demands for the law and be persistent with those demands. It is clear that particular individuals working for govern ment difficult to determine whether such demands will actually be incorporated into policy until the policy is actually ratified. Another key informant who is currently working as a consultant for the government to 16 and suggested they change their proposal to incorporate state initiatives and local agre ements that include all affected sectors. He also pointed to the importance of a clear vision at the local levels in influencing national policy: the problem is that the local level is divided. The key is that the national this requires a state that regulates and puts order to things, but it als o requires the right attitudes from the local level to come together in agreement. This informant also discussed the incapacity of the government in addressing all issues and actors requiring consideration during the development process of environmental la ws. In of resources to help incorporate the relevant actors into the development process for such legislation. Similarly, one of the Deputies working on the Amazon Law emphasized that there is institutions are not responding to environmental issues. She pointed out that the government does 16 This informant
174 not even provide interlocutors, like her, with the funding to disseminate information and gather 17 the discussion of development had a more inclusive start. These key insights suggest that, although BOCINAB may have been stigmatized by the government in the past, potential exists for the government to accept its demands and incorporate these demands into the emerging environmental legislation. Key individuals that work for the demands; however, the results remain dismissal of NGOs, it is impo rtant for BOCINAB to present its demands as independent from these entities. ANMIN A As explained, ANMIN A does not coordinate directly with government institutions, but representatives of SERNAP fulfill this role for ANMIN A. As a result, I do not include the responses to the interview question used to assess the satisfaction of ANMIN these institutions. Through interviews with key informants from SERNAP and others who interests, as they relate to the administration of natural resources in protected areas, have influenced efforts to legalize the co management model. This suggests that the government intends to decrease the level of participation of local populations in decision making processes that affect protected areas. Before concluding, I will briefly summarize this suggestion. 18 17 18 Refer to Chapter 4 for an explanation of efforts made to develop and legalize the co management model.
175 After nearly two years of working on a formal agreement for co management with SERNAP, ANMIN screeching halt. In 2011, the management program, directly ending its support and encouragement of the co management model in protected areas. During my fieldwork, from 2011 to 2013, the four pro co management program were gradually eliminated. Many informants suspect that the reduction represented by ANMIN management committee. Despite the supposed importance of the public participation of communities in in maintaining primary administrative control and decision m aking power over matters related to 19 In addition, SERNAP has failed to take more progressive actions to move proposals forward to legally establish protected area co management. According to ke y informants, some individuals with decision making power in SERNAP fear losing power to the communities in protected areas. These actions and circumstances, taken together, affect the once solid relationship between SERNAP and those social organizations i n protected areas that largely represent the local populations. Furthermore, it seems that the government does not want to lose the control that it has in protected areas, which is a topic that has become more contentious since the most recent conflict, 19 Natural resources in protected areas are very important sources of national revenue. As protected areas make up renewable resources located within its 22 protected areas are highly valu ed.
176 o ccurring in TIPINIS, one of the oldest protected areas and an indigenous territory. 20 One informant who previously worked for SERNAP summarized the current situation: There was a project for co management but TIPNIS happened and it came to a there is no way to advance it now because it comes from a public institution with a lot of conflict inside. The other problem is that the proposal developed for co ever get finished cooking obsolete. The government wants the process to fall apart because it has witnessed the power of the people and their management in protected areas particularly with what happened in TIPNIS. On top of every thing else, there are no longer professionals in SERNAP who can help guide the process of moving the co management proposal forward [at the national level or strengthen its function within protected areas]. The ethnographic evidence presented suggests that the actions taken to stymie efforts to advance the co intentions to decrease the participation of local populations in decision making regarding matters related to protected a rea management. In addition, recent events in the environmental sector al actions to undermine the legitimacy of these forums contradicts its emphasis on participatory decision making. esponse to the TIPNIS conflict exposed a tension within the government as well as its resentment of NGOs. The tension that was exposed its vision for economic devel opment. Both this tension and the resentment of NGOs have shaped 20 Refer to Chapter 2 for an explanation of the conflict surrounding TIPNIS.
177 Ethnographic findings highlighted here reveal this dynamic as it is has played out in the environ mental sector. Institutional constraints I identified several institutional conditions that also environmental concerns. These conditions incl ude the weaknesses of the current environmental legislation, a lack of sufficient economic and human resources a lack of technical competence, and the institutional instability of government institutions. To illustrate how these conditions affect the gove Thro ugh interviews with government officials and direct observation in public events, informants discussed the weaknesses of the current environmental regulations in the mining sector indicating tha t they are insufficient 21 in dealing with the rapid increase in small mining operations. One informant summarized this problem : What is happening is that the Environment Law was designed so that the large corporations can comply with the environmental regul actions and control, but it has not taken into consideration that the small operations ol their In interviews with public officials from the M inistry of Mining, I learned that the large mining companies, unlike the small operations (or cooperatives), all have their environmental licenses and implement measures to control their practices. 22 Informants point to the irresponsibility of small operati 21 The Mining Code of 1997 is currently under reform and is expected to address some of the issues regarding the control of environmental licenses. 22 Obtaining environmental lic enses requires money and time, neither of which a small mining operation has.
178 infract ions. In addition, these cooperatives are also hard to control because they continue to multiply in number and tend to migrate from place to place, where they engage in temporary, opposed to permanent, activity. lity to fulfill its role in the environmental sector is the issue of institutional instability. During my fieldwork, the government carried out a number of executive decisions to replace the environmental ministers and their technicians. While I understand that such changes occurred in previous government administrations, informants confirmed that these types of changes are much more common in the current previou s administrations. As such, the inconsistency of authorities and public officials affects the institutional memory 23 and the progress made in the environmental sector. To illustrate, the Minister of the Environment has changed three times since September 2 011. As a consequence, the advances made in her term were paralyzed with the entrance of the following ministers. This forced CORIDUP to make efforts to establish relationships with the new public officials, which, as previously discussed, has presented ch allenges. Furthermore, I found that institutional memory has deteriorated since 2009 when authorities and public officials worked closely with CORIDUP to formulate and he lp promote the passage of the ED 0335. 24 23 This term refers to organizational memory or the consistent knowledge of organizational affairs over time. When technicians and public officials are changed often, there is li ttle time to learn and gain knowledge of progress made in the past to help orient them as to where they should direct their attention. Therefore, the loss of institutional memory is expected to affect the progress made in the future. 24 The public officials in the VME began their work in 2011 and were thus not involved in the development of the ED 0335 with CORIDUP.
179 Section Conclusion Ethnographic findings reg arding the relationship between forums and government animosity towards NGOs, in particular, seems to affect the legitimacy of these forums. As for ANMIN natural resources in protected areas has primarily served to reduce the legitimacy of management committees. My ethnographic research indicates that the influence of the national political climate one characterized both by rapid legislative change and reconstruction as well as conflict ands at the national level. These dynamics reveal how the nature of governance at the national level often fails to respond to contrary grassroots demands. In addition to these findings, I also identified several institutional conditions that limit the go such as the weaknesses of the current environmental legislation, a lack of sufficient economic and human resources, a lack of technical competence, and the institutional instability of gover nment institutions. Meso Micro Relations In this section, I examine the relationship between forums and the populations they ostensibly represent in order to gauge the degree of recognition of the forum at the micro level. Although I originally set out to explore local level awareness and knowledge of forums, gaining access to the communities and local organizations was quite challenging. 25 As such, I asked 25 Besides the geographic distance and distribution of communities as well as few transportation options available for traveling to communities, I faced other challenges. In the case of CORIDUP, for example, neither its leaders nor CEPA, the NGO, were able to provide me with a list of communities from which I could select a random sample. Limited by time and money as well, accessing communities was particularly challenging.
180 forum members in follow up interviews to indicate how effectively the forum communicates with the pop ulations they represent. Across cases, I found very low levels of information dissemination and articulation between the forums and their communities. I also identified several challenges to information flow in each case. CORIDUP CORIDUP represents approx imately 80 communities in the four river sub basins affected by mining activities. While I was unable to gather the necessary information to determine how forum membe existence. Forum members representing the respective sub basins in CORIDUP are responsible the sub basins. However, through follow up and key informant interviews, I learned that members infrequently communicate with the communities they represent and that the communities are losing confidence in CORIDUP. According to forum members, support and con fidence of communities is very important for CORIDUP. Members contend that, while CORIDUP was once effective in getting the ED 0335 to pass, a lack of information regarding subsequent progress of the ED 0335 and a lack of tangible results 26 from the ED 0335 has weakened its relationship with the affected communities. confidence in CORIDUP and developed the impression that CORIDUP does not do anything. Informants cont end that this sentiment is related to the lack of information disseminated by CORIDUP and its failure to come up with concrete solutions to the problems resulting from the 26 Remediation efforts on the ground
181 contamination. O ne member t with the When asked about the infrequent communication with communities, most forum members suggested that they were not inclined to share information with communities because they have very little to share. As one that good things will come, but that some lose faith because they are tired of waiting. Nevertheless, others understand that the process is lengthy, and so they continue to wait for positive results. The weak nexus between CORIDUP and the communities is clearly reflected in the communication patterns among CORIDUP members. As discussed in Chapter 8, CORIDUP leaders fear that people will become frustrated and lose faith in their leadership if they inform the people of the actions they take where concrete, direct, and immedi ate results have not been realized. The failure to share information, however, has just this effect, as it leaves people uninformed and frustrated, breeding further misinformation. In this way CORIDUP risks losing its credibility by failing to maintain con tact with its communities. To illustrate, the communication gap between CORIDUP and the affected communities has not only resulted in a lack of confidence in CORIDUP, it has also generated a loss of trust in sonal interests. One informant captured these many years ago did we get the ED 0335 to pass and did the Vice Ministry
182 not all community members are affil iated with CORIDUP since many have chosen to work for of receptivity in these communities. One informant explained this dynamic work with the mining companies disqualify CORIDUP because they have been convinced to he To further illustrate this point, one member reflected on her experiences organizing within her community where a major mining comp any is based and where more than half of the community is employed by the company. She explained how it became impossible to share information regarding CORIDUP with her community, as many community members receive m the companies. Four years ago, she and other community members fighting against the contamination stopped meeting publicly in the community and began to meet privately because of threats from the local mining company. Emphasizing the influence of this co mpany in her community, she added that even the municipal representatives represent the indigenous authorities that will defend their demands as
183 To further depic t the dynamic between CORIDUP and the affected communities, I turn to certain public evaluation events involving affected communities and other actors attending these events in October 2011 and November 2012. Responding to the large turnout of communities in suggested this could represent a posi tive step forward for CORIDUP: this helps us a lot with our rela tionship with the communities. Yet, during my field infrequently, even when there was a considerable amount of information to share. The infrequen cy and members as well as between CORIDUP and the affected communities. When meetings occur more frequently, the idea is that the r epresentatives would be informed and share this information with their respective communities. In my return visit in December 2012, I attended another meeting, or evaluation, of the ED embers, in addition to residents from the different municipalities were present, notably less community members to represent all affected communities they made few efforts to publicize the event so as to encourage participation from the relevant municipalities. 27 information with the active members and not with those who rarely or never attend meetings. In the public event, the few community members present resulted in a lack of support for 27 Days prior to the event, I was traveling to a large municipality with one of the members who asked the president if she could announce the event on the radio. Although it is widely known tha t radio is the most effective means of co mmunication in these rural towns, he did not permit her to make such an announcement.
184 CORIDUP and its leaders, particularly during the most heated moments in which the miners BOCINAB BOCINAB represents 15 social or ganizations in two departments that make up the Northern Amazon region. As the forum members are presidents of their social organizations, they communicate forum issues to the presidents of their sub organizations and these individuals are then expected to share information with the presidents of the respective communities. I found that this information, however, tends not to make its way to the communities and therefore ons identified for this dynamic is the geographic expansiveness of the region, which challenges the local relevance and interest in issues as well as the personal interests of the leaders responsible for sharing information. In addition, the geographic dis tance and remoteness of the communities makes spreading information a particular challenge for local leaders. Based on responses from forum members, I found that some forum members tend to share only partial or basic information with their sub organizatio ns and, in some cases, they simplify the information to make it more understandable for their communities. This may also pertain to the limited knowledge of, or interest in, such issues at the local level, as discussions of policies are often considered ob scure until results are felt on the ground. Some informants also recognize the tendency and preference of leaders to be the most syndicate or union like organizations, 28 which are structured hierarchically, the personal interests of the highest level leader may be the most influential factor interfering with information 28 Refer to Chapter 2 for an explanation of unions and their development in Bolivia.
185 dissemination. Some informants pointed to the example of the member organizations that are co opted by the government, manifesting how the politics of the national government limits the amount and type of information shared with communities. In this way, informants argued that the representativeness of these leaders. 29 Significant travel distances and limited public transportation characterizing the distribution and remoteness of communities in the northern Amazon region in addition to the associated time and costs required for tra vel also explain the limitations to information sharing in complicate traveling, as there are often no roads (only walking trails) and thus no forms of public transportation. The most common form of transportation is motorcycle, which requires gasoline and maintenance. Given the voluntary nature of leadership positions in so cial organizations, there is typically no funding with which to finance this travel. The distance and limited organizations who are also invited to attend meetin gs. 30 ANMIN A Similar to CORIDUP and BOCINAB, ANMIN no information with their organizations and local populations. When asked, m ost forum members agreed that the communities are not even aware of the role of the man agement forum. The 29 Considering the larger social and political context in Bolivia, especially in the context of environmental demands, the close relationship between the government and leaders of some social organizations has both weakened their connection with communities a s well as limited their representation of them. 30 Several representatives from every organization are sometimes invited to BOCINAB meetings, but costs accrued are not reimbursed by the supporting NGOs.
186 director of the area suggested that the leaders ultimately make decisions on behalf of the members they represent without informing or consulting the members. He suspected that the protected area management seem to result in a lack of recognition of the forum in the area. In my visits to the area, I learned that communities seem to be much more familiar with the park rangers who are often present i n community meetings and provide information concerning the protected area. Key informants suggested that local level knowledge of the forum reflects local interest in protected area management. Similar to BOCINAB, local interest seems to be related to the geographic distribution of the communities in the area, which in this case implies that communities benefit from protected area management. For example, in the highlands of the area, communities are more aware of protected area management because they man age vicua and have been involved in projects and activities focusing on vicua management. As the protected area was expanded to include Bautista Saavedra and the lower areas in 2000, it has included communities that have become much more focused on mini ng and coca production. Informants information regarding protected area management. In the case that issues are not considered relevant or of interest to the communities, several members indicated that they are given little Key informants noted that the information most widely disseminated (and of increasing interest) is that which pertains to mining activities in the area. I also learned that the permanence of several forum members in ANMIN A that no longer hold their local leadership positions tends to restrict the amount and quality of information
187 shared with their populati ons. These individuals seem to be driven to continue participating because of their interest in remaining informed rather than their interest in representing their ittle information sharing occurs between forum members and their populations. One informant refers Also similar to BOCINAB, significant travel distances and limited public transportation limit information sharing by l eaders as well as the participation of community members in forum meetings when held near their hometown. Because of the communities away from the major municipalities where forum meetings are held on a rotating basis the members further away from the municipalities likely know little about the management committee. Section Conclusion In view of the purpose of the forums to represent local populations in addressing local issues and those emerging at the national leve l there seems to be a notable disconnect between the forum and its populations. A variety of explanations for the lack of information sharing between forum members and their con stituents were discussed and among these explanations, the interests of leader s and that of their populations have the greatest effect on information flow. In addition to these factors, geographic distribution also seems to explain the difference between the recognition of CORIDUP and that of BOCINAB and ANMIN A. Moreover, the minim al efforts to engage local level stakeholders in information sharing may clearly affect the mobilization of local support for the f orums and the legitimization of them.
188 CHAPTER 10 CONCLUSIONS This study provides insights into emerging collaborative network ed arrangements involving civil society actors in participatory governance. This research focuses on the legitimacy of such governance arrangements by examining interactions across levels of governance, from the forum level to the national and local levels To operationalize legitimacy, I focus on the democratic quality of forums (internal legitimacy) and the recognition of the forum by external actors (external legitimacy). I use social network analysis to explore the structure of interactions at the meso level as well as the cross level linkages that I hypothesized may influence forum legitimacy. The quantitative and qualitative results presented in Chapter 8 and 9 provide answers to my three research questions. First, the forums have varying levels of leg itimacy based on forum legitimacy scores. Third, ethnographic findi ngs suggest that the most central network actors hold particular influence and power in controlling communication patterns and decision making, a dynamic emphasized by their relationship with supporting institutions, such as NGOs. Finally, ethnographic fin collaboration with NGOs and their lack of efforts made to engage local level stakeholders in information sharing threaten their external legitimacy. I conclude with a discussion of the implications of my findings for participatory governance in Bolivia. Findings from the social network analysis and ethnographic research indicate that the dynamics at the meso level and the cross level linkages to the macro and micro levels are int imately related and must be integrated for a more legitimate form of governance, particularly
189 because the most central actors play important roles in providing resources to forums. Findings suggest that the construction of forum legitimacy requires more th an just inclusiveness and equality within forums, as forums are nested within a broader socio political context. In line with theoretical propositions and emerging empirical inquiry by network governance scholars, my findings indicate the importance of con sidering the internal and external environment in examining the legitimacy of such governance arrangements. Internal Legitimacy cases, and that forum legitimacy is l argely reflected in network positions, particularly network centrality. There are several insightful observations related to these findings, however that came out of the ethnographic research. First, information, influence, and centrality interact to prod uce the forums as legitimate because they are more influential and their relationship with supporting institutions is both constitutive of their greater acc ess to information as well as an important feature of why they are leaders. First, I observed that less central actors perceive the forum as less legitimate in part because they do not have the same influence. In this way, issues of power inequality seem t most influential in spreading information, shaping the agenda, and controlling resources. This finding corroborates previous research indicating that the stratification of th e social system into different network locations is associated with an unequal distribution of status and access to social power (Baron and Pfeffer 1994; Thye 2000; Fiol et al. 2001). The underlying logic in the interpretation of power here is that these d ifferent social status and power conditions lead to differences in perceptions of the social system (Lamertz and Aquino 2004). Although all forums are dependent on supporting institutions due to their limited technical capacity required to
190 accomplish their goals, this dynamic has played out differently across cases. For example, the cases studies examined present variations in leadership experience and skills that have implications for the variation in forum legitimacy across cases. The other important obs ervation is that although the influence of these leaders may indicate tha level linkages that provide the support and resources necessary to the viability of forums. As such, relationship with external actors. These findings support previous research that found positive impacts of network characteristics that cross network boundaries on resource governanc e outcomes ( Ernstson et al. 2010; Newman and Da le 2007 ). Moreover, while it is important for leaders to have adequate experience and skills to fulfill the expectations of other actors outside the network, the democratic norms of inclusion and equal opportunity may have been comprised to get there. Evidence therefore supports th e proposition that networks face a serious tension between the need for both inclusive decision making and administrative efficiency (Provan and Kenis 2007). External Legitimacy F indings related to external legitimacy also reveal insightful observations E vidence principles of participatory democracy, meaning that all civil society actors theoretically have the opportunity to contribute to the development of public po licy, there are restrictions as to which voices from civil society are heard. Although supporting institutions, such as NGOs in the case of BOCINAB and CORIDUP, provide the forums the technical assistance required to influence public policy at the national level, this very collaboration is shunned by the government and
191 unresponsiveness to forum demands. essentially determine whether or not the forums will achieve their objectives through their incorporation into policy and tangible outcomes that are expected to benefit their rural constituencies. This dynamic produces feedback to affect the internal engag ement processes of the forums themselves, and may affect the trust, confidence, and commitment of members to the forums. I found that what some scholars refer to as intermediate outcomes or critical process outcomes tangible outputs are ess ential for building the momentum in the collaboration and encouraging a cycle of trust building and commitment ( Ansell and Gash 2008, 561; see also Cangen and Huxham 2003b; Rogers et al. 1993 ). re are several institutional environmental demands, such as a lack of economic and human reso urces, lack of technical competence and institutional instability at th e national level. The latter pertains to the change of authorities and public officials, which affects its ability to respond to specific environmental issues and demands. Findings also indicate the significance of the disconnect between the forums and th e populations they are to represent. A lack of information flow from the meso to the micro levels affects the mobilization of local support for the forums and the legitimization of them. While there are a variety of explanations for this dynamic, the lack of tangible outcomes felt and seen by local populations as a result of forum achievements helps explain the lack of motivation of
192 work. This disconnect and lack of r eyes of the government. The evidence presented supports the idea that cultivating internal and external legitimacy are among the most challenging aspects of network governance, as both internal a nd external legitimacy affect the participation of members, which is crucial to the legitimacy of forums. On the one hand, forums must respond to the needs of its members, however, it is also important that outside actors see the network as an entity in it s own right and not simply a group of organizations that occasionally get together to discuss common concerns. Moreover, the cross level linkages prove invaluable to sustaining forum legitimacy. Conclusions With the rise of participatory governance, a wid e range of civil society actors is increasingly in the position to influence the policy process at a number of political levels due to the numerous access points into the institutional process. It is expected that the creation of legitimate spaces for mul ti stakeholder involvement in policy development provides greater opportunities for participation in the policy process while increasing the quality of information ev idence on deliberative public participation and may provide insights regarding the c hallenges involved in participa tory processes that integrate openness, inclusivity and influence (Barnes et al. 2007). public sphere to include previously underrepresented populations in decision making have created spaces and opportunities for citizen engagement, these developments have been subordinated by other policy imperatives that have ultimately shaped the politic al opportunities of civil society actors in participating in public policy development (Newman et al. 2004). The contradiction lies in the political context, in
193 which the government demands the technical capacity of civil society actors to influence policy yet restricts the involvement of NGOs that have traditionally played an instrumental role in providing this assistance to social organizations. Such political preferences ultimately affect the spaces and opportunities available for participatory governanc e. Moreover, this research reveals the tension between a focus on building internal network interactions versus building the credibility of the network to outsiders. The tension arises when the internal legitimacy needs of network members conflict with the dem ands and expectations of outside actors. Today, all of the forums struggle to sustain themselves and their legitimacy, which I attribute to factors that work to affect both their internal and external legitimacy. Among these factors is the importance of leadership roles in well as the recognition of the forums by outside actors. It is clear that the role of NGOs has also been critical in sustaining the forums, however, if NGOs are expected to serve the role of interlocutors between forums and the national level, they may do more harm than good Furthermore, this research illustrates how the external context creates opportunities and constraints and influences the general parameters within which collaborative governa nce unfolds (Emer son et al. 2012 ). I found this to be the case in my research where the political interests and limited capacity of the government affect the responsiveness of the government to forum demands, limiting the possibilities for the forum to achieve its goals. I n turn, these constraints Moreover, this context has created institutional constraints and limits the capacity of these initiatives to shape policy and practice f rom below, which then affects the forums themselves as members lose motivation without concrete results and faith in its leadership to attain results. In this way, while it is necessary to build internal legitimacy through fair decision making
194 processes, d eveloping and enhancing external legitimacy can also result in reinforcing the commitment of network participants, who are more likely to see themselves as part of a viable network. Contributions P articipatory decision making processes have gain ed considerable attention due to the expectation that they produce higher quality and long lasting solutions to environmental problems affecting multiple actors and multiple scales (Reed 2008). T his research has practical applications and theoretical signific ance for those seeking to understand the increasingly multifaceted relationships between civil society and government actors involved in constructing public policy especially in a context characterized by poverty and natural resource depletion Findings c ontribute to our knowledge regarding multi level interactions and legitimate forms of governance, revealing the opportunities and constraints affecting participatory governance. Findings suggest that efforts to engage actors acros s levels in decision makin g may have important implications for the legitimization and mobilization of local support for new policy goals. The lack of information flow and resource equality as well as the institutional and organizational weaknesses at different levels of governance challenge s the purported outcomes of decentralization reforms Moreover, I have also demonstrated the significance of the broader socio political context in affecting the opportunities and constraints of civil society actors in influencing public policy. By operationalizing interactions with social network variables at the meso level and cross level interactions to the micro and m a cro levels, this research also contribute s to the larger question of how interaction across levels or organizational borders i mpacts information sharing and coordination through particular individuals located at different levels. Social network analysis serves as a powerful methodological tool to analyze the relationships between group
195 structures, processes, and the corresponding mechanisms that influence locally legitimate regulatory systems. The integration of the network analytic approach also contributes to our understanding of the collaborative networked arrangements formed by and around grassroots actors for environmental ad vocacy. was helpful because it provides implications as to the viability of networks that depend on building legitimacy both internally and externally. In addition, the pr ocedural approach to legitimacy used to measure internal legitimacy contributes to our understanding of the importance of democratic norms in the collaborative governance process. I have provided internal and external legitimacy, suggesting the tensions between inclusiveness and what some scholars refer to as efficiency and effectiveness. Limitations The scale used to measure internal legitimacy in this research presented several challenges due to the way the instrument was designed and administered. First, although I intended to include both positive and negative statements, results from the pre test indicated that only positive items were adequately reliable. The exclusion of negative statements may affect the l range of perceptions tested. However, pre testing did indicate that the dimensions used to construct the scale were, in fact, much more similar to each other and thus produce a much more reliab le measure of legitimacy when considered altogether. Next, I found that the amount of time I had for administering the legitimacy scale affected the variability of the responses and thus the final legitimacy scores. Given the limited availability of most forum members, I frequently had no other option than to interview members during my first meeting with them, allowing little time to establish a trusting relationship. As a result, I found
196 that the responses to the legitimacy scale failed to capture the full ran ge of perceptions identified in the follow up interviews. Similar circumstances may have also affected the social network analysis, as it was administered at the same time as the legitimacy scale. Finally, the fact that I was unable to locate several of th e forum members who no longer participate in forums may have further limited the variations in perceptions captured. Another limitation characteristic of social research is that administering surveys and interviews at one point in time fails to capture c hanges in perspectives over time. In this way, longitudinal studies can address this weakness (Concley and Moote 2003). Characteristic of governance processes, I found the relationship between the study variables iterative or cyclical as they are important across all the stages of the collaboration. As such, the collaborative governance process is difficult to represent because of its non linear character (Ansell and Gash 2008 ). Similarly, examining social network features at one point in time is also limited as this may not reveal the complex, non static nature of social interactions. Some authors contend that measuring network features at one point in time may limit our understanding of the conditions and mechanisms that allow networks to function in certain ways i n the context of natural resource governance (Bodin and Crona 2009), while other authors argue otherwise (Ramirez Sanchez 2011). Comparing relational data collected for the same system at different points in time could, however, potentially reveal differen ces that and not a result of social processes leading to network change ( Bodin and Prell 2011). Future Research More research is needed regarding the process of building network legitimacy. Based on the understan ding that legitimacy management is an ongoing process, a longitudinal examination of internal and external legitimacy would serve to illuminate the intricacies of this process.
197 Future research should also continue to develop measures of both internal and e xternal legitimacy to provide both quantitative and qualitative data to enrich our understanding of network legitimacy. Such research would also contribute to our knowledge of legitimacy by as their identification with it (see Human and Provan 2000; Raab and Kenis 2009). There is also a need for research to focus on the accountability relations among multi level actors in exploring the mechanisms that enable and inhibit information flow. Con sidering the claimed effects of participation, such as greater access to decision makers and higher levels of participation by various social groups in decision making (Abelson et al. 2007), a focus on the accountability of decision makers would expand our understanding of multi level governance. Future research should examine how spaces opened up by participatory processes have or have not been able to use their agency to demand transparency, accountability, and responsiveness from government institutions Cornwall and Gaventa 2001). To deal with the complexity of scale and the involvement of multiple stakeholders in such governance arrangements, one could use can influence policy through democratic institutions. These research inquiries may offer implications of such articulation for the degree to which civil society actors can harness the accountability of national institutions. Considering the growth of par ticipatory initiatives involving different actors both civil society and state to address environmental issues, there is a need to further examine issues of inequality that influence the legitimacy of a participatory process. As evidenced in this research, issues of power inequality may have implications for the leaders are most influential in spreading information, making decisions, and controlling
198 resources. Scholars suggest that participatory processes succeed only when there is a relative balance of power among the stakeholders (Fung and Wright 2001; Baiocchi 2005). As such, future research should also consider the significance of network centrality with respect to issues of power and inequality in governance networks. S uch research would contribute to literature on collaborative governance, as little empirical research in the field has explicitly addressed issues surrounding power inequalities (Choi and Robertson 2013).
199 A PPENDIX A IRB NOTICE S
201 APPENDIX B INS TRUMENTATION Social Network Survey Nombre: ____________________________ Organizacin: ________________________ I. Informacin Personal 1. Sexo: M F 2. Edad: _____ 3. Nivel de Educaci n: 1. primario 2. secundario 3. Universidad II. Afiliac i n 4. A qu organizacin representa Ud? _______________________ 5. Cuanto tiempo (meses/anos) est involucrada su organizacin en el COMITE DE GESTION ? ________ y Ud? _____ __ III. Al respecto a los relaciones entre actores en la plataforma: 1) En una escala de 1 5 (1 =no a menudo a 5 =m uy a menudo), c on qu frecuencia Ud. se relaciona con los o tros miembros del foro? Miembros (Nombre y Apellido) 1 No a menudo 2 poc as veces 3 regular 4 constante 5 Muy a menudo 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16 17.
202 17. 19. 20. IV. Al respecto a los relaciones con actores fuera de la plataforma: 1) Nivel Micro a) Ud. representa ciertos intereses en el foro Nombre 25 personas de su organizacin que Ud. representa como miembro del foro. b) En una e scala de 1 3 (1=dbil a 3=fuerte) por favor indique que tan fuerte siente cada persona qu e Ud. representa sus intereses. Constituyentes (Nombre y Apellido) Dbil Regular Fuerte 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16 17. 17. 19. 20. 21. 22. 23. 24. 25.
203 2) Nivel Macro a) En una escala de 1 5 (0=nada a 5=muy a menudo) por favor indique, c on qu frecuencia Usted (no como foro ) se reuni o coordin con cada institucin e organizacin en los ltimos dos aos ? Institucin/organizacin 0 Nada 1 No a menudo 2 pocas veces 3 regular 4 constante 5 Muy a menudo 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. (b) En una escala de 1 5 (1=nada satisfecho(a) a 5=muy satisfecho(a)), por favor indique su nivel de satisfaccin con la r espuesta de la institucin a las preocupaciones del foro. Institucione s/Organizaciones 1 Nada Satisfecho(a) 2 poco 3 regular 4 satisfecho 5 Muy Satisfecho(a) 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15.
204 Legitimacy Scale En una escala de 1 5 (1=muy desacuerdo a 5=muy de acuerdo), indique su nivel de acuerdo con cada frase. 1 Muy Desacuerdo 2 3 4 5 Muy de Acuerdo Esta organizacin incluya todas las organizaciones que deben ser representa das. Las normas de la organizacin estn claramente establecidas para garantizar el respeto mutuo entre los participantes. Ud. se siente cmodo(a) expresando sus puntos de vista en las reuniones. Ud. es consciente de sus derechos y respon sabilidades como miembro de esta organizacin. Todos los participantes tienen la oportunidad de hablar abiertamente. La misma gente siempre participa en la elaboracin de las agendas. Todos los miembros tienen acceso a la informacin que necesitan. Le dan suficiente tiempo y espacio para expresar sus opiniones en las reuniones. La informacin presentada durante las reuniones es clara. Ud. entiende cmo se toman decisiones en las reuniones. Hay tiempo suficiente para discutir todos los temas pertinentes en las reuniones. Ud. es consciente de las reglas del debate. Las discusiones siempre son dominadas por las
205 mismas personas. Ud. tiene la oportunidad de sugerir temas para la agenda de las reuniones. Si Ud. tiene una pregunta, Ud. siente cmodo(a) al preguntar. Todos los participantes tienen la misma oportunidad de influir en las decisiones. La informacin que recibo me ayuda a tomar decisiones. Si Ud. no est de acuerdo con al guna decisin, se siente cmodo(a) dar su opinin. Si Ud. necesita ms informacin sobre un tema, Ud. sabe cmo conseguirla. Los debates permiten compartir diferentes opiniones. Ud. recibe la informacin necesaria para contribuir de maner a efectiva en la toma de decisiones. Esta organizacin es accesible a diferentes grupos organizados.
206 Key Informant Interview Nombre: ____________________________ Ubicacin:___________________________ Cargo: ______________________________ IMPETU / PROPOSITO: 1. Que es lo que ha motivado la formacin del COMITE DE GESTION ? En qu ao se form el COMITE DE GESTION ? a. Como fue la visin del COMITE DE GESTION en el inicio? (ie.resolver conflictos de inters?) b. Cuales temas disputaron ms? (L istar) 2. Hubieron actores especficos (entidades, individuales) en la formacin/creacin del COMITE DE GESTION ? 3. Cules son los objetivos del COMITE DE GESTION hoy? a. Que tan bien logra sus objetivos? Que contribuye a eso? b. Qu clase de temas ne gocian en las reuniones/sus actividades? (regulaciones hdricas, justicia comunitaria, usos y costumbres?) 4. Cul es la intencin del COMITE DE GESTION ? afectar la poltica (en el nivel local o nacional) o crear un espacio para discutir entre actores? a. Usted podra proveer algunos ejemplos de las metas y avances del COMITE DE GESTION en los ltimos dos aos? 5. Como ha cambiado el COMITE DE GESTION atreves de los aos? ACTORES: 6. Han cambiado los actores involucrados en el COMITE DE GESTION de sde el inicio? Quien participa ahora? Porque ellos? Como estn elegidos? (Probe: ONGs, municipio, concesionarios, etc y su rol) b. Cada cuanto tiempo cambian los participantes? c. Cuales actores estn encargados de organizar y/o facilitar las reuniones/ actividades del COMITE DE GESTION ? Porque ellos? Quien controla el financiamiento? REUNIONES: TOMA de DECISIONES 7. Como preparan el agenda para el COMITE DE GESTION ? Los participantes se pueden cambiarla/revisarla antes de la reunin? 8. Cuales regla s establecen para guiar la toma de decisin? Hay un estatuto para controlar el comportamiento de participantes en la reunin? (Son reglas democrticas? Ejemplos) a. Los participantes deben respetarlas? Que pasa si no lo respetan (ejemplos)? b. Como se to man las decisiones? (por voto: mayora o acuerdo comn/consenso) Quienes votan? INTERESES: 9. Como varan los intereses de los actores? Como influyen los intereses particulares en la toma de decisiones? Hay algunos intereses que dominan en el COMITE D E GESTION ? Cuales y como?
207 a. Que es lo que pasa si hay intereses y puntos de vista diferentes que dificultan llegar a un consenso? (Hay ejemplos de eso?) b. Hay un sistema para controlar los conflictos de intereses en las reuniones? c. Existen mecanismos para asegurar que los opiniones y intereses de los actores son escuchados? INFORMACION & LA RENDICION DE CUENTAS: 10. El COMITE DE GESTION est reconocida por las comunidades, los municipio, gobernacin, ONGs? Porque/Como? Que tanto? a. Compartan la informacin discutida o decisiones con otros actores que no estn incluidos en el COMITE DE GESTION? Cmo ? Cual tipo de informacin comparten? b. Quien(es) tiene la responsabilidad de llevar la informacin a estos actores? c. Es un requisito que los part icipantes comparten la informacin discutido con los que representa? 11. Fortalezas y debilidades, oportunidades, desafos: a. Cules son las fuerzas del COMITE DE GESTION? b. Cules son las debilidades del COMITE DE GESTION ? c. Cules son los desafos que les han presentado? d. Que es lo que han hecho/estn haciendo para cambiar la situacin? 12. En su opinin, que requiere el proceso democrtico (en reuniones, la toma de decisiones)?
208 Direct Observation Guide f or Forum M eetings Conceptual dimension Questions to consider Stakeholder representation 1. How are members notified of meetings? 2. Who shows up to meetings? 3. Were there missing members at meetings? Why did they not show up? Equal opportunity Transp arency 1. Dominant Voices: Are there people who dominate? Who talks most? 2. Excluded Voices: Are there excluded individuals? Who talks less? 3. Is there sufficient time so that everyone has the opportunity to contribute to the discussion? 4. Does every one have the same opportunity to challenge the information presented? 5. Does everyone have the opportunity to contribute in agenda setting? In decision making? 6. Who has the most influence within the forum? 1. The information presented seems clear and w ell explained? something or if they want to express their opinion? 3. What mechanisms are used so that everyone is heard? 4. How are suggestions from members considered in the discussions? 5. Are there rules to regulate discussion? Who controls the rules? 6. How are decisions made (by consensus, majority)?
209 APPENDIX C MACRO INSTITUTIONS USED IN DATA COLLECTION 1) Macro Institutions, CORIDUP Institution or Organization Classification Ministerio (y Vice) del Medioambiente Government national Minist erio de Mineria Government national Gobernacion Government departmental AM DEOR ( Asociacin de Municipios de Oruro ) Government municipal Univer sidad Tecnico de Oruro Public University FEDJUVE ( Federacin Departamental de Juntas Vecinales de Oruro) Social organization Regantes de Challapata Social organization Sullos Soras Social organization LIDEMA ( Liga de Defensa del Medio Ambiente) NGO U MAVIDA NGO Project NINA NGO INTERSOL NGO CEPA NGO 2) Macro Institutions, BOCINAB Institution or Organization Classification Ministerio (y Vice) del Medioambiente Government national Minis terio de Trabajo Government national INRA (National) Gov ernment national INRA (Departmental) Government departmental CIDOB Social organization Derechos Humanos NGO UNITAS NGO CEJIS NGO CIPCA Pando NGO CIPCA Norte (Beni) NGO IPHAE NGO 3) Macro Institutions, ANMIN A Institution or Organization Class ification Ministerio (y Vice) del Medioambiente Government Ministry of Rural Development Government Departmental Government Government CONAMAQ Social o rganization CSUTCB Social o rganization Tupac Katari Social Organization
210 Bartolina Sisa Social O rganization Via Bolivia NGO Fundisnap NGO Cooperacion Espanola NGO Mapza NGO WCS NGO
211 LIST OF REFERENCES Abels, G. 2007. Citizen involvement in public policy making: Does it improve democratic legi timacy and accountability? The c ase of pTA. Interdis ciplinary Information Sciences 13(1): 103 116. Abelson, J., P. G. Forest, J. Eyles, P. Smith, E. Martin, and F.P. Gauvin 2003. Deliberatio ns about deliberative methods: I ssues in the design and evaluation of public participation processes. Social Science a nd Medicine 57: 239 251. Abelson, J., P. G. Forest, J. Eyles A. Casebeer, E. Martin, and G. Mackean 2007. Examining the role of context in the implementation of a deliberative public participation experiment: Results from a Canadian comparative study. So cial Science & Medicine 64 (10): 2115 2128. Abrahamson, E. and L. Rosenkopf. 1997. Social network effects on the extent of innovation diffusion: A computer simulation. Organization S cience 8 (3): 289 309. Adcock, R. and Collier, D. 2001. Measurement validity : A shared standard for qualitative and quantitative research. American Political Science Review 95(3): 529 546. Agranoff, R. and M. McGuire 2003. Inside the matrix: Integrating the paradigms of intergovernmental and network management. International Jou rnal of Public Administration 26 (12): 1401 1422. Agrawal, A. and M. C. Lemos 2007. A gree ner revolution in the making?: E nvironmental governance in the 21st century. Environment: Science and Policy for Sustainable Development 49 (5): 36 45. Agrawal A. and J. C. Ribot. 1999. Accountability in dece ntralization: A framework with South Asian and West A frican cases Journal of Developing Areas 33: 473 502. Albo, X. 1994. And from Kataristas to MNRistas? The surprising and bold alliance between Aymaras and neolibe rals in Bolivia. In Indigenous Peoples and Democracy in Latin America, ed. D.L. Van Cott, 55 81. New York: St Martin's Press. Andersson, K. and E. Ostrom 2008. Analyzing decentralized resource regimes from a polycentric perspective. Policy Science 41:71 93. Ansell, C. and A. Gash 2008. Collaborative governance in theory and practice. Journal of Public Administration Research and Theory 18:543 71. Arellano Lpez, F. S. 2012. Conflicting land use agendas: environment, indigenous land rights and developmen t in central Bolivia. The case of the Isiboro Scure Indigenous Territory and National Park (TIPNIS) Land Deal Politics Initiative. http://www.cornell landproject.org/download/landgrab2012papers/arellano.pdf (accessed 30 March 2013) Ary, D., J. C. Jacobs and A. Razavieh. 2002. Introduction to research in education 6th ed. CA:
212 Wadsworth Group, Thompson Learning, Inc. Baiocchi, G. 2005. Militants and citizens: T he politics of participatory democracy in Porto Alegre Stanford, CA: Stanford University Press Barnes, M J. Ne wman, and H. Sullivan 2006. Discursive arenas: Deliberation and the c onstitution of identity in public participation at local level s Social Movement Studies 5(3): 193 207. Ba ron, J. N. and J. Pfeffer. 1994 The social psychology of organizations and inequality. Social Psychology Quarterly 57(3) : 190 209. Beaumont, J. and W. Nicholls 2008. Plural Governance, participation and democracy in c ities. International Journal of Urban and Regional Research 32(1): 87 94. Beetham, D. 1991. Max Weber and the legitimacy of the modern State. Analyse & Kritik 13 (1) : 34 45. Benhabib, S. 19 96. Toward a deliberative m o del of democratic l egitimacy. I n Democracy and d ifference: Contesting boundaries of the p olitical ed. S. Benhabib, 67 94. Princeton, NJ: Princeton University Press. Berkes, F. 2009. Evolution of co management; Role of knowledge generation, bridging organizations and social learning. Journal of Environmental Management 90: 1692 1702. Bingham, L. B. and R Bi g ideas in collaborative public management Armonk, NY: M.E. Sharpe. rlson. 2008. Frameshifting: Lateral thinking for collaborative public management. In Big ideas in collaborative public management, eds. 16. Armonk, NY: M. E. Sharpe. Blair, H. 2008. Innovations in participatory local govern ance in participatory governance and the millennium development goals (MDGs) Publication based on the Expert Group Meeting on Engaged Governance: Citizen participation in the implementation of the developmental goals including the millennium development g oals (MDGs) New York: United Nations. Bodin, ., B. Crona, and H. Ernstson. 2006. Social networks in natural resource management: What is there to learn from a structural perspective? Ecology and Society 11 (2): 2. Bodin, and B. Crona 2009. The role of social networks in natural resource governance: What relational patterns make a difference? Global Environmental Change 19: 366 374. Bohman, J 1996. Public deliberation: Pluralism, complexity, and d emocracy. Cambridge, MA: MIT Press.
213 2004. Realizing d eliberative democracy as a mode of inquiry: Pragmatism, social facts and normative theory. Journal of Speculative Philosophy 18(1): 23 43. Borgatti, S. P., M. G. Everett and L. C. Freeman 2002. UCINET 6 for Windows. Harvard: Analytic Technologies Brass, D. J. and M. E. Burkhardt.1993. Potential power and power use: An investigation of structure and behavior. Academy of Management Journal 36(3): 441 470. Brondizio, E., E. Ostrom, and O. Young 2009. Connectivity and the governance of multilevel social ecolo gical systems: The role of social capital. The Annual Review of Environmental Resources 34:253 278. Bryson, J. M., B. C. Crosby, and M.M. Stone 2006. The design and implementation of cross sector collaborations : Propositions from the literature Public Ad ministration Review 66 : 44 55. Burt, R 1992. Structural holes: The social structure of c ompetition. Cambridge, MA: Harvard University Press. 2004. Structural holes and good ideas Ame rican Journal of S ociology 110 (2): 349 399. Canessa, A. 2012 New indi genous citizenship in Bolivia: Challenging t he liberal model of the state and its subjec ts. Latin American and Caribbean Ethnic Studies 7 (2): 201 221. Carpini, M. D., F. L. Cook, and L. R. Jacobs. 2004. Public deliberation, discursive participation, and ci tizen engagement: A review of empirical literature. Annual Review of Political Sci ence 7:315 44. Cash, F. W. Adger, and F. Berkes. 2006. Scale and cross scale dynamics: Governance and information in a multilevel world. Ecology and Society 11:8. CEPA/LIDEMA /UTO. 2010. Anlisis ambiental en la transferencia de arsnico, cadmio, estao y zinc, del agua a suelo y especies vegetales de consumo humano y animal, a lo largo del canal de riego Sora Sora. (cuenca Poopo) Oruro, Bolivia. Chambers, S. 2003. Deliberativ e democratic theory. Annual Review of Political Science 6: 307 326. Chaplin, A. 2010. Social movements in Bolivia: F rom strength to power. C ommunity Development Journal 45(3): 346 355. Chhatre, C. 2008 Political articulation and accountability in decentral ization: Theory and evidence from India. Conservation and Society 6(1): 12 23. Choi, T. and P. J. Robertson 2013. Deliberation and decision in collaborative governance : A simulation of approaches to mitigate power imbalance Journal of Public Administrat ion Research and Theory 23 (1): 1 24.
214 Cohen, J. 1996. Procedure and subst ance in deliberative democracy. In Democracy and Difference ., ed. S. Benhabib, 103 32. Princeton, NJ: Princeton University Press. Colton, D. and R. W. Covert. 2007. Designing and const ructing instruments for social research and evaluation Jossey Bass. Connelly, S., T. Richardson, and T. Miles 2006. Situated legitimacy: Deliberative arenas and the new rural governance. Journal of Rural Studies 22 (3): 267 277. Connelly, S. 2009. Deli beration in the face of power: S takeholder planning in Egypt. Policy and Society 28 (3): 185 195. 2011 Constructing legitimacy in the new community governance. Urban Studies 48 (5): 929 946. Conservation Strategy Fund. Cost benefit analysis of the Cachuela Esperanza hydroelectic project. http://conservation strategy.org/en /project/cost benefit analysis cachuela esperanza hydroelectric project (accessed on 2 November 2012). Cornwall, A. and V. S. Coelho. 2007. Spaces for change?: T he politics of citizen participation in new democratic arenas London: Zed Books. Cornwall, A and J. Gaventa. 2001. Bridging the gap: C itizenship, participation and accountability. PLA notes 40 : 32 35. Bodin, ., B. Crona, and H. Ernstson. 2006. Social networks in natural resource management: d e Jong, W., S. Ruiz, and M. Becker. 2006. Conflicts and communal forest management in northern Bolivia. Forest Policy and Economics 8: 447 457. d e Jong, W. and S. A. Ruiz 2012 Strangers among trees: Territoriali z ation and forest policies in the northern Bolivian Amazon. Forest Policy and Economics 16 : 65 7 0. DeV ellis, R. F. 1991. Scale development: Theory and application Applied Social Research Methods Series. Newbury Park, CA: Sage Publications. De Vaus, D. A. 1996. Surveys in social r esearch 4th ed., London: UCL Press. Edmunds, D. and E. Wollenberg. 20 03. Local forest management: The impacts of devolution policies. London: Earthscan. Emerson, K P J. Orr, D L. Keyes, and K M. McKnight. 2009. Environmental conflict resolution: Evaluating performance outcomes and contributing factors. Conflict Resolut ion Quarterly 27:27 64. Emerson, K., and P. Murchie. 2010. Collaborative governance and climate change: Opportunities for public administration. In The future of public administration, public management, and
215 public service around the world: The Minnowbrook perspective Slyke, and S. Kim, 141 53. Washington, DC: Georgetown Univ. Emerson, K., T. Nabatchi, and S. Balogh. 2012. An integrative framework for collaborative governance. Journal of Public Administration Research and Theory 22 ( 1): 1 29. Emirbayer, M. and J. Goodwin. 1994. Network analysis, culture, and the problem of agency. American Journal of S ociology 99(6): 1411 1454. Ernstson, H., S. Sorlin, and T. Elmquist. 2008. Social mo vements and ecosystem services T he role of social ne twork structure in protecting and managing urban green areas in Stockholm. Ecology and Society 13 (2): 39. Ernstson, H., S. Barthel E. Andersson, and S. T. Borgstrm. 2010. Scale crossing brokers and network governance of urban ecosystem services: The case of Stockholm. Ecology and Society 15 (4): 28. Ferroukhi, L. 2003. La gestin fores tal municipal en Amrica Latina. Bogor, Indonesia: Center for International Forestry Research, International Development Research Center. Fischer, F. 2006. Participatory gover nance as deliberative empowerment the cultural politics of discursive space American Review of Public Administration 36 (1) : 19 40. Freeman, L.; D. White, and K.A.Romney 1992. Research methods in social network analysis New Brunswick, New Jersey, Transact ion Publishers. Friedkin, N. E. 1993 Structural bases of interpersonal influence in groups: A longitudinal case study. American Sociological Review 58(6): 861 872. Fung, A. and E. O. Wright. 2001. Deepening democracy: Institutional innovations in empower ed participatory governance. New York, NY: Verso. 2003. Thinking about empowered participatory g overnance. In Deepening democracy: Institutional i nnovat ions in empowered p articipatory g overnance, ed s A Fung and E O. Wright, 3 42. New York, NY: Verso Fung, A. 2006. Varieties of participation in complex governance. Public A dministration R eview 66 (s1): 66 75. Gastil J. and Levine P. 2005. The deliberative democracy handbook San Francisco: Jossey Bass. Gaventa, J. 2004. Towards participatory governance: Assessing the tr ansformative possibilities. In From tyranny to t ransformation ed s S. Hickey and G. Mohan, 25 41 London: Zed Books Goetz, A. M., and J. Gaventa, J. 2001. From consultation to influence : Bringing citizen voice and client focus into serv ice d elivery IDS Working Paper 138, Brighton: IDS.
216 Goodale, M 2006. Reclaiming modernity: I ndigenous cosmopolitanism and the coming of the second revolution in Bolivia. American Ethnologist 33: 634 649. Goodwin, M. 1998. The governance of rural areas: So me emerging research issues and agendas. Journal of Rural Studies 14:5 12. Granovetter, M. 1973. The strength of weak ties. American Journal of Sociology 78(6) : 1360 1380. Granovetter, M. 1985. Economi c action and social structure: T he problem of embeddedne ss. American Journal of S ociology 91(3): 481 510. Habermas, J. 1996. Between facts and n orms: Contributions to a d iscou rse theory of law and d emocracy Cambridge, MA: MIT Press. Hajer, M. and H. Wagenaar, eds 2003. Deliberative policy analysis: Understand ing governance in the network s ociety Cambridge: Cambridge University Press. Hanneman, R. A. and M. Riddle. 2005. Introduction to social network m ethods University of California, Riverside. Published in digital form at http://faculty.ucr.edu/ hanneman/ Hatcher, L. 1994. A step by step approach to using the SAS system for factor analysis and structural equation modeling SAS press. Hendriks, C. M. 2002 Institutions of deliberative democratic processes and interest groups : Roles, tensions and incentives. Australian Journal of Public Administration 61(1): 64 75. democracy. Political Studies 54(3): 486 508. 2008. On inclusion and network governance: The democratic disc onnect of Dutch energy transitions. Public Administration 86(4): 1009 1031. Hritier, A. 1999. Elements of demo cratic legitimation in Europe: A n alternative perspective. Journal of European Public P olicy 6 (2): 269 282. Hirst, J. 2000. Democracy and governanc e. In Debating governanc e, ed. J Pierre 13 35 Cambridge, UK: Oxford University Press. Hooghe, L. and G. Marks. 2003. Unraveling the central stat e but how? Types of multi level governance. Ameri can Political Science Review 97: 233 243. Human, S. E. and K. G. Provan 2000. Legitimacy building in the evolution of small firm multilateral networks: A comparative study of success and demise. Administrative Science Quarterly 45 (2): 327 365. Huxham, C. and S. Vangen. 2005. Managing to Collaborate London: Routl edge.
217 Ibarra H. 1993. Power, social influence, and sense making: Effects of network centrality and proximity on employee perceptions. Administration Science Quarterly 28(2): 277 303. Instituto Nacio n al de Estadstica (INE). 2001. Censo de Poblacin y Vivie nda 2001. La Paz, Bolivia: INE. .2013. http://www.ine.gob.bo/pdf/boletin/NP_2013_2.pdf ( accessed 20 March 2013 ) INRA (Instituto Nacional de Reforma Agraria). 2010a. Breve historia del reparto de tierras en Bolivia. LaPaz: INRA. 2010b. La tierra vue lve a manos indi genas y campesin as.resultados de la reconduccio n comunitaria de la reforma agraria en Bolivi a. La Paz: INRA. Keast, R., M. P. Mandell K. Brown and G. Woolcock 2004. Network structures: Working differently and changing expectations. Publi c Administration Review 64 (3): 363 371. Khol, B. 2003. Democratizing decentralization in Bolivia: The law of popular participation. Journal of Plan ning Education and Research 23: 153. Kickert, W J. M., E. K. Klijn, and J F. M. Koppenjan. 1997. Managing com plex n etworks : Strategies for the public sector London: Sage Publications. Killeen, T. J., A. Guerra, M. Calzada, L. Correa, V. Cal deron, L. Soria, and M. K. Steininger. 2008 Total historical land use change in eastern Bolivia: W ho, where, when, and how much. Ecology and Society 13 (1): 36. Kooiman, J. 1993. Social political governance: Introduction. In Modern governance: New government s ociety i nteractions ed. J. Kooiman, 1 6. London: Sage 2000. Societal governance: Levels, models and orders o f socia l political interaction. In Debating Governance ed. J. Pierre, 138 166. New York: Oxford University Press Inc Kuckartz U. 2001. MAXqda qualitative data analysis introduction Berlin: Verbi Software. Consult. Sozialforschung. Laing, A. F. 2012. Beyond th ost ne heory in Latin America: The politics of anti colonial struggles in Bolivia. Antipode 44 (4): 1051 1054. Lamertz, K. and K. Aquino. 2004 Social power, social status and perceptual similarity of workplace victimization: A so cial network analysis of stratification. Human Relations 57 (7): 795 822. Larson, A. M. and J. Ribot. 2004. Democratic decentralization through a natural resource lens. European Journal of Development Research 16 (1): 1 25 Larson, A. M., and F. Soto 2008. Decentralization and natural resource governance regimes. Annual Review of Environment and Resources 33:213 239.
218 Larson A. M., P. Pacheco, F. Toni, and M. Vallejo. 2007. The effects of decentralization on access to livelihoods assets. Journal of Environmen tal Development 16:251 68. Laserna, R., and A.G. Linera. 2009. Poder y cambio en Bolivia: 2003 2007. Embajada del Reino de los Pases Bajos. Leach, M., R. Mearns and I. Scoones. 1999. Environmental entitlements. World Development 27(2) : 225 247. Llanque, V 2004. Captulo 9. Extraccin forestal, procesos migratorios y deforestacin en el norte amaznico boliviano. In Retos y perspectivas del nuevo rgimen forestal en el norte amaznico boliviano W.de Jong. Centro Internacional para la Investigacin Foresta l. Llanque, V 2006. Forest product extractions, migr ation and deforestation in the N orthern Bolivian Amazon. In The social ecology of tropical forests : Migration, populations and frontiers eds. W.d e Jong, L. Tuck Po and A. Ken ichi. Kyoto University Pre ss, Trans Pacific Press Lupien, P. 2011. The incorporation of indigenous concepts of plurinationality into the new constitutions of Ecuador and Bolivia. Democratization 18(3): 774 796. Mandell, M. P. and R. Keast 2008. Evaluating the effectiveness of inte rorganizational relations through networks: Developing a framework for revised performance measures. Public Management Review 10 (6): 715 731. Marsh, C. 1982. The survey method: The contribution of surveys to sociological explanation London: Allen & Unwin. Martin, G. P. 2012. Public deliberation in action: E motion, inclusion and exclusion in participatory decision making. Critical Social Policy 32 ( 2): 163 183. Mrquez, M. C. 2006. Historia de una crisis: la minera en Oruro a fines del periodo colonial La Pa z: IEB/IFEA/ASDI. Massey, J. E. 2001. Managing organizational legitimacy: Communication strategies for organizations in crisis. Journal of Business Communication 38 (2): 153 182. Mayorga, F. 2007 Encrucijadas: Ensayos sobre democracia y reforma estatal en B olivia Editorial Gente Comn. Mayntz, R. 2006. Governance Theory als forten twickelte Steuerungstheorie? In Governance Forschung: Vergewisserung uÂ¨ b er Stand undEntwicklungslinien, ed. G.F. Schuppert, 11 20. Baden Baden, Germany: Nom os Verlag. McGinnis, M D., ed. 1999a Polycentric g overnance and development. Readings from the workshop in political theory and policy a nalysis In Institutional a nalysis ed s M.D. McGinnis and E. Ostrom. Ann Arbor: University of Michigan Press.
219 Meadowcroft, J. 2004. Delibe rative democracy. In Environmental governance reconsidered: Challenges, choices, and o pportunities, ed s R F. Durant, D J. Fiorino and R Cambridge, MA: MIT Press. Med eiros, C. 2001. Civilizing the p opular? The l aw of popular p artici pation and the design of a new c ivil s ociety in 1990s Bolivia. Critique of Anthropology 21(4): 401 426. Mendelberg, T. 2002. The deliberative citizen: Theory and evidence. Political Decision Making, Deliberation and P articipation 6 (1): 151 193. Mollo, N. 2009. Nueve a os del derrame de petrleo en el Desaguadero. Oruro, Bolivia: CEPA/LIDEMA Nabatchi, T. 2010. Addressing the citizenship and democratic deficits: The potential of deliberative democracy for public administration. The American Review of Public Administratio n 40 (4): 376 399. Net emeyer, R. G., W. O. Bearden, and S. Sharma 2003 Scaling procedures: Issues and applications SAGE Publications, Incorporated. Newman, J., M. Barnes H. Sullivan and A. Knops 2004. Public participation and collaborative governance. Journal of Social Policy 33 (2): 203 223. Newman, L. and A. Dale. 2007. Homophily and agency: C reating effective sustainable development networks. Environment, Development and Sustainability 9 (1): 79 90. Nygren, A. 2005. Community based forest management wit hin the context of institutional decentralization in Honduras. World Development: 33(4). B. Bingham, and C Gerard. 2006. Special issue on collaborative public management. Public Administration Review 66:1 170. Orias R. 2011 Report on law s and regulations governing civil society organizations in Bolivia. World Movement for Democracy http://www.wmd.org/sites/default/files/Bolivia_EN_2011.pdf (accessed 30 March 2013) Pacheco, P., W. de Jong, and J. Johnson 2010 The evolution of the timber sector in lowland Bolivia: Examining the influence of three disparate policy approaches. Forest Policy and Economics 12 (4): 271 276. Pgina Siete 2011. presidente indgena 24 September 2011. http://www.paginasiete.bo/2011 09 25/Nacional/Destacados/33Nac00121.aspx (accessed 20 March 2013). Pahl Wostl, C. 2009. A conce ptual framework for analyz ing adaptive capacity and multi level learning processes in resource governance regimes. Global Environmental Change (9) 3: 354 365.
220 Papadopoulos, Y. 2003. Cooperative forms of governance: Problems of democratic accountability in co mplex environments. European Journal of Political Research 42 (4): 473 501. Papado po u lo s, Y. and P.Warin. 2007. Are innovative, participatory and deliberative procedures in policy making democratic and effective? European Journal of Political Research 46 (4) : 445 472. Parkins, J. R. and R. Mitchell 2005. Public participation as public debate: A deliberative turn in natural resource management Society and Natural Resources 18:529 540. Parkinson, J. 2003. Legitimacy problems in deliberative democracy. Political Studies 51: 180 196. Perreault, T. 2012. Dispossession by accumulation? Mining, water and the natur e of enclosure on the Bolivian a ltiplano. Antipode 1 21. Prell, C. 2011. Some basic structural characteristics of networks. In S ocial networks and natural resource management: U ncovering the social fabric of environmental governance ed s Bodin, O. and C. Prell, 29 43. Cambridge, UK : Cambridge University Press. Provan, K. and P. Kenis. 2007. Modes of network governance: Structure, management, and effectiveness. Journal of Public Administration R esearc h and T heory 18: 229 252. Provan, K. G. and H. B. Milward. 1995. A preliminary theory of interorganizational effectiveness: A comparative study of four community mental health systems. Administrative Science Quarterly 40:1 33. Pretty J. and H. Ward 2001. Social capital and the environment. World development 29 (2):2 09 227. Raab, J. and P. Kenis 2009. Heading toward a society of networks empirical developments and theoretical challenges Journal of Management I nquiry 18 (3): 198 210. Ragalsky, P. 2010. Political processes and the reconfiguration of the state in Bolivia. Latin American Perspectives 37:35. Reed, M. S. 2008. Stakeholder participation for environmental management: A literature review. Biological C onservation 141 (10): 2417 2431. Ribot, J. C. 1999. Decentralization, participation and accountability in Sahelian forestry. Africa 69(1): 23. 2002. Democratic decentralization of natural resources, institutionalizing popular participation. Conference proceedings from World Reso urces Institute organized conference on decentralization and environment, Bellagio, Italy, February 2002. Washington, DC: World Resources Institute.
221 2003. A theory of a ccess. Institutions and Governance Program World Resources Institute Rural Sociolog y 68(2): 153 181. 2004. Waiting for democracy Washington, DC: World Resources Institute. Ribot, J. C., A. Agrawal and A. M. Larson 2006. Recentralizing while decentralizing: How national governments reappropriate forest resources. World Development 34:1864 1886. Rosenberg S, ed. 2007a Deliberation, participation and d emo cracy: Can the people g overn? Basingstoke, UK: Palgrave Macmillan. Rosenberg, J. P., and P. M. Yates. 2007 Schematic representation of case study research designs. Journal of advanc ed nursing 60 (4): 447 452. Rowe, G., and L. J. Frewer. 2004. Evaluating public participation exercises: A research agenda. Science, Technology & Human values 29 (4): 512 556. Rowley, T. J. 1997. Moving beyond dyadic ties: A network theory of stakeholder infl uences. Academy of M anagement Review 22 (4): 887 910. Rubin A., and E. Ba b bie 1997 Research methods for social work 3 rd ed. Pacific Grove, CA: Brooks/Cole. Ryfe, D. M. 2005. Does deliberative democracy work? Annual R eview of Political S cience 8: 49 71 Scharpf. F. W. 1999. Governing in Europe: Effective and democratic? Oxford: Oxford University Press. Schmidt, H. 1997. Healing the wounds of war: Memories of violence and the making of history in Zimbabwe's most recent past. Journal of Southern African St udies 23(2): 301 310. Sen, A. K. 1999. Democracy as a universal value. Journal of Democracy 10 (3): 3 17. Servicio Nacional de reas Protegidas (SERNAP). 2006. Plan de manejo: Area natural de manejo integrado nacional APOLOBAMBA. https://dlweb.dropbox.com/get/Apolobamba/70_PM_ANMI_Apolobamba.pdf?w=AABK ZL bkqGoVWAH5bCUvjP7xoCZI3qrSjNq1zf20lln5Q (accessed on 3 November 2012). 2012 Reportes del programa de monitoreo integral del ANMIN Apolobamba. La Paz, Bolivia: SERNAP. Skelcher, C., E. Klijn D. Kbler N. Mathur, E. Srensen and H. Sullivan 2006. Governance networks, democratic anchorage, and the impact of national po litical context. A Performing Public Sector: The Second Transatlantic Dialogue, Katholieke Universiteit, Leuven, Belgium. Workshop 3: Performance of multi sector/multi organization
222 Srensen, E. and J. Torfing 2007. Theoretical Approaches to Democratic Network Governance. In Theories of democratic network governance eds. E. Srensen and J. Torfing, 233 46. London: Palgrave Stirling, A. 2006. Analysis, participation and power: Justification and closure in participatory multi criteria analysis. Land Use Policy 23 (1):95 107. Suchman, M. C. 1995. Managing legitimacy: Strategic and institutional approaches. Academy of Management Review 20 (3) : 571 610. Tacconi, L. 2007. Decentralization, forests and liv elihoods: Theory and narrative. Global Environmental Change 17:338 348. Thede, N. 2011. Democratic agency in the local political sphere: Reflections on inclusion in Bolivia. Democratization 18(1): 211 235. Thompson D F. 2008. Deliberative democratic theor y and empirical political scienc e. Annu al Rev iew of Polit ical Science 11:497 520. Urioste, M., D. Pacheco A. Zoomers, and G. Haar 2000 Land market in a new context: T he INRA law in Bolivia. In C urren t land policy in Latin America: R egulating land tenure under neo liberalism ed A. Zoomers, 259 268. Royal Tropical Institute. Urioste, M., D. F. Tierra and L. Paz 2003. The abandoned agrarian reform: V alleys and high plains. La Paz, Bolivia: Fundacin Tierra Urioste, M. 2011 Concentracin y extranjeriza cin de la tierra en Bolivia. La Paz, Bolivia: Fundacin Tierra Van Cott, D. L. 2000. Party system development and indigenous populations in Latin America: The Bolivian c ase. Party Politics 6:155. Wasserman, S. and K. Faust 1994 Social network analysis: M ethods and applications. Cambridge, UK: Cambridge University Press. Webber, J. R. 2011. From rebellion to reform: Class struggle, indigenous liberation, and the politics of Evo Morales Chicago : Haymarket Books Yashar, D. 2005. Contesting c itizenship in Latin America: The r ise of indigenous movements and the post liberal challenge Cambridge UK : Cambridge University Press. Yin, R. K. 1994. Case study research: Design and m ethods Sage Publications. Young, I. M. (2000) Inclusion and Democracy Oxford: Oxf ord University Press. Zeller, R. A., and E. G. Carmines. 1980. Measurement in the social sciences: The link between theory and data Cambridge; New York: Cambridge University Press.
223 Zuidema, P. A., and R. G. Boot. 2002 Demography of the Brazil nut tree ( Bertholletia excelsa) in the Bolivian Amazon: I mpact of seed extraction on recruitment and population dynamics Journal of Tropical Ecology 18 (1): 1 31.
224 BIOGRAPHICAL SKE TCH Laura F. Kowler was born and raised in Richmond, Virginia She graduated with an A. Her undergraduate studies also included several memorable anthropology courses and a summer semester in Tanzania through the African Studies Department at the Un iversity of Georgia. After graduation, Laura taught conversational English to college students at the University of Agriculture and agricultural journalists in Prague, Czech Republic for a year. She then spent a few months learning about organic farming pr actices in both Spain and Portugal before moving of international conservation and development initiatives grew from her work involving small scale farmers an d local institutions in Ecuador from 2006 2007. This experience inspired the Basin. She received her Master of Science degree from the University of Florida in the su mmer 2009 and her Doctor of Philosophy degree from the Univ ersity of Florida in spring 2013