A POLITICAL ECOLOGY OF COMMUNITY-BASED FOREST AND WILDLIFE MANAGEMENT IN TANZANIA: POLITICS, POWER AND GOVERNANCE A THESIS SUBMITTED TO THE UNIVERSITY OF CAMBRIDGE IN CANDIDACY FOR THE DEGREE OF DOCTOR OF PHILOSOPHY KATHRYN ELIZABETH HUMPHRIES DOWNING COLLEGE DECEMBER 2012 Author’s Declaration This thesis is the result of my own work and includes nothing which is the outcome of work done in collaboration, except where specifically indicated in the text. The text does not exceed 80,000 words and no part has been submitted to any other university in application for a higher degree or diploma Kathryn E. Humphries Summary My research is focused on investigating the socio-political processes taking place within Community Based Natural Resource Management (CBNRM) in Tanzania. I draw on a political ecology approach in an investigation of the politics of struggles over natural resources, their management and the benefits that can be derived from this. I bring together theories of policy processes, African politics and scale into an examination of power within two case studies of CBNRM from the wildlife and forestry sectors. I carry out a comparative analysis of these case studies, employing a qualitative methodology based on semi-structured interviews, focus groups, participatory activities, participant observation and document analysis. My research is clustered around three core themes. Firstly, I trace the process of policy reform that introduced CBNRM in both the forest and wildlife sectors, and examine the differences between the governance systems prescribed in policy as a result of these processes. The contrasts between the two sectors in Tanzanian CBNRM are important and multiple. Different policy pathways were adopted, relating to the distinct political economies of forest and wildlife resources and their politicisation within the context of power devolution for CBNRM. The prescribed governance systems in the two sectors contain important differences in the processes by which local communities can apply to participate in CBNRM, the mechanisms of revenue distribution, and the ways in which power is devolved to the local level. Secondly I examine the implementation of these prescribed governance systems and their performance in reality through an exploration of the configurations of power set out in CBNRM, and the struggles that take place around these in ‘politics of scales’ as actors attempt to benefit from CBNRM. I examine the ways the governance systems have been adopted and adapted from those set out in CBNRM policy. I argue that the distinctions between the prescribed governance systems in the two sectors produce separate contexts of re-configuration into the performed governance systems within the case studies. However, I also argue that while the contexts are specific to each sector, both the case studies revealed the same underlying socio-political process of struggles over power to both manage and benefit from natural resources. These struggles to control and benefit from CBNRM are closely linked to the unequal distribution of benefits that were witnessed in both case studies. Finally I examine the performance of CBNRM as an integration of systems of power set out in policy and hidden, often unacknowledged, local contours of power. I address the themes of how the reality of CBNRM differs from that set out in policy, examine the processes ongoing within the projects that permit and maintain elite capture and unequal distribution of benefits, and investigate the socio-political processes of corruption taking place within devolved environmental management. I argue that the struggles over power, combined with hidden aspects, especially neopatrimonialism, local moral economy and the cultural context of corruption, are central to these unequal outcomes and the capture of benefits by a small group of individuals. My research highlights that power, the politics of its devolution to the local level, the struggles that take place around it, and its subtle, hidden forms, lie at the heart of gaining further understanding of the ways in which policies develop, the unexpected outcomes they produce and the inequalities these often entail. Acknowledgements I have been able to carry out my research thanks to a scholarship provided through the Economic and Social Research Council in the UK, which funded four years of research training, language study, fieldwork and writing up. I have been lucky enough to study something that I am fascinated with during my time at Cambridge, both during my MPhil and as a PhD student, and I am very grateful to the ESRC and to the Geography Department at Cambridge for providing me with this opportunity. I have been very privileged in having Bill Adams as my supervisor, and I thank him hugely for his unwavering patience and support and his ability to see exactly what it is that I am trying to say, often before I am sure myself. When I started my PhD, it felt like a big gamble to be taking on a completely new area of study in a country I had barely visited. I see that the risk was also great for him, and I thank him for taking it with me. Cambridge and the geography department have been a very happy home for me, and I am grateful to the friends that I have made here and the housemates that have shared the ups and downs and embraced my ‘Danish’ ways. Thanks and love especially to Han, Jen, Sally, Emily, Lauren, Ariadna, Arshiya, Isabel, Rohini and Girija. In Tanzania I made my home in several places, and am indebted to many. Firstly, without the Riverside Campsite and the teachers at its Language School I would not have learnt Kiswahili anywhere near as quickly, nor had so much fun doing it. Thanks especially to Theo and Mai for many tiring but happy hours. I benefitted greatly from working as a research associate in the Department of Wildlife Management at Sokoine University of Agriculture in Morogoro, with guidance from Professor Alex Songorwa, to whom I am very grateful for his support. During fieldwork I also worked with a number of research assistants, who set me straight and smoothed the path for me. Thank you to Theo, Jackson, Filemon and lastly to Bonny, who came with me to Idodi, will always be remembered for his impression of a chatu, and in whose memory this thesis is dedicated. I met some wonderful people in Iringa, and would particularly like to send all my love to Dave, Leslie, Andy and Angela, who took me under their wing and under their roofs, fed me wonderful cakes and let me be a part of their families whilst far from home. I must also say thank you to my family for never failing in their support and interest in what I’m doing, and for putting things in perspective. Finally, to Jonathan, who is always steady as a rock and untiring in his insistence that it will all be better than alright. Thank you for joining me in my adventures and for taking me on yours. Table of Contents: List of Tables:.......................................................................................................................... i List of Figures: ........................................................................................................................ i List of Acronyms .................................................................................................................... ii Chapter 1: Introduction .......................................................................................................... 1 1.1 People and Conservation: The Premise and Promise of Community-Based Natural Resource Management ............................................................................................ 1 1.2 People, Benefits and Power: Community-Based Natural Resource Management ...... 5 1.3 From Theory to Practice: Mixed Results, Disenchantment and the Crises of CBNRM ................................................................................................................... 8 1.4 The Socio-Politics of Natural Resource Management: A Research Agenda ............ 10 1.5 Thesis Structure ......................................................................................................... 15 Chapter 2: Theoretical Context............................................................................................16 2.1 Devolved Environmental Management ..................................................................... 17 2.1.1 Devolved Environmental Management: Power in CBNRM ............................. 17 2.1.2 Devolved Environmental Management: Communities in CBNRM .................. 21 2.1.3 Studying the Informal: Disorder, Corruption and Neopatrimonialism .............. 26 2.2 Policy Processes ........................................................................................................ 32 2.2.1 Linear Policy and Elite Decision-Making.......................................................... 32 2.2.2 Policy Network and Policy Processes Views..................................................... 34 2.2.2.1 Epistemic communities.............................................................................. 36 2.2.2.2 Policy Narratives and Discourse coalitions ............................................... 37 2.2.2.3 The Advocacy Coalition Framework ........................................................ 38 2.2.2.4 Policy space ............................................................................................... 39 2.2.3 Policy Processes: An Integrated Approach ........................................................ 40 2.3 Power ......................................................................................................................... 41 2.3.1 Structure and Agency ......................................................................................... 44 2.4 Politics of Scales ....................................................................................................... 50 2.4.1 Geography and Scale ......................................................................................... 51 2.4.2 Politics of Scales: Transformations ................................................................... 53 2.4.2.1 The Discursive Construction of Scale & Scalecraft .................................. 54 2.5 A Political Ecology of Nature Conservation ............................................................. 56 2.5.1 Political Ecology Perspectives ........................................................................... 57 2.5.2 Nature Conservation in Political Ecology.......................................................... 60 2.5.3 Political Ecology, CBNRM, Power and Politics of Scales ................................ 62 2.6 Synthesis.................................................................................................................... 64 Chapter 3: Tanzanian Context ............................................................................................. 66 3.1 Introduction ............................................................................................................... 66 3.2 Devolved Environmental Management ..................................................................... 67 3.2.1 Devolved Environmental Management: Colonial Appropriation of Resources, Socialism and Ujamaa ...................................................................................................... 68 3.2.2 Devolved Environmental Management: Neoliberalism, Decentralisation and Local Government Reform ............................................................................................... 70 3.3 Nature Conservation Paradigms, Policy, and Protected Areas in Tanzania.............. 75 3.3.1 Colonial Protected Areas: Strict Protection, Wildlife Hunting & Forest Reserves… ........................................................................................................................ 75 3.3.2 Post-independence Continuation: Commitment to Conservation ...................... 77 3.4 The Introduction of CBNRM .................................................................................... 81 3.4.1 Participatory Forest Management in Tanzania .................................................. 82 3.4.2 Wildlife Management Areas in Tanzania .......................................................... 83 3.5 Summary ................................................................................................................... 86 Chapter 4: Methodology........................................................................................................87 4.1 Ontology, Epistemology & Methodology ................................................................. 87 4.2 Research Design ........................................................................................................ 88 4.2.1 Case Study Selection.......................................................................................... 88 4.2.1.1 WMA Case Study ...................................................................................... 92 4.2.1.2 CBFM Case Study ..................................................................................... 97 4.2.2 Fieldwork Timetable ....................................................................................... 101 4.2.3 Research Outside of Tanzania ......................................................................... 101 4.2.4 Areas of study .................................................................................................. 101 4.3 Methods employed .................................................................................................. 102 4.3.1 Document Analysis .......................................................................................... 104 4.3.2 Village Preliminary Survey .............................................................................. 104 4.3.3 Focus Groups ................................................................................................... 105 4.3.4 Semi Structured Interviews .............................................................................. 107 4.3.5 Participatory Activities .................................................................................... 108 4.3.6 Participant Observation .................................................................................... 109 4.3.7 Local Assistant Reports ................................................................................... 110 4.4 Triangulation ........................................................................................................... 110 4.5 Language ................................................................................................................. 111 4.5 Recording and Note-taking Strategies..................................................................... 112 4.5 Transcription and Analysis...................................................................................... 112 4.6 Constraints and Reflections ..................................................................................... 113 4.6.1 Positionality ..................................................................................................... 113 4.6.2 Sensitive Issues ................................................................................................ 114 4.6.3 Ethics................................................................................................................ 115 4.6.4 Reflexivity........................................................................................................ 116 Chapter 5: Policy Processes in Tanzanian Community-Based Natural Resource Management ........................................................................................................................ 117 5.1 Introduction ............................................................................................................. 117 5.2 Prescribed Governance Systems ............................................................................. 118 5.2.1 Community-Based Forest Management Institutions ....................................... 118 5.2.2 CBFM Revenue Systems ................................................................................. 121 5.2.3 Wildlife Management Area Institutions........................................................... 122 5.2.4 WMA Investment Agreements and Revenue Arrangements ........................... 127 5.2.5 A Comparative View of the Governance Systems in Tanzanian CBNRM ..... 129 5.3 Community-Based Natural Resource Management Policy ..................................... 135 5.3.1 External Influences in Agenda Setting ............................................................ 136 5.3.2 The Rationality of Community-Based Conservation ....................................... 138 5.4 Policy Pathways ...................................................................................................... 140 5.4.1 The Politics of Devolution: Political Economies of Natural Resources .......... 140 5.4.2 The Politics of Devolution: Resistance and Recentralisation .......................... 143 5.5 Policy processes ...................................................................................................... 145 5.5.1 Wildlife Policy Networks: Competing Narratives and Coalitions in Policy Processes......................................................................................................................... 146 5.5.2 Policy Entrepreneurs ........................................................................................ 155 5.5.3 Pilot Projects in the Issue Network: Roles of Local Level in National Policy Development................................................................................................................... 159 5.5.3.1 REWMP’s Place in Policy ........................................................................ 162 5.5.3.2 REWMP: Networks, Relationships and the Creation of a National Example….. ................................................................................................................ 164 5.6 Chapter Summary .................................................................................................... 166 Chapter 6: the Performance of Policy I: Power Systems in Governance and Politics of Scales.....................................................................................................................................169 6.1 Introduction ............................................................................................................. 169 6.1.1 Devolution and Topographies of Power .......................................................... 169 6.1.2 Scalar Framings in Wildlife Management Areas and Community-Based Forest Management ................................................................................................................... 175 6.2 Benefit Distribution and Scalar Configurations of Power ...................................... 176 6.2.1 Elite Capture through Prescribed Governance ................................................. 179 6.3 Politics of scales ...................................................................................................... 181 6.3.1 Discursive Construction and Reinforcement of Scales: The Protection of Power…. ......................................................................................................................... 181 6.3.2 Scalar Transformations: The Struggle for Power ............................................ 185 6.3.2.1 CBFM: Scaling Down ............................................................................. 185 6.3.2.2 WMA: Scaling Up ................................................................................... 187 6.3.3 Politics of Scales: Same Politics, Different Scales .......................................... 190 6.3.4 Scalar Transformations: Conflicts and Power Struggles ................................. 191 6.3.4.1 CBFM Conflicts: Resource User Groups ................................................ 192 6.3.4.2 WMA Conflicts: Inter-Village Politics and Struggles for Revenue ........ 196 6.4 The Performance of Policy: Power in Reality ......................................................... 203 6.5 Summary ................................................................................................................. 206 Chapter 7: Performance of Policy II: Hidden Aspects of Governance and Participatory Spaces....................................................................................................................................209 7.1 Introduction ............................................................................................................. 209 7.2 Deviations from Policy: Rule Breaking in the Performance of Policy ................... 211 7.2.1 Disincentives within the Prescribed Governance System ................................ 213 7.2.2 Discourses of Rule-Breaking ........................................................................... 215 7.3 Enforcement, Moral Economy & Reciprocity ........................................................ 216 7.4 The Corruption Complex ........................................................................................ 218 7.4.1 The Corruption Complex: Neopatrimonialism ................................................ 218 7.4.2 The Corruption Complex: Bribery ................................................................... 223 7.4.3 The Corruption Complex: Clientelism............................................................. 227 7.4.4 Scales of Corruption ........................................................................................ 230 7.5 Cultural Aspects of Power and Corruption ............................................................. 232 7.5.1 Cultural Aspects of Corruption: The Tanzanian Big Man ............................... 235 7.6 Summary ................................................................................................................. 237 Chapter 8: Conclusions........................................................................................................241 8.1 Politics, Power and Governance: Empirical and Theoretical Research Contributions ............................................................................................................................. 243 8.1.1 A Comparative Perspective of CBNRM in the Wildlife and Forestry Sectors 243 8.1.2 Winners and Losers in Tanzanian CBNRM .................................................... 245 8.1.3 The Reality of Community-Based Natural Resource Management ................ 246 8.1.4 The Discursive Process of Policy Reform ....................................................... 249 8.2 Practical Research Contributions, Implications and Future Directions .................. 250 References.............................................................................................................................255 Appendix 1: Bundles of rights in natural resource management (adapted from Agrawal & Ostrom, 2001) ........................................................................................................................ 288 Appendix 2: Schedule of Interviews, Focus Groups and Qualitative Surveys Carried out during Fieldwork .................................................................................................................... 289 Appendix 3: List of Documents Collected and Analysed..................................................... 292 i List of Tables: Table 1.1: Research Questions Addressed in this Thesis by Chapter Table 2.1: Seven forms of social power Table 3.1: Local Government Structure and Functions in Mainland Tanzania Table 3.2: Forest Reserve Categories in Tanzania according to the Forest Act 2002 Table 4.1: Villages Participating in MBOMIPA by Administrative Division Table 4.2: Semi-Structured Interviews, Qualitative Surveys and Focus Groups Carried out in the Wildlife Sector Table 4.3: Semi-Structured Interviews, Qualitative Surveys and Focus Groups Carried out in the Forestry Sector Table 4.4: Participatory Activities Carried out During Fieldwork Table 5.1: Principal Roles and Responsibilities of Institutions within Community Based Forest Management Table 5.2: Principal Roles and Responsibilities of Institutions within Wildlife Management Areas Table 5.3: Competing Discourse Coalitions in Tanzanian Community Based Conservation Prior to National Policy Development Table 6.1: Monetary allowances derived from the case studies of CBFM and WMA 2008-10 Table 6.2: Kiwere CBFM Income Derived From Permits Issued in 2010 Table 6.3: MBOMIPA Authorised Association Committee Structure and Representatives’ Villages Table 6.4: Axes of Power within MBOMIPA WMA Table 7.1: MBOMIPA Authorised Association Expenditure by Category 2007-9 List of Figures: Figure 2.1: The Linear Model of Policy Reform Figure 2.2: Policy Communities and Discourse Coalitions within an Issue Network Figure 3.1: Map of the United Republic of Tanzania showing the Administrative Region of Iringa Figure 3.2: Local Government Reporting Structures in Tanzania using Participatory Forest Management as an Example Figure 4.1: Map of Iringa Region, Tanzania Figure 4.2: REWMP/MBOMIPA Case Study Area Figure 4.3: CBFM Case study area Figure 5.1: Institutional Governance Structure Prescribed in Policy for CBFM in Tanzania Figure 5.2: Institutional Governance Structure Prescribed in Policy for WMAs in Tanzania Figure 6.1: Power structure prescribed in policy in CBFM in Tanzania Figure 6.2: Power structure prescribed in policy in WMAs in Tanzania Figure 6.3: Graph showing Kiwere CBFM Income Derived From Permits Issued in 2010 Figure 6.4: Map of MBOMIPA Participating Villages and Land within the Wildlife Management Area Figure 6.5: Re-scaled Power Structure as Performed in CBFM Figure 6.6: Re-scaled Power Structure as Performed in WMA Figure 7.1: Government Anti-Corruption Poster ii List of Acronyms AA Authorised Association CAMPFIRE Communal Areas Management Programme for Indigenous Resources CBFM Community Based Forest Management CBNRM Community Based Natural Resource Management CCS Community Conservation Service DANIDA Danish International Development Agency DNRAB District Natural Resources Advisory Board IUCN International Union for the Conservation of Nature GCA Game Control Area GDP Gross Domestic Product GTZ Gesellschaft für Technische Zusammenarbeit JFM Joint Forest Management LAMP Land Management Programme LGA Local Government Act MBOMIPA Matumizi Bora Maliasili Pawaga na Idodi (sustainable use of natural resources in Pawaga and Idodi) MEMA Matumizi Endelevu Maliasili (sustainable use of natural resources) MNRT Ministry of Natural Resources and Tourism NFP National Forestry Policy NGO Non-Governmental Organisation NWMP Natural Woodlands Management Project PA Protected Area PAWM Planning and Assessment for Wildlife Management PFM Participatory Forest Management REDD Reduced Emissions from Deforestation and Degradation REWMP Ruaha Ecosystem Wildlife Management Project SCP Selous Conservation Programme SIDA Swedish International Development Cooperation Agency SRCS Serengeti Regional Conservation Strategy TANAPA Tanzanian National Parks Authority TANU Tanganyika African National Union TAWICO Tanzanian Wildlife Corporation TNCO Transnational Conservation Organisation UMFM Udzungwa Mountains Forest Management and Biodiversity Conservation Project UNEP United Nations Environment Programme URT United Republic of Tanzania VLFR Village Land Forest Reserve VNRC Village Natural Resources Committee WCA Wildlife Conservation Act WMA Wildlife Management Area WPT Wildlife Policy for Tanzania WWF Worldwide Fund for Nature Chapter 1: Introduction 1 “If community conservation is the answer in Africa, what is the question?” (Adams and Hulme, 2001a: 193) In this introduction, I present my research agenda. I aim to justify this research focus and to show what it is that researchers should be asking about community conservation projects. My research, presented in this thesis, is concerned with the socio-political complexities of Community-Based Natural Resource Management (CBNRM) as it is currently being implemented in rural Tanzania. CBNRM is a form of community conservation which has been widespread across sub-Saharan African, and is particularly important in Tanzania, where it is one of two major national policies for community conservation, but remains relatively little studied. Since the 1980s, the varied programmes and types of projects within community conservation, including CBNRM, have grown to become a new orthodoxy (Hulme and Murphree, 2001a), but it is not without its critics and the results of three decades of community conservation have left many disappointed (e.g. Brandon et al., 1998; Terborgh, 1999). In this thesis, I argue that research needs to focus on understanding the social, institutional and political dimensions of CBNRM to add depth of insight into the (often unexpected) outcomes. I position my research in this field, as an in-depth study of two CBNRM case study projects, one in the wildlife sector and the other in forestry. I begin by discussing the paradigm shift to a community approach to conservation, its historical development, justifications and the foundational concepts on which all types of community conservation are based. In 1.2, I then focus on the core characteristics of CBNRM within this broader paradigm of community conservation. In 1.3, I review the literature discussing whether or not CBNRM ‘works’ and the assertion that CBNRM is in crisis. Finally in 1.4 and 1.5 I outline a research agenda for CBNRM that introduces socio- political processes and power as a lens for understanding the reality of CBNRM. 1.1 People and Conservation: The Premise and Promise of Community-Based Natural Resource Management The term community conservation is applied to a broad spectrum of approaches that are centred around “ideas, policies, practices and behaviours that seek to give those who live in rural environments greater involvement in managing the natural resources...and/or greater access to benefits derived from those resources” (Hulme and Murphree, 2001a: 4). Such 2 approaches, whilst varied, began to emerge in the 1980s, and by the 1990s had become a standard model of conservation programmes (Leach et al., 1997a; Fabricius, 2004; Hutton et al., 2005). This represented a dramatic shift in conservation ideology and practice. Historically conservation had been dominated by a protectionist model of conservation, which sought to remove the influence of people from wild spaces and safeguard natural resources under the management of the state (Leach et al., 1997a; Adams and Hulme, 2001a). The Protected Area (PA) model, in which areas of land are set aside for protection and conservation purposes, dominated conservation efforts. Whilst PAs encompass multiple strategies to delimit land for conservation purposes, the national park, first introduced at Yellowstone in the USA in 1872, and the productive nature reserve, first introduced in colonial India, are the typical expressions of this model (Barton, 2001; Adams, 2004). These ideas reflect the conceptual separation of people from nature, contrasting nature to the ‘civilised’ spaces which people inhabit, and advocate protecting wilderness by creating wild, undisturbed spaces from which human impact has been removed (Marsh, 1865; Adams and Hulme, 2001a; Pretty, 2002; Neumann, 2004a). People, and the ways in which they used such lands and natural resources, did not fit comfortably into this construction of wild nature; conservation narratives identified these people as deadly hunters and poachers, pitched against sentient, innocent and scared animals in a ‘war for biodiversity’ that threatened these wild spaces (Neumann, 2004a). Colonial narratives of the ignorant ‘noble savage’, particularly with reference to hunting and forest management (MacKenzie, 1987; Barton, 2001), alongside the reification of Western scientific knowledge for effective management, created powerful arguments and justifications for the removal of natural spaces into state- managed PAs and the exclusion of local people who lacked the knowledge to manage these spaces effectively (Adams and Hulme, 2001a). Conservation became dominated by narratives that pitched people against nature, therefore (Peluso, 1993). Whilst PAs still form the backbone of conservation efforts to this day, conservation developed a ‘human face’ (Bell, 1987) and, from the 1970s, new narratives of community conservation “stressed the need not to exclude local people, either physically from PAs or politically from the conservation policy process, but to ensure their participation” (Hutton et al., 2005: 342). This paradigm shift to community conservation was based on several lines of argument. Firstly, in a context of increasing concern over human impacts on the biosphere, extinction rates, the global scale of environmental issues, and arguments that conservation must move beyond the borders of PAs, protectionist conservation came under criticism for its 3 high costs, inefficiency and poor results (see Western et al., 1994; Adams, 2004; Fabricius et al., 2004; Adams, 2009). Secondly, the rise of community conservation narratives drew on ideas of the economic impacts of PAs on local communities and the exacerbation of poverty resulting from biodiversity conservation (Western and Wright, 1994; Sunderlin et al., 2005). The impoverishing impacts of PAs were highlighted through the polarisation of costs and benefits at the local and national/international levels respectively (Ferraro, 2002), generating increased concerns over the human rights impacts of PAs, especially forced evictions and militaristic management (Neumann, 2004a), and drawing attention to the social impacts issues arising from these (Neumann, 1998; Brockington, 2002; Attfield et al., 2004). Accounts of conservation and PAs also highlighted the invalidity of arguments that removed people from PAs, highlighting the socially-constructed nature of concepts such as wilderness and pristine landscapes free from human influence, and the political motivations for state control of natural resources and the annexation of land for conservation purposes (Neumann, 1998; Scott, 1998; Neumann, 2001). The rapid rise and adoption of community conservation was also a question of context and timing, however (Hutton et al., 2005). By tying together conservation and poverty reduction objectives, community conservation narratives brought conservation in line with ideas about sustainable development in the 1980s (Roe, 1991; Adams and Hulme, 2001b). The principle of integrated conservation and development and the ultimate objective of sustainable development was enshrined in the Brundtland Report (1987) and the UN Conference on Environment and Development, held in Rio in 1992 (Western et al., 1994; Adams et al., 2004; Hutton et al., 2005). This integration of conservation and development objectives has often been termed a ‘win-win’ solution and ‘pro-poor’ conservation (Adams et al., 2004), that represents major changes in both conservation and development discourses that were taking shape during the 1970s and 1980s (Roe, 2008). The shift to the incorporation of people in nature conservation and the management of natural resources drew heavily upon the paradigm of community participation from within the field of development, which called for “rural people’s direct involvement in development activities while at the same time promoting both economic and social development” (Wainwright and Wehrmeyer, 1998: 933). This shift in development thinking away from state-led, top down and technocentric development that dominated the development era post World War II was described as a shift from ‘things’ to ‘people’ (see Escobar, 1995; Leys, 1996). The new paradigm was conceptualised as ‘bottom-up development’ (Chambers, 1983; Chambers, 4 1997; Adams and Hulme, 2001b; Parfitt, 2004; Williams, 2004a). Participatory development emerged as a backlash against the failures of state-led development programmes (Pieterse, 1998), but was also undoubtedly a consequence of the emerging interests in the capacity of local communities to manage natural resources, specifically the development of common- pool resource theories (see 2.1.2; Ostrom, 1990), but also the new international policy agenda and the amalgamation of neo-classical economics and liberal democratic theory in neoliberalism (Virtanen, 2005). Neoliberalism “revolves around the restructuring of the world to facilitate the spread of free- markets1” (Igoe and Brockington, 2007: 433). Neoliberalism is a broad term with multiple meanings, but which is synonymous with both deregulation and, in later forms, the emergence of hybrid forms of environmental governance (Peck and Tickell, 2002; Brooks et al., 2006; Charnley and Poe, 2007; Tacconi, 2007; Andersson and Ostrom, 2008; Brockington, 2008). Under neoliberalism, businesses, Non-Governmental Organisations (NGOs) and communities, are seen to share responsibility for conservation with the state (Peck and Tickell, 2002; Igoe and Brockington, 2007). Neoliberal forms of environmental governance presented an opportunity for the state to save financial resources through devolution, and was also supported by populist justifications for community participation which argued that local communities are best-placed to conserve natural resources (with external support) because they already use and rely upon them (Dressler et al., 2010). Through market mechanisms and new forms of environmental governance, neoliberalism promised “a world in which it is possible to eat one’s conservation cake and have development dessert too” (Grandia, 2007 in Igoe and Brockington, 2007: 434). Community conservation was supported by neoliberal arguments that ‘win-win’ scenarios for conservation and development are possible; for example it was argued that the development of ecotourism would open up conservation spaces to tourism markets and generate revenue for investment in conservation and development, creating a sustainable system of continuous benefits (Igoe and Brockington, 2007). The confluence of these factors during the 1980s set the scene for the emergence of a new paradigm of community conservation which drew on the ascendance of neoliberalism, the inclusion and participation of local communities in bottom-up development and the 1 The ‘New Policy Agenda’ for foreign aid emphasised the role of market mechanisms, rather than state planning, to correctly set the economic incentives necessary to achieve policy aims, including conservation (Hutton et al., 2005). 5 integration of conservation and development objectives. Following the institutionalisation of community approaches to conservation at the international level, for example in the World Congresses on National Parks and Protected Areas in 1982 and 1992, a plethora of large- scale attempts at community conservation emerged, including the Man and Biosphere Programme, Integrated Conservation and Development Projects (see Brandon and Wells, 1992; Wells, 1998), and ‘conservation with development’ projects (Hutton et al., 2005). Community conservation, however, is a term that covers a wide spectrum of approaches that vary from PA outreach to devolved natural resource management, and can be distinguished by the values they place on nature and the powers devolved to communities (see Adams and Hulme, 2001b; Barrow and Murphree, 2001; Hulme and Murphree, 2001b; Adams and Hutton, 2007). Within this spectrum of community conservation, CBNRM is an approach that focuses on conservation for use values, alongside the devolution of power to manage natural resources to the level of the participating community, as discussed in the next section (Adams and Hulme, 2001b). 1.2 People, Benefits and Power: Community-Based Natural Resource Management The core principles of CBNRM emerged “out of a desire to rectify the human costs associated with coercive conservation...to return the stewardship of biodiversity and natural resources to local communities through participation, empowerment and decentralization” Dressler et al. (2010: 5). My focus is therefore upon approaches that involve power sharing, not just benefit sharing (Charnley and Poe, 2007) and I use the following three characteristics to define a CBNRM approach: (1) the inclusion of local communities in conservation practice, specifically through the devolution of management rights and responsibilities to the local level; (2) the central aim of sustainable use of resources in such projects (i.e. anthropocentric and utilisation values of nature); (3) the incorporation of economic benefits at the local level to compensate for the costs of conservation and further incentivise it (Adams and Hulme, 2001a; Hulme and Murphree, 2001b). CBNRM was the most widely implemented form of community conservation across southern Africa in the 1990s, and it focused almost exclusively on the management of wildlife resources and the generation of revenue through hunting (Fabricius et al., 2004). CBNRM later broadened to incorporate the management of other natural resources, and has become a widely implemented approach in the forestry and non-timber forest products sectors (Hutton et al., 2005). In this thesis I 6 employ the term CBNRM in an investigation both the wildlife and forestry sectors in Tanzania. The term CBNRM remains most commonly associated with wildlife programmes, such as Zimbabwe’s Communal Areas Management Programme for Indigenous Resources (CAMPFIRE; Murphree, 1996), but is entirely appropriate to the study of community forestry, which Charnley & Poe (2007) define according to the same three characteristics. There is a clear issue of terminology surrounding CBNRM, including a variety of approaches claiming to be CBNRM and an equally large variety of approaches that can be classified as CBNRM but are not labelled as such (Roe et al., 2009). I follow a definition of CBNRM that alongside these three key components, “inherently means local groups of people (‘communities’) managing resources in an active manner and with some degree of formal (de jure) or informal (de facto) control or tenure over those resources (see 2.1.3; Roe et al., 2009). The term is uncommon in West and Central Africa, where ‘gestion de terroir’ or sustainable resource management approaches have been more widely implemented (Roe et al., 2009). In East Africa the defining characteristics of CBNRM are more commonly practiced, but often fall under sectored policies of ‘community-based conservation’ in wildlife and ‘participatory forest management’ in forestry (Roe et al., 2009). Both community conservation and CBNRM in the developing world vary widely in their characteristics (Agrawal and Ostrom, 2001; Charnley and Poe, 2007; Shackleton et al., 2010). For example, whilst East Africa’s programmes have focused very much around the existing PA network, in Southern Africa there has been a larger focus on private land in community conservation (Barrow et al., 2001). Within the forestry sector, distinct policies have developed in several examples of national policy reform to incorporate CBNRM, including Nepal’s high profile Community Forest User Groups, Community-Based Forest Management in Tanzania, Lesotho, Namibia, Malawi and Uganda (Charnley and Poe, 2007). In the wildlife sector, Namibia is heralded as one of the most successful examples of CBNRM through its Communal Conservancy Programme, which devolves broad and secure management rights for natural resources to local communities, who retain full revenue from diverse income sources, including hunting, tourism and non-timber forest products (Roe et al., 2009). In 1999 a report by the Food and Agriculture Organisation of the United Nations reported the devolution of authority to manage forest natural resources to the local level had begun in over 50 countries (cited in Agrawal, 2001). 7 CAMPFIRE is an excellent example of a model CBNRM approach, which has been highly praised for its innovativeness and for its commitment to decentralised authority (Jones, 2006). This praise has been accompanied by equal measures of criticisms however (see 1.3). CAMPFIRE provides a good example of the governance systems and management structures typical of CBNRM projects; Village and Ward Development Committees are set up, which, in theory, empower the local community and enable them to derive benefits from natural resources by entering into contracts with private investors (most commonly hunting and photoraphic tourism operators; Murombedzi, 2003). CAMPFIRE follows the market-based approach of the CBNRM model by focusing particularly upon the management of high-value wildlife resources (Songorwa et al., 2000), which are integrated into markets as consumptive and non-consumptive resources, to provide economic benefits and incentives for sustainable use to the local communities (Jones, 2006). Market-based approaches within community conservation and CBNRM became much more common in the 1990s, leading to what Hulme & Murphree (1999) termed a ‘new conservation’ that has been popular in both the wildlife and forestry sectors due to the high value of these resources. Such a relationship between CBNRM and markets is an important development, described by Dressler et al. (2010: 6) as a process of “perverse hybridiz[ation] with wider neoliberal structuring”. Multiple strategies have been employed in community conservation to channel benefits from conservation to local communities, including compensation and the initiation of alternative livelihood strategies (Turner et al., 2004), but within those that emphasise sustainable use and the devolution of power over natural resource management to the local level, this market-based approach has been key (Magome and Fabricius, 2004). This type of CBNRM has placed a lot of emphasis on creating sustainability through nature ‘paying its way’, and ensuring that the full costs of conservation can be covered, including the social costs (Smith and Scherr, 2003). CBNRM became an excellent example of neoliberal conservation therefore, through its fostering of close relationships with the private sector and especially natural resources and tourism markets, and through producing a hybrid form of environmental governance that focuses upon merging neoliberal ideas of private markets with populisms arguments for devolved governance (Peck and Tickell, 2002). In the United Republic of Tanzania, CBNRM has followed this market-based approach and implemented twin policies in the wildlife and forestry sectors (see 3.4). These are called Wildlife Management Areas (WMAs) and Community-Based Forest Management (CBFM) 8 respectively, and it is these policies and the implementation that are under consideration in this thesis. 1.3 From Theory to Practice: Mixed Results, Disenchantment and the Crises of CBNRM “For all the idealism inherent in CBNRM, it is never actually ideal in practice.” (Dressler et al., 2010:6) As CBNRM became increasingly popular and present around the world, debates over how to balance the objectives and needs of conservation and development within CBNRM, and community conservation approaches more widely, developed rapidly over the 1990s (Büscher and Dressler, 2007). Criticisms of the community conservation paradigm have been aimed at its ideology, the assumptions made and its implementation, and have often been focused on the incompatibility of people and nature (see Barrett and Arcese, 1995; Songorwa et al., 2000), the need for rigidly protected conservation areas, the moral imperative of biodiversity conservation, the crisis of biodiversity decline and the myth of eco-friendly local communities (Büscher and Dressler, 2007). Such critiques and criticisms have led many to identify a trend of going ‘back to the barriers’ (Hutton et al., 2005), a return to fortress conservation, which Büscher & Dressler (2007) refer to as the neoprotectionist turn. Criticisms of CBNRM have been part of this neoprotectionist turn. Whilst the promises of CBNRM made it the “darling of funding agencies” (Shackleton et al., 2010: 2), reports from studies of CBNRM in practice often showed that its twin objectives have not been met and/or not balanced, producing a crisis of CBNRM. Evidence from CBNRM projects implemented around the world since the 1980s is mixed and many have expressed disappointment with the outcomes and disillusionment with the ideology (Hutton et al., 2005; Jones, 2006; Roe, 2008). Case studies have highlighted the diversity in objectives regarding conservation and development within such approaches and identify failures to deliver the expected outcomes for either conservation or development (Hackel, 1999; Sundar, 2001; Francis and James, 2003; Berkes, 2004; Hutton et al., 2005; Selfa and Endter-Wada, 2008). It has become accepted that win-win scenarios in reality usually consist of a set of trade-offs between conservation and development (McShane et al., 2011). 9 Whilst some individual projects remain celebrated examples, others never get off the ground, some fall into obscurity and some are seen as catastrophic failures. CAMPFIRE has produced several extremely successful examples of local benefits in terms of both revenues generated and resulting development (e.g. the Masoka example in Murombedzi, 2001). However, the complex nature of interpreting the results from CBNRM is clear from CAMPFIRE: Frost & Bond (2008) found that between 1989 and 2001, across Zimbabwe the revenue generated amounted to over $20 million transferred to participating communities. However, this impressive figure hides large inequalities between the amounts generated in different projects, wherein just 12 of the participating 37 districts generated an overwhelming 97% of these revenues (Frost and Bond, 2008). CAMPFIRE has not generated significant benefits everywhere it is implemented, but success was identified as concentrated in specific areas with small, homogenous communities combined with high value, desirable game hunting opportunities (Bond, 2001; Jones, 2006). Furthermore the CAMPFIRE example and other studies of CBNRM have shown that the predicted win-win outcomes are often not apparent, or disappointingly small, especially at the level of the individual household (Magome and Fabricius, 2004; Ministry of Natural Resources and Tourism, 2009). Asking questions about how much revenue is generated from CBNRM is clearly not enough to try to understand the complexities of its implementation or its outcomes. Generating revenue from CBNRM is one side of the battle for development objectives and many investigations into CBNRM have argued that getting this revenue to the community level is another. The CAMPFIRE projects often had their revenue taxed at the regional level at a rate of 50% (Bond, 2001). Beyond this, local people’s participation and the devolution of power in such projects has been repeatedly questioned (Gibson and Marks, 1995; Gillingham, 1998; Murombedzi, 1999; Goldman, 2003). The outcomes of CBNRM for conservation objectives are no clearer. Whilst many hail the increase in the area of land under some form of protected status through CBNRM as a major success of the approach, others argue that the influence upon people’s attitudes towards hunting or other illegal uses of natural resources, and wider conservation have not necessarily been changed by the introduction of community benefit schemes through CBNRM (e.g. Gillingham and Lee, 1999). Gibson & Marks’ (1995) study of CBNRM in Zambia concluded that the tactics of hunting had changed, shifting to alternative areas and alternative species in response to the introduction of CBNRM, but the incentives that led people to hunt remained unchanged. It is clear that assumptions about how CBNRM works in practice often ignore the politics of 10 power devolution and of the costs, benefits and balance between conservation and development for the people participating in CBNRM. There are ongoing debates about the ways communities have been portrayed within CBNRM ideology and strategy. This topic is discussed in further detail in 2.1.2, but it is important to note that a key area of research has begun to identify problems and look at how CBNRM takes place at the local level. The issues of elite capture by wealthy groups and individuals, and unequal distribution of costs and benefits within a community has been widely identified (e.g. Songorwa et al., 2000; Goldman, 2003; Hadiz, 2004; Meshack et al., 2006). CBNRM remains a leading strategy in the conservation of natural resources around the world, is aligned with the global paradigm of sustainable development and has received countless amounts of funding and effort over its history, yet its results are generally mixed, often disappointing and little-understood (Dressler et al., 2010). In section 1.4, I outline a research agenda for CBNRM that focuses specifically upon understanding the complex nature of implementation of such an approach. Research must focus on investigating the politics of natural resource management because natural resources, their management, the communities that undertake this and the changes in state-society relations that this involves are all political in their nature (Campbell and Vainio-Matila, 2003; Blaikie and Springate- Baginski, 2007). Sikor & Nguyen (2007) provide a clear example of this kind of research agenda in their study of the unequal distribution of benefits within CBNRM in Vietnam’s forestry sector. They highlight two critical areas for further research: (1) the political-social- economic context in which the devolution of rights to local communities to manage natural resources takes place; and (2) to investigate the clear patterns of costs and benefits at the local level resulting from locally-specific contexts of rights and access to natural resources. My research responds to these priorities, and I further argue that to develop our understanding of the complex outcomes that continue to result from CBNRM, research must address questions of the socio-political processes taking place within CBNRM projects. 1.4 The Socio-Politics of Natural Resource Management: A Research Agenda My research agenda is anchored in a belief that natural resource management is intrinsically and inescapably political. This political nature is often left unrecognised and de-politicised by focusing instead on technical aspects of policy or project implementation, or what Ferguson (1990) refers to as the ‘anti-politics machine’. Anti-politics serves to simplify complex socio-political issues and simultaneously legitimise the work of conservation and 11 development as outside of politics through ‘mobilizing metaphors’ such as participation, good governance and ownership (Büscher, 2010). Issues such as elite capture, the exacerbation of poverty through CBNRM and undesirable outcomes for conservation are often explained away as poor technocratic policy design, improper implementation or a lack of political will to devolve power to local communities (Hadiz, 2004). I argue that a better understanding must focus on the workings of CBNRM, move beyond naive assumptions as to how it works in practice and apply an explicitly socio-political lens to investigate the processes at work around natural resource management that lead to these outcomes. There is a large focus within CBNRM studies on the economic benefits being channelled to the local community, and is often used as a benchmark of success and failure within these projects (Murphree, 2000; Songorwa et al., 2000; Murombedzi, 2001; Brockington, 2002; Murombedzi, 2003; Murphree, 2005). However, there has been little focus upon relationships between different stakeholders involved in CBNRM, power relations, rights, equity and justice (Ribot, 1999; Shackleton et al., 2010). Questions need to be asked about the socio-political processes that govern how money is distributed and used, how decisions are made and how inequalities remain unchanged. The crisis of CBNRM lies in the need to understand the ‘how’ of governance processes across multiple levels (Büscher and Dressler, 2007). It is clear that in order to understand the successes and failures and the unexpected outcomes of CBNRM, attention must be paid to several socio-political factors including the nature of policies for management of natural resources, the governance systems and rules of natural resource use that they impose on communities for their management, and the socio- political nature of that management within the communities. This research agenda must also incorporate an investigation of the micro-politics in CBNRM and situate such practices within political economy and local context, therefore (Büscher and Dressler, 2007; Büscher, 2010; Dressler et al., 2010). CBNRM is a process of policy reform that necessarily involves the devolution of power from the state to the local level, mandating communities with the authority to manage natural resources and to derive benefits from this. Power is a central theme in the socio-politics of natural resource management, therefore. There are two clear areas relating to power that require further investigation: the first relates to power devolution, and the complex reality of how this is being implemented. If CBNRM begins with the devolution of power, we must investigate how this power is devolved, the politics involved in this process and what it means for the communities that take on the responsibility of managing such natural resources 12 (see 2.1.1); the second revolves around communities that participate in CBNRM, and explores questions relating to who such communities are and how they manage natural resources. If the management of natural resources is intrinsically political, there is no reason why their management at the local, rather than the state level is any less political (see 2.1.2). In this thesis I aim to investigate the ways in which CBNRM is driving socio-political struggles at the local level for access to and benefits from natural resources. Here again power is a core theme, based on the idea that the devolution of power over natural resource management takes place within existing and complex local systems of power. Kumar’s (2002) study of Joint Forest Management in India and Gillingham’s (1998) study of Community-Based Wildlife Management in Tanzania both argue that power structures lie at the heart of facilitating and maintaining elite capture (of both income and influence) within communities. Whilst the reporting of socio-political factors as important has become common-place, marking an important shift towards the research agenda presented in this thesis, investigations into the processes at work are rare. It is to this gap in research that the research project is orientated. CBNRM is a mixture of the policy that sets it out, creating a framework for implementation and the structures and systems of management, alongside its implementation, by people and in reality. I refer to these as the prescription of policy and its performance. CBNRM highlights that there are differences between the two, and it is important to ask what these are, why they occur and how, or through what processes. One of the central goals of my research is to identify the differences between prescribed and performed governance systems in Tanzanian CBNRM, and to uncover the socio-political processes by which the former is re- shaped into the latter. I employ the term governance, defining it as “the activity of every individual or collective that both through formal and informal means tries to create steering mechanisms, set goals, frame demands and develop policy with the intent of bringing these into practice” (Rosenau, 2001 in Büscher and Dressler, 2007: 593). This definition is a broad view of governance that draws on the political nature of resource management, its policies and the communities engaged in it. This definition highlights the fact that governance takes place between states, policies and communities, in processes of both policy prescription and performance, none of which are neutral, harmonious or altruistic, but are engaged in processes of socio-political struggle. This view acknowledges that the politics of natural resource management are not static, but are under constant re-negotiation, and that power 13 struggles and socio-political processes also take place within an informal setting (see 2.1.3 and 2.4). There are three over-arching questions that formed the basis of my research interest at the outset of this research project: • How and why have the wildlife and forestry sectors implemented CBNRM differently in Tanzania? • How and why do the same small group of individuals benefit disproportionately from CBNRM? • How is the reality of governance in CBNRM different from that set out in policy? I have approached these questions through an investigation of the prescribed and performed governance systems within Tanzanian CBNRM, comparing and contrasting the wildlife and forestry sectors. My specific research questions are detailed in Table 1.1. These questions are split into three categories that correspond with the empirical chapters presented in the thesis. In chapter 5, I address the first of the questions listed above through a discussion of the processes of policy reform that introduced CBNRM in Tanzania. In this chapter I compare the formal, prescribed governance systems for CBNRM in the forestry and wildlife sectors and discuss the forces that shaped policy processes in these two sectors in different ways. My investigation of the second and third questions shown above is split between chapters 6 and 7. In chapter 6, I begin to examine the performance of policy through an analysis of power systems within CBNRM and the struggles and transformations taking place around these. I utilise the power structures in place within CBNRM and the struggles taking place around it to investigate who benefits from CBNRM and what inequalities are witnessed in the distribution of benefits within both case studies. I also outline the configurations of power that were witnessed in the case studies, and the differences between these and the arrangement of power set out in the prescribed governance system. Chapter 7 continues this examination of the performance of policy by addressing the informal aspects of power that form part of the governance system in reality, but are hidden from view. In this final empirical chapter I discuss the integration of CBNRM into local systems of power, and outline the deviations from policy and the prescribed governance system that took place in the case studies. In this chapter I ask what processes of corruption and rule-breaking are taking place in the case studies, and what processes underlie them. I focus on understanding 14 the socio-political processes that permit and maintain both differences between the prescribed and performed governance systems, and the unequal distribution of benefits from CBNRM. Thesis Chapter Research Topic and Questions to be Addressed 5 Policy Processes in Tanzanian CBNRM • How have the conservation policies underlying WMAs and CBFM developed in Tanzania? • Why have the policies developed in different ways? • What discursive processes did this development entail? • How have wildlife and forest resources become politicised, and in what different ways? • What are the formal governance structures set out by the policies and how do these differ between the wildlife and forestry projects? • How are accountability, participation and democratic processes prescribed within these systems of governance? 6 The Performance of Policy I: Power Systems in Governance and Politics of Scale • How is power devolved within WMAs and CBFM? • How do the configurations of power differ between the WMA and VLFR • Which actors become powerful and how is power re-spatialised from previous arrangements? • Who is identified as benefiting the most/least from the WMA/VLFR project? • How and why is power being contested and struggled over within the case studies? • What power structures can be indentified in the performance of policy and how do these differ from those prescribed in policy? 7 The Performance of Policy II: Hidden Aspects of Governance and Participatory Spaces • How are the governance structures being shaped by existing power relationships within the case studies? • How do actors create and use influence and power within the projects and adjust power systems? • How do the informal aspects of the performed governance system differ from that prescribed in policy? • What opportunities exist to challenge the formal and informal systems of power? • How is corruption taking place within the case studies? • How are such processes maintained? Table 1.1: Research Questions Addressed in this Thesis by Chapter 15 I have approached these questions through a study of power in Tanzanian CBNRM, positioning this as a central and linking theme that runs throughout these questions and the empirical chapters that follow. In the following chapter I set out a theoretical framework to address this research agenda within this thesis. This framework is necessarily broad, drawing on areas of study that are rooted in sociology, human geography, policy studies, conservation and development. I then draw from these further disparate strands of study and theory to consider the intersection of power with policy and its implementation in the Tanzanian context of CBNRM. I have framed my research as fundamentally a political ecology approach that investigates the politics of natural resource management within CBNRM. Political ecology forms the umbrella that brings together these different areas of scholarship, therefore (see 2.5). Political ecology has always been focused upon “probing how the politics of access to and control over land and resources were related to environmental change” and argues that social and political factors, not technical or managerial issues, lie at the core of ecological problems (Neumann, 2005). Political ecology has politics at its heart, therefore and has developed since the 1970s in response to the perceived apolitical nature of environmental research, driving a new research field that adopts the underlying assumption that “politics and environment are everywhere thoroughly interconnected” (Bryant, 1998: 82). 1.5 Thesis Structure Following this introductory chapter, the thesis is set out to introduce the theoretical framework applied in this study in chapter 2, followed by a discussion of the research context and specifics of forest and wildlife conservation in the United Republic of Tanzania in chapter 3. The methodology applied in this study is outlined in chapter 4. Chapters 5, 6 and 7 are the empirical chapters presenting my own research findings and addressing the research questions shown in Table 1.1. Finally, chapter 8 is a summary and discussion of the research findings and conclusions of the thesis. Chapter 2: Theoretical Context 16 In this chapter I explore the theoretical framework that I use to investigate the research agenda and questions set out in chapter 1, drawing on a wide body of literature to investigate the socio-politics of Community-Based Natural Resource Management (CBNRM) in Tanzania. This study is framed around the emergence and experience of CBNRM, and the challenges that have surfaced as a result of these. Whilst there is considerable interest in the promises of CBNRM and its integration of conservation and development, there has been widespread disappointment at the unexpected results that have not lived up to the predicted ‘win-win’ scenarios, as discussed in section 1.3. My theoretical approach places politics and power at the centre of devolved environmental management, and aims to add depth of understanding to the processes taking place within Tanzanian CBNRM, rather than a black and white perception of CBNRM in which it either works or does not. This has been highlighted as an area for further research (Adams and Hulme, 2001a; Berkes, 2004), and it is to these calls for a deeper, more socio-politically orientated approach that this thesis is orientated. I have framed my research as a political ecology investigation, which is discussed in the final section of this chapter (2.5). Within this approach, I draw on many different areas of literature and theory. I trace the socio-political processes within CBNRM from the initiation of policy reform through to the implementation of policy and its outcomes and complexities. In section 2.2, I discuss the literature from policy processes, on which I draw in chapter 5, to gain insight into the discursive processes that shaped CBNRM in Tanzania, and the sectoral policies that underlie it. I use this literature to address the question of how and why the governance systems prescribed in CBNRM policy differ between the forest and wildlife sectors, and the ways in which power is devolved in distinct ways. In section 2.3 and 2.4 I discuss my theoretical framing of both power and scale. These are critical features of a political ecology approach (Bryant and Bailey, 1997), and are closely linked to the discussion of policy processes, where power is an often under-theorised aspect of policy reform, and where the scalar configurations of power prescribed in policy form the basis of political struggles for power. In chapter 6, I draw on the politics of scales literature in my investigation of the reality, or performance, of CBNRM in Tanzania and the ways in which people are engaged in struggles to benefit from CBNRM, producing patterns of unequal benefits from CBNRM, and creating distinctions between the performed and prescribed governance systems. 17 In chapter 1 I argued for a socio-political research agenda and, in section 2.1, I begin this chapter with a discussion of the acknowledged complexities and theoretical developments associated with devolved environmental management and CBNRM. I outline a theoretical framework for devolved environmental management that firstly emphasises the politics and intricacies of power devolution from the state to the local level. Secondly, I draw on literature from participation, institutions and common pool resource theory to propose a concept of community within CBNRM that avoids the ‘local trap’ (Born and Purcell, 2006) and addresses the micro-politics of community management, participation and decision- making. Finally, in section 2.1.3 I discuss the importance of incorporating theoretical insights from the literature of African politics, particularly with respect to the inclusion of the informal realm in studies of governance. This is an important feature of the discussion in chapter 7, where I address unacknowledged aspects of power within Tanzanian CBNRM and further socio-political processes taking place around the governance system. 2.1 Devolved Environmental Management 2.1.1 Devolved Environmental Management: Power in CBNRM As discussed in the preceding chapter, a central tenet of CBNRM is the devolution of power2 to the local level, and the initiation of community-level decision-making for natural resource management. This form of power devolution generally involves the conferment by the state of authority3 to local organisations. The local level becomes authorised in the management of natural resources and receipt of benefits from this therefore. My focus is specifically upon programmes of devolution, and specifically the transfer of powers to community organisations. This is a component of the broader term decentralisation, which involves the transfer of powers to lower levels of political administrative and territorial hierarchy (Shackleton and Campbell, 2001; Larson and Soto, 2008). I use the term decentralisation as an umbrella under which several different processes occur, including democratic decentralisation4, deconcentration5 and devolution. Devolution is distinct from decentralisation in the level and types of powers transferred, with devolution 2 Power is a complex term that is discussed further in section 2.3 3 Authority refers to compliance based on the recognised legitimacy of the decision-maker (Lukes, 2005). 4 The transfer of powers to lower level governments, for example district or regional level administrations, or to regional autonomous governments (Larson & Soto, 2008). 5 The transfer of powers from central ministries to branch offices outside of the capital or administrative centre (Larson & Soto, 2008). 18 representing the fullest and least restrictive type of power transfer to a community group (Crook, 2003; Smoke, 2003). Alongside the arguments made for the paradigm shift to community conservation and the integration of conservation and development objectives discussed in chapter 1, the justifications for the devolution of power to the local level in the natural resources sector are similarly based on the perception of local people less as a problem, and more as a potential solution (Larson and Soto, 2008). The handing over of authorisation to manage natural resources and channelling benefits from this to the local level was predicted to reduce costs for the state, whilst at the same time improving compliance with natural resource rules and creating incentives for long-term sustainable management of resources (Sundar, 2001; Andersson and Ostrom, 2004; Brockington, 2008). There are two key assumptions within this conceptualisation of power devolution in natural resource management that have been crucial to the mixed results and disappointments with CBNRM discussed in chapter 1: the meaning and types of power in power devolution; and the assumption that local support can be fostered through providing incentives for conservation (Gibson and Marks, 1995). Addressing these assumptions focuses research on the nature and politics of power devolution within specific CBNRM policies Multiple examples across the community conservation spectrum have indicated that power devolution, even when central to objectives, can take many forms, incorporating different levels and types of power and associated management rights at the community level (Agrawal and Ostrom, 2001). The restriction of community management and ownership rights to that of implementation of an externally-designed system is seen as one of the biggest barriers to successful CBNRM (Agrawal and Gibson, 1999; Charnley and Poe, 2007). Such restrictions can result in communities receiving handouts from CBNRM activities, instead of achieving meaningful control and empowerment at the local level, and such handouts may result in the supplementation of livelihoods, rather than the instigation of long term development (Murphree, 1996; Murombedzi, 2003). Reviews of CAMPFIRE have often been critical on exactly this point, claiming that a lack of proprietorship rights for communities, and the lack of devolution to institutions at the village level results in income handouts rather than local management (Murphree, 1996; Murombedzi, 2003). Similarly, studies of community conservation projects in Tanzania have concluded that they fail to create empowered community managers and frequently result in handouts to local communities, who are passive recipients in a government project (Songorwa et al., 2000; 19 Brockington, 2002). An assessment of CBNRM initiatives therefore needs to carefully consider the ways in which power is devolved to local communities, the management rights they hold and the decision-making processes that take place at the local level. Such an analysis can usefully draw on the analysis of different ‘bundles of rights6’ associated with communally managed resources outlined by Agrawal & Ostrom (2001): • Withdrawal rights that provide right of entry and attainment of resources or products • Management rights that are characterised by the ability to regulate use of natural resources and to make decisions to transform the resources and make improvements • Exclusion rights that allow the rights-holders to decide and implement a system of rules and eligibility for access and use of natural resources • Alienation rights that refer to the rights to sell or lease resources or the rights to manage those resources (Campbell, 2007). Rights of exclusion, associated with proprietorship of land is advocated as the minimum requirement for successful community management (see also 1.3; Charnley and Poe, 2007). The transfer of authority to community organisations is clearly a sensitive matter. Just as the annexation of land and natural resources by the state has been described as an essential part of state-making (Ferguson, 1990; Neumann, 2001; Crook, 2003; Neumann, 2004b), states are often reluctant to relinquish control over these resources, and may curb devolution wherever possible (Agrawal and Ostrom, 2001). This reluctance is evident in Nepal, where two distinct types of community forestry have developed (Thoms, 2008). In areas where the value of forestry resources is low, the devolution of power to local communities has been celebrated as full and meaningful, whereas more valuable timber resources in the highlands remain under state ownership and control with compensation and alternative livelihood provisions for communities instead of devolved management (Agrawal and Ostrom, 2001). The unwillingness of states to devolve power appears strongly linked to the political and economic value of the resources in question, therefore. Several examples of devolved natural resource management have indicated that recentralisation can actually take place within a programme of decentralisation or devolution (e.g. Pulhin and Dressler, 2009). ‘Recentralising whilst decentralising’ is a process of the extension of state control which occurs through the restriction of power that is devolved to the local level, or through the devolution of power to local level institutions that serve the interests of the central authorities 6 Categories of property rights and operational use rights. See Appendix 1 20 (Olowu, 1999; Schafer and Bell, 2002; Ribot et al., 2006). This may take place through the procedures that communities are required to complete in order to receive state approval of their rights to manage natural resources, for example through compulsory attendance of approved training schemes, and a lack of recognition of other forms of knowledge (Goldman, 2003). The resulting form of CBNRM is one where the local community is restricted to the role of beneficiaries within a discourse of empowerment (Goldman, 2003). The second assumption that I wish to address is that of local support. This is often assumed to be both critical to the success of CBNRM, and to be created through the construction of appropriate incentive structures to prioritise sustainable management of natural resources. However, Gibson & Marks (1995) argue that local support, seen as crucial to successful community conservation, is not necessarily created in this way. Instead they highlight the importance of different meanings of resources and conservation in different contexts and for communities, and point to community power issues as crucial to understanding local attitudes towards conservation, and the outcomes of projects. Similarly, in Songorwa et al.’s (2000) study in Tanzania they argue that local support for community wildlife management initiatives cannot be assumed, and that the nature of wildlife resources and the way they are perceived within local communities may make them inherently incompatible with rural development strategies. CBNRM is being implemented in a reality in which the support of the state to devolve power, but also the willingness of the community to adopt the ideology of CBNRM and responsibilities it involves, must not be assumed. Empowerment of local communities to manage resources as envisaged by advocates of CBNRM necessarily involves a large risk that such communities choose to manage those resources in a way that is not in line with state priorities and models, and potentially not in line with conservation objectives (Redford and Sanderson, 2000). Murombedzi’s (2001) analysis of CAMPFIRE revenues showed disparities in the priorities of central authorities, conservation organisations and local communities, with communities preferring to use funds they accumulated for development opportunities, rather than investing in wildlife management. Empowering the community to set their own agenda and make their own management decisions is seen as a crucial component of meaningful power devolution, and local autonomy is critical to the construction of natural resource management systems that are adapted and appropriate to the local context (Agrawal and Ostrom, 2001). However, the risks involved with devolution has 21 led many states to recentralise and include limitations on the decision-making powers of the communities involved (Schlager and Ostrom, 1993; Behera and Engel, 2006). The uncertainties involved with community autonomy over conservation and development decisions discussed above may also de-incentivise power devolution within CBNRM. Both the assumptions discussed in this section hide powerful incentives for the state to retain ownership and control of natural resource management. CBNRM is shaped by these factors, which can account for the discrepancies between the CBNRM that is envisaged and that which emerges in practice, both in terms of the ways in which power is devolved to a local community to manage its natural resources, but also in the ways that a community chooses to do so. The socio-politics of CBNRM must address these issues of the intricacies of power devolution. Important questions for research include the nature of the powers that are devolved to local communities, the shifts in roles for both the state and the local level that occur within it, the restrictions that are placed upon devolution and community autonomy in decision-making, and the ways in which power may be recentralised through CBNRM. The prescribed governance system needs to be analysed from a power devolution perspective, therefore, to consider these complexities and their impacts on the authorisation of communities to manage natural resources and benefit from this. Secondly, it is clear that communities do not necessarily share the same priorities for CBNRM decision-making as other managers, and investigating the decision-making processes within communities practising CBNRM is vital to further understanding of how the prescribed governance system takes shape in reality, to create the performed governance system. 2.1.2 Devolved Environmental Management: Communities in CBNRM The restriction of powers that are devolved to local communities, as discussed above, has a corresponding impact upon community-level participation and engagement with CBNRM. The same processes by which states may retain and recentralise power over natural resources through administrative and legislative structures (see Ribot, 1999), shapes the space that is available for community participation in such projects. Analysis of participation in community conservation projects, and even within devolution-focused CBNRM projects has commonly identified a lack of meaningful participation as a result of these processes (Zacharia and Kaihula, 2001; Nelson et al., 2007). However, the dynamics of participation in CBNRM are not solely controlled by the powers that are devolved to the community. As discussed above, it cannot be assumed that communities that participate in CBNRM will 22 prioritise conservation objectives or act in the same way as if a state, NGO or private enterprise were managing those resources. Songorwa (1999) found that community priorities around the Selous Game Reserve in Tanzania were centred on access to meat, which drove participation in community conservation projects. He argues that rather than the incentive structure set out in the project, it was this access that determined people’s perception of success and benefits from the project, and was largely responsible for people’s engagement, trust in the project and their support for it (Songorwa, 1999). Within the paradigm of community conservation and development, participation and empowerment have been large areas of study (Cleaver, 2001; Cooke and Kothari, 2001; Mosse, 2001; Matta et al., 2005). Whilst participation is seen as a medium for the articulation of social and political needs in a context of state hegemony7 (Nagothu, 2001; Kelsall and Mercer, 2003; Hargreaves et al., 2007), the theoretical underpinning to such arguments is one that considers participation through ‘conscientisation’8 (Midgley et al., 1986; Rose, 1997; Mohan, 2001; Carr, 2003). Similarly, participation is often assumed to take place within a context of ‘ideal free speech’ (Habermas, 1994 in Puri and Sahay, 2003: 183) and deliberative democracy where “democratic will-formation draws its legitimating force…from the communicative presuppositions that allow better arguments to come into play in various forms of deliberation and from the procedures that secure fair bargaining processes” (Habermas 1996a in Kapoor, 2002: 105). Opening up the opportunity for participation and empowerment through the implementation of CBNRM, for example, is assumed to allow for fair, democratic participation by all those within a community, resulting in efficient and beneficial management of natural resources. In reality, participation does not appear to take place like this, varies both spatio-temporally and socio-politically, and involves costs for participants (Rose, 1997; Williams et al., 2003; Williams, 2004b; Williams, 2004a; Kapoor, 2005). Cooke & Kothari (2001) famously wrote about participation as tyrannical, rather than emancipatory. Their criticisms of participation emphasise that it is a socio-political process, not a neutral event, and participation holds the potential for negative as well as positive outcomes (Cooke and Kothari, 2001). The participatory role of communities in CBNRM 7 A concept discussed by Gramsci as ruling class domination that is both intellectual as well as a material force in society, and whereby the working class adopt a psychological state of acceptance and internalisation of the socio-political order (Lukes, 2005). 8 Describing empowerment achieved through self-reflection and the production of knowledge. The concept was made famous by Freire (1970). 23 must acknowledge that participation and empowerment do not take place evenly across a community, but take place within an existing context of power relations (Nelson and Wright, 1995; Mohan and Stokke, 2000). Such power relations can govern those who are able to speak at meetings, those who are able to gain positions within the management system and those who are listened-to in decision-making processes (Cleaver, 1999; Cleaver, 2001). The identification of elite capture across the spectrum of community conservation approaches illustrates how this process is taking place and having material consequences for the communities and individuals involved. In their study of Joint Forest Management in India, Hildyard et al. (2001) concluded that the outcomes of elite capture, gender inequalities in participation, intra-community differential outcomes and the importance of socio-economic and cultural groups seen in the project reflect deep-seated, structural, and often unintentional flaws in participatory theory and implementation (Hildyard et al., 2001). Critics of the participation movement argue that its socio-political nature is hidden and it is riddled with localist discourse, (Hickey and Mohan, 2004b; Williams, 2004b). Communities are often conceptualised according to Tonnie’s idea of gemeinschaft and ideas of a mythical community consisting of small spatial units, and homogenous social structures that are based upon shared norms (Agrawal and Gibson, 1999; Nunan, 2006). Community has become a vague term that is used to describe a group of people who are thought to intrinsically possess high levels of social capital9, capacity to participate and ability to make decisions in the best interests of the group as a whole. This essentialisation of the community has created a privileged solution within a wider trend of ‘localism’, ‘enchantment with community’, or the ‘local trap’ (Mohan and Stokke, 2000; Adams and Hulme, 2001a; Mercer et al., 2003; Purcell and Brown, 2005; Born and Purcell, 2006). These views romanticise and simplify the community into a harmonious, organic and virtuous construct, wherein the nuances of reality are simplified and masked from view by rhetoric (Agrawal and Gibson, 1999; Mohan and Stokke, 2000; Blaikie and Springate-Baginski, 2006). Such characteristics may be present in some cases (Jones, 2006), but cannot be conceived as intrinsic to local communities, which are generally complex and messy groups of people representing different axes of power and multiple understandings and agendas with regards to natural resources and their management (Cleaver, 1999; Kapoor, 2002; Kapoor, 2005). 9 This is a complex term that is widely debated, but Putnam’s (1993) famous discussion described social capital as a feature of civil society in which“networks of trust and shared norms in society facilitate cooperation for mutual benefit” (Adams, 2009: 380). 24 In an attempt to recognise the complexity of communities, the environment and the relationships between the two, Leach et al. (1997b) argued that politics must be at the very heart of the approach to understanding community conservation and its varied outcomes. They introduced an institutional perspective called ‘environmental entitlements’ (Leach et al., 1997b; Leach et al., 1999), which instead of seeing communities as harmonious, started from a viewpoint that they are socially differentiated and highly diverse, imbued with power relations around boundaries of gender, age, wealth and social identity. Instead of seeing community management of natural resources as consensual and based upon shared beliefs and priorities, their approach conceptualises such processes as taking place through negotiation and bargaining, usually involving conflict and struggle (Leach et al., 1997b). The entitlements framework built upon the arguments of Amartya Sen (1981; Sen and Press, 1982), who argued that institutions governing access to food were responsible for creating famine situations, rather than an actual shortage of food. Human behaviour in these processes of decision-making, management and access to natural resources is governed by institutions, defined as “regularised patterns of behaviour between individuals and groups in society”, commonly referred to as the ‘rules of the game’ (see also North, 1990; Mearns 1995a in Leach et al., 1999: 226). Institutions matter because there is a cost (financial or social) involved with violating them, and individuals and groups must abide by these rules in order to be playing the game legitimately therefore (North, 1990; Leach et al., 1997b). Importantly, Leach et al. (1999: 12) recognised that institutions are both formal (enforced by an endogenous organisation, e.g. laws) and informal (“upheld by mutual agreement among the actors involved, or by relations of power and authority between them”, e.g. social norms). Institutions are also not fixed, but can shift over both space and time as actors engage in negotiation and struggle (Kepe and Cousins, 2001; Nunan, 2006). Institutions reflect values and norms, but they are only fully understood in the context of existing power systems as they are simultaneously a reflection of and are upheld by power relations (North, 1990; Leach et al., 1999). Here there is an important link to the theorisation of power discussed in 2.3; power, and hence institutions, are not static features within a structural system of domination, but are socially-constructed and negotiated through constant struggle. Closely linked to the concept of institutions, a large body of work has focused on investigating the capacity of community organisations to manage common-pool resources10. 10 Such resources are defined as subtractable natural or human-made resources that are available for use by more than one person, and where degradation may be caused by over-use (Ostrom, 2002). 25 Elinor Ostrom’s many works on this subject, including Governing the Commons and The Drama of the Commons (Ostrom, 1987; 1990; Ostrom, 1992; see also Ostrom et al., 1999; 2002) set out a strong argument for the successful management of such resources under group property rights, and developed a set of design principles outlining the characteristics under which success was witnessed. Ostrom, amongst many authors, has created a strong narrative in support of CBNRM, arguing against powerful narratives of the tragedy of the commons and the idea that rational actor theories can fully explain community management, to show that the paradigm shift to approaches such as CBNRM was well-founded, but they also highlighted the complexities of this kind of management, including the importance of the wider socio-economic context (Schlager and Ostrom, 1993), and the integration of community management into a supportive governance structure across multiple levels (Ostrom, 1992). There are close links between the common pool resource literature and studies of the complexities of power devolution discussed in 2.1.1 as both argue that the types of powers held by or devolved to a community group are a vital characteristic of the management of natural resources. Ostrom (1990) and Agrawal & Ostrom (2001) argued that community authority and autonomy lies as the heart of many of the design principles identified, including locally-adapted rules of use, meaningful participation in decision-making (see also Hayes, 2006), and acceptable mechanisms of conflict resolution and sanctions for violations of community rules (Ostrom, 1990). Common-pool resource theorists have added to debates about CBNRM significantly by outlining the complexities and capacities of communal management of natural resources, but the literature on common pool resources does not deal sufficiently with power to provide a framework for investigation of the socio-political processes at work within such management (see Agrawal, 2005). I acknowledge the important contributions made by both the institutions and common pool resource literatures, and I draw on the broad themes that they highlight within my own study, which is centred around a more serious investigation of power within the socio-political processes taking place in both power devolution and community management of natural resources. In this thesis I focus on power, and I utilise institutions as a lens through which to study the power relations at work. To this end, I argue that it is critical to investigate both the formal institutions in place within CBNRM, as set out in the prescribed governance system, and how these combine with the informal institutions that are present to create the performed governance system. The role of informal institutions is especially important in Africa, as 26 power is concentrated in these institutions (Hyden, 2008). In particular, informal institutional arrangements may provide valuable insight into the processes underlying the outcomes of elite capture and unequal distribution of benefits from CBNRM (Songorwa, 1999; Songorwa et al., 2000; Igoe and Croucher, 2007). However, an in-depth investigation of the informal institutions in place within a performed governance structure is a particular research gap in CBNRM and community conservation research (Nygren, 2005; Bawa et al., 2007; Nelson and Blomley, 2007; Nelson et al., 2007; Jones et al., 2008). A key aspect of CBNRM that has been highlighted by the literatures on devolution, common- pool resource and African politics is the role of the informal realm. Power to manage natural resources is made up of both de jure rights and responsibilities (formal institutions and rules e.g. laws) and by informal rights and responsibilities, which may be referred to as customary rights, informal institutions, norms of access or de facto rules in use (Robbins, 2000). Hyden (2008: 12) argues that “any study of politics in Africa ... involves by definition the role of informal institutions. Together with formal ones, they shape the articulation of power”. De facto rules in natural resource management are moulded around local contexts of power and may operate alongside de jure rules, but they may also contradict these rules. Studies of the role of informal institutions in communal management of natural resources have also produced a large literature on the extra-legal aspects of governance and the ways in which rules are broken. Ostrom (1990) maintains that the de jure and de facto rules of use or formal and informal institutions do not stand in isolation, but are combined to produce the ‘operational rules of use’, which is an important component of the performed governance system which I investigate. African politics literature in particular has looked in detail at the illegal aspects of governance being performed, including rampant corruption and the systems of neopatrimonialism operating within the continent (Olivier De Sardan, 1999; Robbins, 2000). In the next section, I address these concerns and ask how such issues should be considered with regard to CBNRM and its performance. 2.1.3 Studying the Informal: Disorder, Corruption and Neopatrimonialism “A focus on power is particularly relevant in countries like Tanzania where it is more inherent in informal than in formal institutions” (Hyden, 2005: 1) 27 Chabal & Daloz (1999: xix) propose a theory of African politics that functions through disorder, a “condition which offers opportunities for those who know how to play [the] system”. This disorder does not represent irrationality; “it is merely to make explicit the observation that political action operates rationally, but largely in the realm of the informal, uncodified and unpoliced” (Chabal & Daloz, 1999: xix). They argue that this has developed in Africa through the informalisation of politics, whereby the state and civil society are not separate, but straddle each other (Chabal and Daloz, 1999). The informal economy remains central to both social and political life, and the basic unit of reference remains family and kin- based (Chabal and Daloz, 1999). The importance of the informal realm and a moral economy has previously been discussed with reference to Tanzania as an ‘economy of affection’11 (Hyden, 1980; 2005). The premise of the concept is that the economic system is at least partly based on provision through non-economic institutions, for example kinship and reciprocity (Booth, 1994). The theorisation of the role of the informal realm by Chabal & Daloz (1999) provides a frame with which to integrate both informal institutions and unofficial forms of power into the socio-political processes of natural resource management. I employ their theory of disorder and conceptualisation of power in African societies to frame my investigation of power and how this is connected to rule-breaking and corruption, which are a focus of the discussion of hidden aspects of the performed governance system in chapter 7. I utilise Chabal & Daloz’s (1999) theory of the informal nature of African social and political systems, resting on a moral economy and neopatrimonial systems of power, to examine rule-breaking and corruption as socio-political processes, rather than poor implementation. Here I discuss this view of corruption and rule breaking and the neopatrimonial types of power which support it. “Corruption is not the absence of state institutions, but the presence of differing institutions, which vie for legitimacy and trust amongst diverse players within both the state and civil society” (Robbins, 2000: 426) There is a large literature on corruption, but little consensus as to how to define it. Most discussions employ a Western view of such processes and utilise definitions that are infused 11 The concept has strong links to the gemeinschaft perception of communities (Booth, 1994) and must be considered in light of the discussion of community dynamics in 2.1.2. 28 with morality (Theobald, 1990). Salient characteristics of common definitions12 place it in the realms of politics and public affairs and rely on deviance from the rules that are set out by states and found in laws. I adopt a definition of corruption that recognises it as a form of rule-breaking and illicit behaviour, but removes the label of illegitimacy from such practices and focuses on the social context of corruption. Numerous authors have outlined typologies of corruption that attempt to address its characteristics across different scales, rather than a uniform definition. Olivier de Sardan (1999) argues that it is both important to distinguish between the different types of corruption13 and the levels at which corruption is practised. Corruption literature tends to focus on political corruption (see Theobald, 1990; Rose- Ackerman, 1999); however, my interest is in the lower levels of corruption, the petty forms of corruption that pervade normal life, rather than high level, more economically significant corruption usually associated with governments (Olivier De Sardan, 1999). Leaving behind the Western notions of corruption, such processes have more recently been viewed in a social context (Chabal and Daloz, 1999), with efforts to outline the social, cultural and psychological foundations of corruption or, as Olivier de Sardan (1999) phrases it, how these ‘communicate with corruption’. Secondly, efforts should focus on understanding the impacts of corrupt practice across levels of society (Chabal and Daloz, 1999). Corruption must, therefore, be seen as part of a moral economy of production, without falling into the trap of saying that it is a cultural inevitability (Chabal and Daloz, 1999; Olivier de Sardan, 2008). There are important questions as to why corrupt practice is generally upheld within African societies. It is simultaneously decried and banalised, anti- corruption discourse is largely rhetoric and social acceptance is high (Chabal and Daloz, 1999; Olivier De Sardan, 1999). Pepinsky (1992: 27) summarised such attitudes by arguing that “there is nothing wrong with being reasonably corrupt in a corrupt system”. I consider corruption not as a driving force behind the failure of CBNRM, but as a hidden socio- political process taking place within it that requires further understanding. My aims concerning corruption are three-fold: Firstly, to investigate the socio-political processes of corruption and how it is taking place within Tanzanian CBNRM; secondly to consider the 12 Huntingdon (1968 in Theobald, 1990: 2) defines it as “behaviour of public officials which deviates from accepted norms in order to serve private ends”. Nye’s famous definitions claims that corruption is “Behaviour which deviates from the formal duties of a public role because of private-regarding (personal, close family, private clique) pecuniary or status gains; or violates rules against the exercise of certain types of private-regarding influence” (1967 in Olivier de Sardan, 1999: 27). 13 For example the abuse of power, nepotism, embezzlement and misappropriation of funds, influence pedalling and insider trading. 29 ways in which corruption mechanisms and power systems are being altered by the implementation of CBNRM, as the scaling-down of corruption to local communities is often associated with devolved natural resource management (Harriss-White and White, 1996; Robbins, 2000; Véron et al., 2006). Finally, I wish to further address the issue of power within corruption and to ask how corruption is facilitated and why it goes largely unchallenged and accepted. Corruption is essentially a form of rule-breaking or departure from the prescribed governance system. I intend to investigate such processes, drawing on the concepts of the role of informal institutions and the social norms and rules in use to identify additional aspects of the governance system and power relations that are in place within local communities undertaking CBNRM. The pervasive forms of corruption that I focus on are closely tied to the nature of social power in African, and particularly to the concept of neopatrimonialism, which Kelsall (2011) agues dominates the political economies of Africa14. Neopatrimonialism can be defined as a “system...held together by the personal distribution of material resources and perks...distributed and consumed as if they were the private property of the ruler and/or his staff” (Kelsall, 2011: 77). The term originates from patrimonialism, which Weber defined as an ideal type of rule where authority is constructed from loyal personal relations between a leader and his staff (Kelsall, 2011). The term neopatrimonialism refers to “a political economy in which this basic authority system is combined with, or exists behind some formal, impersonal elements of governance” (Kelsall, 2011: 77). Bayart (1993) famously described such a system of authority as ‘the politics of the belly’, arguing not only that neopatrimonial systems dominate in Africa, but also that they are a quest for hegemony by those in power, and that such processes are socially accepted and normalised by the wider population (see also 2.3). The dominance of the informal realm and relationship-based institutions in African socio- political life facilitates the expansion of patron-client relations, especially as social and political power is constructed through the network of supporters surrounding an individual15 (Chabal and Daloz, 1999). To maintain this support requires relationships that are based on reciprocity, however, and powerful individuals are required to use their power and the 14 Clientelism and rent-seeking can be thought of as components within a neopatrimonial system (Kelsall, 2011). 15 This kind of social power is similar to the concept of the ‘big man’, characterised as a charismatic individual, generous and with significant personal power, which has been discussed particularly with reference to Melanesia (Sahlins, 1963). 30 resources generated by it to purchase people’s affection, thus maintaining and expanding their network of support (Chabal and Daloz, 1999). A system of patrons and followers is created wherein the power of the patron rests on his/her ability to redistribute resources to the satisfaction of his clients (Chabal and Daloz, 1999). This system of redistribution is an essential characteristic of neopatrimonialism. Authority over natural resources is a significant source of power, and the politics that surround the management of natural resources demonstrate the instrumental use of this authority to protect and expand social power (Chabal and Daloz, 1999). It is critical to examine CBNRM within a context of the political nature of natural resources, and their intrinsic relationship to power. The relationship between neopatrimonialism and corruption is also important. Neopatrimonialism allows powerful individuals to supplement their incomes through corruption (Kelsall, 2011), but these powerful individuals must also strive to maintain their positions of power through the redistribution of natural resources, the benefits derived from access to them, and their management. Through these processes, the prescribed governance system will clearly be altered in performance, as the formal institutions of governance meet and are integrated with local systems of power and informal institutions surrounding this. These concepts have become so central to theories of African politics that current policy thinking within development has begun to re-evaluate the prevalent forms of governance in Africa in light of these differences in the basis of power and the informal institutional context, and to question the applicability of Western models of development and ‘good governance’ in such contexts. The African Power and Politics Programme16, for example, has begun to research alternatives to the standard developmental governance in Africa, drawing on the distinctions discussed above to assess the potential of ‘good enough governance17’ (Grindle, 2004; 2007) in an approach termed ‘going with the grain’ (Booth, 2011; Kelsall, 2011). This approach includes the exploration of the potential “harnessing of neo-patrimonialism for development ends”, termed ‘developmental patrimonialism’ (Kelsall, 2011: 76). Here there is a close link to the argument made by Chabal & Daloz (1999) that disorder is not necessarily irrational; neopatrimonialism is a different system, and there are glaring issues of inequality that are associated with it (see Bayart, 1993), but it can also be 16 See www.institutions-africa.org, accessed 15/6/2012. 17 “Good enough governance directs attention to considerations of the minimal conditions of governance necessary to allow political and economic development to occur” (Grindle, 2004: 526). 31 conceptualised as a legitimate upholding of socially accepted norms (Hyden, 2008). My argument is that, setting aside arguments about whether it is right or not, it is crucial to investigate the workings of informal institutions, the basis of power and the processes of neopatrimonialism at work within CBNRM, as they constitute a critical component of the performed governance system. The complexities of devolved natural resource management discussed in this section indicate four key areas for research, which are closely linked to the research questions and empirical chapters set out in Table 1.1. The first is the nature of power devolution, and the politics surrounding this process that work to resist devolution, restricting it and recentralising power to the state where possible. This is addressed in chapter 5 through an investigation of the policy processes leading to CBNRM and the different forms of power devolution this entails. The second refers to the ways in which communities manage natural resources under their own authority. It is clear that the powers devolved to a community are one part of the equation. However, a second part is taken up by the micro-politics of community management that must acknowledge that local communities are socially-differentiated groups of individuals, not a unanimous voice. Community management of natural resources, decision-making and individual participation takes place within this political setting, not in a political vacuum. Thirdly this political space of CBNRM is filled with institutions, both formal and informal, that represent the rules of the game which individuals are playing in CBNRM. My exploration of the performed governance system in chapters 6 and 7 addresses these two areas of research, focusing specifically on the issues of unequal distribution of benefits within CBNRM, the capture of benefits by certain groups, the political struggles taking place around natural resources and CBNRM, and the integration of local contexts of power into CBNRM. Finally, and closely linked to this, the fourth area for focus is an examination of the informal aspects of power operating within CBNRM. These are not represented in the prescribed governance structure set out in policy, but represent an important part of the governance system that is being performed in implementation. Chapter 7 focuses specifically on these hidden and informal aspects of power, and their role in the socio-political processes of rule-breaking and corruption. In the next four sections I further outline the theoretical framework that I draw on in my investigation of the socio-politics of CBNRM. I begin at the start of CBNRM in Tanzania with a discussion of the process of policy reform. I argue that the nature of power devolution, decisive in the reality of CBNRM, is a product of these policy processes and that 32 an investigation of the characteristics of this power devolution must begin with an understanding of the policy pathways adopted. Specifically I address the questions of how power is devolved within CBNRM in Tanzania, and identify the driving forces and power relations that shaped policy in to its current form. Research must focus on the politics that take place within and around policy “or, more specifically, how power shapes policy. The insertion of power as the missing variable in the link between institution and policy adds texture to the analysis” (Hyden, 2008: 11). 2.2 Policy Processes “Policy is the ghost in the machine – the force which breathes life and purpose into the machine of government and animates the otherwise dead hand of bureaucracy” Shore & Wright (1997: 5) Policy is an unclear concept. Is policy the strategy of government, the document which sets this out, the language of politicians and party manifestos, or the institutional mechanisms of decision-making? How does policy come about and what drives it in the directions it follows? All of these questions have occupied the minds of policy scholars at different times. Here I discuss the conceptualisation of policy, how this has changed over time and the lens through which I examine policy processes in this study. As a broad concept that is touched upon and claimed as its own by several disciplines and areas of study (John, 2003; Borzel, 2011), a history of policy studies and processes is not feasible. Instead, I aim to lay out some of the broad debates that have influenced the many changes in this field and to discuss the roots of the framework employed in this study. I follow the aims of Grindle & Thomas (1991) and Keeley & Scoones (2003) in being concerned primarily with the issue of policy reform, conceptualising this as an “interrelated processes of agenda setting, policy making and implementation” (Grindle and Thomas, 1991: 4), inherently tied to issues of power and the result of both “structural constraints but also opportunities for agency, action and change” (Keeley and Scoones, 2003: 39; see also 2.3.1). 2.2.1 Linear Policy and Elite Decision-Making Policy processes were traditionally envisaged as a top-down, elite-dominated system and controlled by rent-seeking and competition for control over resources (Thomas and Grindle, 1990; Baldus and Cauldwell, 2004). These instrumentalist views saw policy as a tool of the ‘few’ to control the wider ‘many’ by imposing rewards and sanctions, and considered policy 33 to be a rational, action-orientated instrument that developed in linear stages (Shore and Wright, 1997). These views were represented in the rational stages heuristic of policy by Hofferbert (1974), exemplifying the structural approaches that dominated the social sciences at this time. This linear view conceptualised policy as a problem-solving process in which problems are identified, actions is discussed, a decision is made concerning the best course of action, which is then implemented through policy (Fig. 2.1). Any problems or shortcomings with the resulting policy are seen as an issue of poor implementation, lack of resources or a lack of political will (Thomas and Grindle, 1990). This model removes the role of politics in decision-making, and fails to address the questions of how and why change occurs (Thomas and Grindle, 1990). Within this frame, as policy scholars began to address the issue of the factors that drive policy, several different theories of policy creation were suggested. Whilst some argued that policy was created in response to crises (e.g. Hirschman, 1963; see also Grindle and Thomas, 1991), Lindblom (1959; 1979) argued that policy was incremental, involving small adjustments and not sweeping revisions, which he famously described as a process of ‘muddling through’. Etzioni (1967; 1986) attempted to draw upon both these concepts and form a middle ground, in which he posited that policy was created through ‘mixed scanning’ whereby, a wide range of potential courses of action are considered and narrowed down to a few for deeper consideration. Within these views, policy studies began to make room for the role of actors, introducing the ideas of policy entrepreneurs (see 2.2.2.4), and the unpredictable nature of policy through trigger events, but none of these approaches said anything about the role of power in policy (Keeley and Scoones, 2003), which has become a major focus of an approach which frames policy as discourse18, and is inextricably tied to power (Keeley and Scoones, 2003). 18 Discourse is defined according to Hajer (1995: 44) as “a specific ensemble of ideas, concepts, and categorizations that are produced, reproduced, and transformed in a particular set of practices and through which meaning is given to physical and social realities”. 34 Fig. 2.1: The Linear Model of Policy Reform (Thomas and Grindle, 1990: 1165) A significant shift away from this linear model of policy-making has taken place across political science, policy scholars and the wider social sciences. Brock et al. (2001: 3) described the linear model as now widely perceived as a “dignified myth”, with attention focused much more upon policy as an interactive process, involving myriad actors and groups. The changes in how policy is perceived form part of wider changes within the social sciences, which have led to an ‘argumentative turn’ in policy studies (Brock et al., 2001), focused upon poststructuralist19 views of policy as discourse (Fischer and Forester, 1993). Mirroring shifts in theorisations of power to include the organisation of bias and discursive forms of power (see 2.3), the argumentative turn in policy studies has challenged the idea of science ‘speaking truth to power’ in policy making and re-framed debates about policy in terms of the discursive construction of issues and control over agenda setting, raising questions about how certain problems become politically important and the social construction and shaping of knowledge (Shore and Wright, 1997; Keeley and Scoones, 2003). 2.2.2 Policy Network and Policy Processes Views As part of the argumentative turn, policy is re-framed as co-constructed across space, through networks linking the global and the local, the study of which requires “understanding the interaction of networks and relationships, agency and practice, and knowledge and power 19 Poststucturalism is a reaction against deterministic structural interpretations within the social sciences which argues that “the relationship between society and culture is mediated through language” (Kitchin & Tate, 2000: 17). Postructuralism within political ecology has emphasised the roles of knowledge, power and discourse and the multiple interpretations of environmental issues that these create (Watts & Peet, 2004). 35 dynamics in particular contexts” (Keeley and Scoones, 2003: 4-5). Drawing on the work of Marsh & Rhodes (1992), the policy network is a meso-level concept which places emphasis on the relationships between actors, structures, decisions and institutions for policy outcomes (Bulkeley, 2000). Policy networks are comprised neither of all interested parties, nor of a privileged few, but “arise as interest groups gather around one or more government department in the hope of influencing policy, and are integrated into the policy-making process by the state” (Bulkeley, 2000: 728). This network approach to the study of policy is closely linked to the shift towards policy process analysis, rather than policy analysis, which in itself is rooted in contemporary theories of governance (Keeley and Scoones, 2003). These theories involve the substitution of rationalist views with ones that place power as a primary driver of the social world and attempt to integrate structural and actor-based influences (Arts and Tatenhove, 2004). These contemporary theories of governance have also focused on the roles of discourse and the social construction of knowledge over scientific truth (Arts and Tatenhove, 2004). The growth of the networks paradigm in governance is also intrinsically tied to the sweeping changes in the relationships between state and society, exemplified in the growth of neoliberal models of economics, politics and development (Grindle and Thomas, 1991). For some in the political sciences, these changes heralded a new ontological era of ‘governance without government’ (the Max-Planck School), whilst for others the change is methodological and signifies the beginning of a new way of conceptualising and analysing policy process; through networks (Borzel, 2011). The focus in this study is in line with the Anglo-American school which follows the latter of these views (Borzel, 1998). The network paradigm in policy studies is part of an interactive model of policy evolution, which recognises that policies change as they move through the processes involved in their creation (Grindle and Thomas, 1991), and attempts to integrate the roles of structure and agency to the study of such phenomena. This interactive model includes roles for wider groups and actors, beyond bureaucratic policy-makers, generating debate as to the role of ‘interest groups’ in policy (Grindle and Thomas, 1991). Marsh and Rhodes (1992) developed a typology of different networks, distinguished by the nature and pattern of interactions between actors to address these roles. They suggested a continuum of network types that range from a ‘policy community’, a strongly-institutionalised, close knit group of actors centred around a specific government department where shared views, beliefs and policy objectives can be identified (Bulkeley, 2000; Marsh and Smith, 2000). At the other end of the spectrum is the issue network, which constitutes a broader concept of multiple interest 36 groups and multiple views of issues, solutions and policy requirements (Bulkeley, 2000). Marsh & Rhodes (1992) argue that policy networks can involve both of these groups, but they are hierarchical, subject to power relations and the role of a wider issue network is often dependent upon the perceived threat posed to the dominant political and economic group represented in the policy community. In the next section, I discuss an approach to the study of policy processes that draws on the network paradigm and integrates the conceptualisation of both power and policy as discursive. Below I consider four key features and approaches within the network paradigm and how they inform the approach adopted in my research; epistemic communities, policy narratives and discourse coalitions, the advocacy coalition framework and the concept of policy space. This is not an exhaustive list as the policy processes field is very broad (see also Marsh and Smith, 2000; Arts and Tatenhove, 2004; Shanahan et al., 2011), but it includes some of the key concepts and frameworks which I draw on in this thesis. 2.2.2.1 Epistemic communities Haas (1992) builds on the idea that policy is created by an elite group who possess expert knowledge. He defines an epistemic community as “a network of professionals with recognized expertise and competence in a particular domain and an authoritative claim to policy-relevant knowledge within that domain or issue-area” (1992: 3). Key characteristics of the community are that they share normative and causal beliefs, which shape social action in the community, define issues and potential action to address it for desired outcomes, they share notions of validity, which define their criteria for the reification of specific types and sources of knowledge, and they share a ‘common policy enterprise’, whereby they address the perceived problem, using their expertise with the aim of improving the human condition (Haas, 1992). These epistemic communities are key in the discursive framing of issues in policy and in its nature as a political technology, through the privileged nature of their knowledge and information, and the power these communities hold to classify and define issues (Shore and Wright, 1997). Their influence is derived from the uncertainty faced by policy-makers addressing complex issues, which can be used to create space to frame problems and suggest appropriate action (Keeley and Scoones, 2003). The members of an epistemic community are usually highly-educated and well-connected individuals, often with access to international networks and policy-making groups, or may occupy key bureaucratic roles themselves (Keeley and Scoones, 2003). In relation to the policy network concept, an epistemic community is similar to the policy community as set out by Marsh & Rhodes 37 (1992), but incorporates a role for a wider group of actors at different levels, whereas the policy community is closely focused on government officials (Bulkeley, 2000). 2.2.2.2 Policy Narratives and Discourse coalitions Harriet Bulkeley (2000) reviews many theories and conceptualisations of policy network approaches, including those of policy narratives and discourse coalitions. She draws on the ideas developed by Maarten Hajer (Hajer, 1995; Hajer and Versteeg, 2005) and outlines the concept of policy narratives and discourse coalitions, arguing that these incorporate the interaction between actors and groups (epistemic or policy communities and issue networks), but also puts discourse at the centre of the approach (Bulkeley, 2000). She sets out a policy network (termed issue network in accordance with Marsh & Rhodes, 1992) which is made up of multiple groups of individuals connected through discourse coalitions (Bulkeley, 2000). These discourse coalitions exist alongside a policy community and actors compete to position the discourse to which they ascribe as the institutionalised discourse adopted by the policy community, and to influence policy processes, as shown in Fig. 2.2 (Bulkeley, 2000). Placing discourse at the centre of policy processes is in line with poststructural perspectives (Fischer and Forester, 1993) and brings together the theorisation of power discussed in 2.3, and the political ecology approach discussed in 2.5. Poststructuralist and political ecology studies of policy as discourse have often focused on the social construction of environmental problems, the neutral, scientific and rational language utilised in policy and the powerful groups that surround these uses of language in policy (Feindt and Oels, 2005; Goldman et al., 2011). Studies of policy as discourse also need to depict it not only as a means by which the world is created by actors, but also as a means of communication and negotiation (Bulkeley, 2000). The interaction, debates and conflicts between groups and actors with respect to Fig. 2.2: Policy Communities and Discourse Coalitions within an Issue Network (Bulkeley, 2000: 736). 38 policy reform is a crucial discursive process which both constitutes and is constituted by institutions (Bulkeley, 2000). Such an approach must also carefully consider the role of power to such interactions, the emergence of dominant discourses and the ways in which routinised use of discourse removes confrontation and uncertainty and reifies the power structure (Hajer, 1995). The concepts of policy narratives and discourse coalitions attempt to fulfil these requirements. Policy narratives are ways in which ‘policy-speak’ presents simplified, normative and powerful descriptions of a situation (Brock et al., 2001). Roe (1991) likened narrative structures to that of a folktale in which, Brock et al., (2001: 5) argue, “a crisis summons forth a hero, who battles against a series of obstacles, emerges triumphant, and everyone lives happily ever after”. Policy narratives as simple and evocative messages present a clear problem with a clear solution to which policy is orientated. These ideas have been taken up in a discussion of ‘story lines’, which act as the ‘discursive cement’ that facilitates and organises communication among actors (Hajer, 1995). “The key function of story-lines is that they suggest unity in the bewildering variety of separate discursive component parts of a problem” (Hajer, 1995: 56). Actors and structures are positioned around an issue according to such story-lines, on which they draw for their reasoning and understanding (rather than drawing on comprehensive discursive systems; Hajer, 1995). Attention to these story lines and the discourses they are a part of assists in investigating the clustering of knowledge and positioning of actors, how this results in coalitions of actors and control of the issues on the policy table through particular framings which fix and legitimise certain elements, whilst problematising alternatives (Hajer, 1995). Through discourse coalitions actors engage in politics, which is described as a struggle for discursive hegemony (Hajer, 1995). The concept can be used to investigate how actors are involved in argumentative power struggles in which they not only try to align others with their own particular framings, but also try to position other actors in specific ways for the benefit of those framings (Hajer, 1995). 2.2.2.3 The Advocacy Coalition Framework A major and competing contribution to the study of policy networks was Paul Sabatier and colleagues’ advocacy coalition framework (Sabatier, 1988; Sabatier, 1991; Jenkins-Smith and Sabatier, 1994). The framework claims that policy formation is driven by advocacy coalitions that compete within a policy subsystem, consisting of multiple actors and institutions interested in a particular policy area (Sabatier, 1988). Crucially, they define advocacy coalitions according to a shared set of beliefs and values, which they claim is a key 39 element of coalition formation and coherence (Schlager, 1995; Weible and Sabatier, 2005). This has in itself triggered vast areas of literature that investigates the nature of these beliefs, the different types of networks they may produce and the role that narratives may play (see Henry, 2011; Lahat, 2011; Matti and Sandström, 2011; Shanahan et al., 2011). Beliefs are theorised as being three-tiered: core-beliefs, similar to a worldview; policy core-beliefs, relating to the nature of problems and potential policy solutions; and secondary beliefs, which refers to the policy implementation strategy preferred (Bulkeley, 2000). The coalitions compete against one another in an attempt to align policy with their a priori beliefs through co-ordinated activity within the coalition and policy subsystem and through the use of ‘policy brokers’, who attempt to facilitate policy reform through minimising conflict (Sabatier, 1988; Bulkeley, 2000). There are clear similarities between this framework and the concept of competing discourse coalitions and whilst the advocacy coalition framework has been widely adopted, particularly in political and policy science studies (see Marsh and Smith, 2000), it has been criticised for its neglect of discursive aspects of policy formation, lack of consideration of policy learning and the maintenance of a view that is too structural (Bulkeley, 2000). Both discourse coalition perspectives and the advocacy coalition framework do not assume a shared world view between actors, and are open to the possibility of actors being involved in several coalitions and utilising different policy narratives at different times (Bulkeley, 2000). Discourse coalitions, however, conceptualise interests and beliefs as being shaped by and through the policy process and the actors engaged in this, rather than semi-static (Bulkeley, 2000). These discursive processes are vital to the integration of theories of power into policy studies, particularly with reference to the negotiation and struggle that takes place through power relations in the processes of policy reform. 2.2.2.4 Policy space Alongside the growing appreciation of the interactions between “complex constellations of actors” (Brock et al., 2001: 7) in the process of policy evolution, Grindle & Thomas (1991) introduced the idea of ‘policy space’ to describe the arenas in which actors and discourses interact around policy issues. The interactive model of policy as a process set out by Grindle & Thomas (1991) uses the concept of policy space to describe the ability of a political regime and its leadership to pursue policy reform without causing significant conflict or threatening the power of the regime. The extent of policy space is seen as a function of ‘room to manoeuvre’ which can be created and shaped by actors within the policy elite (or 40 community), but, in line with ideas of policy narratives and discourse coalitions, is also shaped by the narratives and discourses surrounding an issue and used by the policy elite (Grindle and Thomas, 1991). The concept argues against deterministic models of behaviour and allows for the integration of structural and agent-centred factors, recognises the importance of discourse to these, and incorporates aspects of change over time in policy learning alongside strategic action and struggle for power and control over policy space (Grindle and Thomas, 1991). I utilise policy space as a term to describe the room for manoeuvre around a political issue, but recognise that this is a product of power relations, resulting in distinct and shifting policy space surrounding different individuals and groups. There are clear similarities between the policy space concept and Kingdon’s (1984) theory of policy streams and windows, in which a stochastic or chance element of policy formation is proposed. Policy windows appear when “in response to a recognized problem, the policy community develops a proposal that is financially and technically feasible, and politicians find it advantageous to approve it” (Sabatier, 1991: 151). Crucially Kingdon’s approach recognises the role of key actors, or ‘policy entrepreneurs’ to creating these windows, and is insightful in explaining how certain issues and policy solutions take centre stage, revealing some of the politics within the process. However, the conceptual map that Kingdon provides, like Sabatier’s advocacy coalition framework, is lacking a serious consideration of power to these issues (Keeley and Scoones, 2003). Whilst I do not employ Kingdon’s theory in my own research for this reason, I do integrate his idea concerning the role of individuals acting as entrepreneurs as key to policy processes and the shaping of policy pathways. I employ the term policy entrepreneur as this incorporates actors from a wider group, and acknowledges that such key individuals may be part of the policy community, discourse coalition or policy network. 2.2.3 Policy Processes: An Integrated Approach In my discussion of policy processes in chapter 5, I employ a conceptualisation of policy that integrates the discussion in the previous sections on network and process views of policy. I utilise the concept of a policy network, consisting of a policy community and multiple discourse coalitions, made up of groups of actors linked by policy narratives and vying for policy space and discursive institutionalisation. I adopt a policy as discourse view that contends that “policy does not move neatly from stages of agenda-setting and decision- making to implementation: policy is often contested, substantially re-shaped or even initiated from a range of places...it is this complexity and dynamism...that may allow for the assertion 41 of alternative story lines and practices, which, in turn, can gradually result in substantial challenges or shifts in knowledge and practices associated with previously dominant discourses” (Keeley and Scoones, 2003: 22). In chapter 5, I use this framing of policy to investigate the process of policy reform to introduce CBNRM, and the characteristics of the policy network, including the policy community, discourse coalitions and policy entrepreneurs in the shaping of policy. In chapter 5 I also examine the role of the local level, and specifically pilot projects, in policy processes through the construction of policy space and position within the policy network. Incorporating theories of policy as discourse necessitates the “analysis of webs of power underlying the actions of different actors in policy processes, as well as practices invested in policy negotiation or contest” (Keeley and Scoones, 2003: 24). I draw on this positioning of power at the centre of policy processes, conceptualising them as a “reflection of structured political interests, the product of agency of actors engaged in a policy area, and also part of over-arching power-knowledge relations that discursively frame practice in different ways” (Keeley and Scoones, 2003: 38). It is to the theorisation of power, and its integration in to both the study of policy processes and the power struggles taking place within CBNRM that this chapter turns next. 2.3 Power Issues of power have become central to the study of policy and, whilst policies are obviously political phenomena, the politics of policy processes, and the power relations involved, are often “disguised by the objective, neutral, legal-rational idioms in which they are portrayed” (Shore and Wright, 1997: 8). This is what Ferguson (1990) referred to as the ‘anti-politics machine’. Hidden beneath apolitical discourses of participation, empowerment and neoliberal win-win scenarios, the crisis surrounding CBNRM is one of socio-politics and governance issues which drive inequalities, re-shape CBNRM according to actors’ own interests and produce unexpected results. Placing power at the centre of these politics and processes requires careful thought however. Power is a term that has occupied the minds of social scientists for many years and in many ways. The debates surrounding power operate across the realms of political science (Haugaard, 2002), policy studies (Keeley and Scoones, 2003), anthropology (Shore and Wright, 1997) and sociology (Foucault, 1975; Foucault, 1982; Clegg, 1989). Whilst concepts 42 and theories of social power abound, highlighting what Clegg & Haugaard (2009) refer to as the many ‘faces’ of power, it is important to acknowledge that there is not one single type of power, and no single explanatory mechanism for it. It is social power, “based upon knowledge and membership of social systems” (Barnes, 1988; Haugaard, 2002: 113), rather than natural power (such as the power to move and object) that has interested social scientists (see Lukes, 2005). Scott (2001) argued that studies of social power form two distinct streams: a mainstream view concerned with the exercise of power and an alternative stream which conceptualises power as a dispositional capacity, without necessarily exercising it. Traditional concepts of power focused upon the first of these streams, describing it as ‘power over’ (Göhler, 2009). For example, this was famously discussed as sovereign power in Hobbes’ Leviathan and Machiavelli’s The Prince (Clegg and Haugaard, 2009). This classical view conceptualises power as ‘zero-sum’, asymmetrical and hierarchical, leading to the concentration of power in the hands of an (individual or group) elite (Scott, 2001). Whilst the classical views of power have been reinterpreted over time, generating different schools of thought and conceptualisations of power, the power debate that gripped the social sciences in the latter half of the 20th century took its starting point from the ideas of elite control of power within this first stream. The debate is often traced back to the work of Mills who, in the 1950s, published two seminal works concerning the power elite and community power structures (Haugaard, 2002). These were famously criticised in behaviouralist or pluralist views of power which argued that different actors and interests dominate over different issues (Dahl, 1961). In the 1970s, these ideas were further expanded, particularly by Bachrach & Baratz’s (1962) study of a second face of power. This key work introduced the idea that power was exercised not only in the outcomes of decision-making, but in the agenda setting created by the ‘organisation of bias’ that led to dominant groups “limit[ing] decision-making to non- controversial matters, by influencing community values and political procedures and rituals, notwithstanding that there are in the community serious but latent power issues” (Lukes, 2005: 5). Schrattschneider referred to this process as the way “some issues are organised into politics, others are organised out” (in Lukes, 2005: 7). These works opened up the study of power beyond sovereign power and to more subtle forms such as authority, coercion, manipulation and influence (see Lukes, 2005), and raised questions about whether conflict was a necessary condition for the exercise of power. 43 The power debate has developed into a second stream of theory identified by Scott (2001): an exploration of power not just as ‘power over’ but as ‘power to’. Parsons (1963) contributed greatly to these theoretical developments with his argument that power is not present only in its exercise, but also as a capacity that is socially produced (Haugaard, 1997). In this thesis, I employ a view of power that draws on this second stream, defining it as “the ability to achieve a desired outcome in competition with other actors who lay claim to the same resources needed to produce that outcome” (Hyden, 2008: 11). I argue that power is both exercised and is a dispositional capacity, therefore. As sociologists and political scientists turned their attention to more subtle forms of power, Lukes (1974; 2005) published a ‘radical view’ that extended these debates into a third dimension; ‘false consciousness’ as a way in which power is working in unconscious ways, affecting people’s values through processes of internalisation (Scott, 2001; Lukes, 2005). This third dimension of power was used by Gaventa (1982) in his study of mining communities in an Appalachian valley, where he argued that quiescence of the dominated community could be attributed to power relations and specifically the influence of consciousness created by information control and socialisation and indirect means, such as anticipated defeat. These ideas linked debates back to Gramsci’s ideas of hegemony, and to the later work of Pierre Bourdieu, who described how “social agents are endowed with habitus, inscribed in their bodies by past experiences: social norms and conventions of the various fields20 are incorporated or inscribed into their bodies, thereby generating a permanent disposition, a durable way of standing, speaking, walking, and thereby of feeling and thinking” (Bourdieu, 1990 in Lukes, 2005: 141). Bourdieu’s idea of habitus is conceptualised as “below the level of consciousness [and, unlike hegemony]...resistant to articulation, critical reflection and conscious manipulation”, which he described as ‘symbolic violence’ (Bourdieu, 2001 in Lukes, 2005: 140). What both these major players in social theory point to is the movement of the consideration of power beyond that of the exercise of ‘power over’ to more complex forms of the internalisation of values that represent the dominant. This raises questions (and has been the basis for myriad theories) as to how this is achieved, what this tells us about power and whether it is necessarily undesirable. The internalisation of socially-constructed power forms a key element of the research project, which attempts to understand the power systems within which CBNRM is operating and 20 “stratified social spaces within which individuals struggle for unequally distributed resources or capital” (Bourdieu, 1990 in Lukes, 2005: 141). 44 specifically, how powerful actors within this are able to negotiate the system, manipulating it for their own ends. In this thesis, I explore the axes of power present within Tanzanian CBNRM therefore, and investigate the internalisation of values that legitimise and support the power systems. I argue that a conceptualisation of power must address debates that have been taking place within power literatures, but also more broadly across the social scientists and address the issue of structure and agency. In addressing the questions of both the internalisation of values and how actors seek to exert power and control over each other and the environment, it is crucial to address the issue of how this takes place, where and how this power is created, and whether it is structurally or socially derived. 2.3.1 Structure and Agency “Theorizing the relationship between power, social knowledge and structure inevitably takes us into the debate on the autonomy of the individual in relation to social order” (Haugaard, 1997:1). Power debates take place amongst wider debates concerned with the respective roles of structure and agency in society, politics and power (Giddens, 1984; Hayward and Lukes, 2008). There is a long-standing tension between structural-functional and Marxist accounts of power and actor-orientated perspectives that emphasise inequalities in the distribution of resources and power, but also acknowledge opportunities for people to shape these relationships through agency (De Haan and Zoomers, 2005). Whilst Gramsci sat firmly at the structural end of the spectrum, Bourdieu attempted to find a middle ground between structure and agency to explain how behaviour is a product of both. It is the relationship between structure and agency that interests me, in terms of how power systems function, how policy is created and how its implementation and the complexities of outcomes witnessed in CBNRM are a process of continual tension and renegotiation between structures of governance and the agency of individuals. I argue that all of these are socio-political processes ongoing within CBNRM that have power at their core and that are all discursive processes (Feindt and Oels, 2005). Several attempts to combine the prospective roles of structure and agency and structural and post-structural accounts of power have become prominent, most notably in the work of Giddens’s ‘structuration theory’. Essentially, Giddens argued against a dualism of subject and object-centred social theories, proposing structuration as an alternative whereby “social 45 structures...exist in the moment that they are reproduced by agents while, simultaneously, social agents constitute themselves as such through structured action. This moment of reproduction of agency and structure is structuration” (Giddens, 1984 in Haugaard, 2002: 146). Such attempts aim to make it possible to investigate both how actors are constrained and shaped by the socio-political structures that they live amongst, whilst acknowledging that these structures are socially constructed and are continually being challenged and re-shaped by actors (2004). I argue that this conceptualisation of power through a combination of structure and agency is critical to understanding the implementation of an externally-designed governance system that is inserted within the socio-political context of a community under CBNRM. This perspective is important to the consideration of institutions within my research, which brings together structure and agency to see both formal and informal institutions as “constituted through social processes, which they in turn enable and constrain. They embody social relations and rules, such as norms, values and behaviour, to which they lend stability and coherence over space and time. Institutions, therefore, encompass concrete and cognitive structures that are habitualised over time and attain a degree of permanence within society” (Bulkeley, 2000: 731). Such arguments form the basis of a research agenda that emphasises the socio-political nature of CBNRM, in terms of both policy and performance of that policy and highlights the need to investigate this performance as continual socio-political processes of re-construction and re-shaping of the governance system. In line with the growth of poststructural theory and the debates surrounding structure and agency, more recent theories of power have focused upon concepts of power that are not zero-sum but see power as created (often attributed to the work of Parsons, see 1963), the relationship between power and knowledge (see Barnes, 1988), and questions of whether and how compliance to a socio-political power regime is accomplished (Foucault, 1975; Dreyfus and Rabinow, 1982; Bourdieu, 1989). Haugaard (1997) argues that in modern thinking on power, four paradigms all concerned with these issues of structure and agency and the nature of power as domination and ‘political technology’ through the social construction of subjects are identifiable in the respective works of Steven Lukes, Michel Foucault, Barry Barnes and Anthony Giddens (as previously discussed). Within these four key thinkers’ work on power, Foucauldian theory has been widely adopted, both within the social sciences more broadly, but especially within poststructural political ecology studies (see 2.5.1). Foucault did not describe his work as being specifically an exploration of power, but did acknowledge that it 46 played a major role in his work. The critical departure for Foucault from Marxist theory and the theories of power that prevailed was that he could not accept that knowledge and power, whilst inter-related, could be separated in the form of an ideologically-free, domination-free form of knowledge; he saw all knowledge, all meanings and all social life as being subject to “petty and ignoble power relations” (Haugaard, 1997: 43). He argued that “it is not possible for power to be exercised without knowledge, it is impossible for knowledge not to engender power” and “the formation of ‘truth’ does not simply happen. Rather, the truth of any discourse formation is the consequence of the struggles and tactics of power” (Foucault 1980 in Haugaard, 1997: 66 and 68). He moved away from concepts of negative-only (sovereign) power, and embraced theories of the construction of power, labelling it essentially positive. He was interested in “the point where power reaches into the very grains of individuals, touches their bodies and inserts itself into their very actions and attitudes, their discourses, learning processes and everyday lives (Foucault, 1980 in Lukes, 2005: 89). Here there are clear similarities to Bourdieu’s (1989; 1992) concept of habitus, but whilst Bourdieu concentrated on where practice comes from, Foucault focused more on the interaction between ‘techniques of domination’ and ‘techniques of the self’ to understand what practice does (Threadgold, 2006). Bourdieu also argued that knowledge is not discursive, but forms part of habitus, whereas Foucault focused on the discursive nature of power (Haugaard, 1997). Foucault argued that the discursive nature and ‘regime’ of power is not necessarily oppressive, but is positive and productive because the subject (himself created by it) must submit to it, in order to be seen to liberate himself from it (Haugaard, 1997). For example, a patient in a mental health hospital is deemed to be cured, and can be freed from the restrictions placed upon him, only once he has accepted his diagnosis and submitted himself to the treatment proposed by the institution, which inherently involves the reification of the power structure in place (Haugaard, 1997). Taking this view of power as positive, Foucault discussed the concept of ‘subjectification’ extensively, and created what is perhaps the most widely-applied aspect of Foucauldian theory; governmentality. Foucault considered the most pernicious form of power to be its normalised forms. In his studies of the deep, intimate connections between power and knowledge, he introduced the concept governmentality to describe “the ways in which subjectivity is constituted within a constellation of powers and in which people continuously and permanently survey and govern themselves as an effect of those powers” (Kesby, 2005: 2042). Governmentality describes the generation of “compliant 47 subjects who actively reproduce assemblages of power without being forced to do so” (Kesby, 2005: 2040). Foucault described this process using the example of Bentham’s ‘panopticon’, a prison in which all inmates are positioned to be continually visible to a central watchtower, where authority (itself unseen) presides (Foucault, 1975). Each inmate has no way of knowing when he is being observed, but begins to modify his own behaviour and observe himself, a process upon the soul which Foucault described as the ‘conduct of conduct’ (Dreyfus and Rabinow, 1982). Whereas Gramsci argued that even under hegemony and the ideological subordination of the working class, resistance and political struggle would always be present, Foucault argued that governmentalised subjects cannot see the power relations at work, which have been normalised and accepted as legitimate, producing passive subjects even in contexts of extreme domination (Gaventa, 1982; Haugaard, 2002). Foucauldian theory has been employed by Arun Agrawal in his study Environmentality (2005), in which he explores the ways in which policy is a political technology. He uses the evolution of community-based forest management in India as an example, focusing on the shifts in relations between state and localities that this involved, the creation of environmental knowledges and the formation of self-identity within devolved environmental management. He argues that in his case study changing relations between the state and society produced new ‘governmentalized localities’ through the construction of new centres of environmental decision-making or ‘regulatory spaces’, and new forms of relationships termed ‘regulatory communities’ (Agrawal, 2005). Within these regulatory spaces and communities he describes the process of governmentality in the production of environmental subjects (Agrawal, 2005: 14). He also focuses on the discursive construction of these environmentalities and governmentalised localities: “regulations and villagers’ practices and words seem to be part of a process that has reshaped people’s understandings of the forest and the basis of forest control itself” (Agrawal, 2005: 12). Foucauldian concepts of discourse focus less upon language and more upon knowledge, and are characterised by four features (Feindt and Oels, 2005): firstly a focus on the productive function of discourse and the ways in which discourse establishes truths about the world, the environment or policy options; secondly that power is present in all forms of social interaction and is both productive as well as repressive. This supports a theorisation of power that is “understood as a web of force relations made up of local centres of power around which specific discourses, strategies of power and techniques for the appropriation of knowledge cluster (Feindt and Oels, 2005: 164); thirdly that discourse is a strategic situation 48 that both forms and is formed by actors, and are locations of conflict. This is an important aspect in the integration of structure and agency; and finally that discourse is important to the construction of subjectivity and governmentality. I employ such a perspective firstly through consideration of the discursive production of CBNRM through policy processes. Secondly my research considers the discursive practices of negotiation and political struggle ongoing within Tanzanian CBNRM, and in 6.3.4 I focus specifically on conflicts taking place and the power struggles these represent. Finally, I address the issue of power and resistance by considering evidence of governmentality within Tanzanian CBNRM, particularly with reference to the informal and hidden systems of power and issues such as corruption and neopatrimonialism. Foucauldian theory has raised many questions and critiques, principally the issue of resistance. Is a governmentalised subject aware of his subjectification? Can this ever be broken? James Scott (1985; 1990) has become one of the most well-known writers on the subject, arguing that resistance is always present and often subtle, quiescence is tactical and hegemony can only ever be paper-thin (Lukes, 2005). Haugaard (2003) recently attempted to reconcile some of these many debates, making room for a nuanced understanding of power that accommodated both negative and positive power, twin roles of structure and agency and room for compliance and resistance. He describes this as the seven ways in which power can be produced (Table 2.1). He attempts to weave a way for power to be considered in its coercive form and as a product of the creation of social order and to bring together the works of many of the influential writers about power, including the work of Barnes, Clegg, Giddens, Lukes, Foucault, Arendt, Bachrach & Baratz and Weber and Dahl, alongside his own earlier work to achieve this. 49 Form of Power Example Description (1) Power created by social order Causal predictability created through the reproduction of meaning: structuration and confirming structuration Structuration is the reproduction of social structure, but this requires recognition to be socially meaningful or powerful. This act of recognition and imbibing of meaning is confirming structuration21 (2) Power created by system bias Order precludes certain actions: de-structuration Similar to the organisation of bias (Bachrach and Baratz, 1962) in that he describes the capacity of such agents to ‘de-structure’ or ‘non confirm-structure’ the practices of others, which has the effect of maintaining existing power relations (3) Power created by systems of thought Certain acts of structuration are incommensurable with particular interpretive horizons Structural practices are sustained by social consciousness. Systemic biases are based upon social meanings, which are created by social meanings and systems of thought make some acts of confirming structuration likely, and others not (4) Power created by tacit knowledge ‘Power over’ based upon social knowledge that is not discursive. Empowerment through the transfer of knowledge from practical to discursive consciousness Rejects the idea of false consciousness and argues that consciousness raising and empowerment (through de- stabilisation of social order) are possible through the creation of discursive practical consciousness that provides us with the ability to recognize patterns of thought and behaviour that we have internalised (5) Power created by reification Social power has to appear as non-arbitrary Drawing upon the work of Foucault and Clegg with respect to internalisation and subjectification, Haugaard argues that the destabilisation of social order described in (4) may not always occur. The reification of structures can lead to confim-structuration because actors cannot perceive the disadvantage to them that this might entail. (6) Power created by discipline Routine is used to make actors predictable through the inculcation of practical consciousness knowledge Practical consciousness knowledge cannot be converted because it is blocked through socialisation involving the routinisation of structuration eliciting automatic confirm- structuration, or what Foucault referred to as disciplining the ‘soul’22 (7) Coercion Natural power as a base: violence and coercion as a substitute for the creation of social power The final form of power is violence and coercion, which, as Arendt argued, represents not the ultimate form of power, but the failure of social power23 21 To illustrate this argument, Haugaard uses the example of a ship that is to be named and launched. Instead of the designated person, a stranger walks up, smashes the bottle on the side of the ship, proclaims a name and launches the ship. We would not consider the ship to have been named or launched, because the act would not be recognised, even though the procedure might have been correct; in Haugaard’s argument an act of structuration has occurred, but it has not been confirmed, and is therefore meaningless (Haugaard, 2003). 22 So that upon leaving the panopticon, the criminal retains the self-governing practices he learnt inside (Haugaard, 2003). 23 “Once the sovereign has to draw their sword it is because the Leviathan has failed to create social power”(Haugaard, 2003: 108). Table 2.1: Seven Forms of Social Power (adapted from Haugaard, 2003: 109) 50 I utilise several of Haugaard’s classification of forms of power to investigate how the power systems in place within CBNRM are operating. I employ the concept of power created by social order (1) and by discipline (6) to investigate the internalisation and confirm- structuration of informal aspects of power, particularly with reference to corruption and neopatrimonialism. In chapter 6 I utilise the ideas of power created by systems of thought (3), tacit knowledge (4) and reification (5) in an examination of the relationship between power and knowledge and the control of information within power struggles. Finally I address the question of resistance and the roots of empowerment through evidence of power created by tacit knowledge (4). My analysis of power is one that weaves throughout the empirical chapters, and I argue that power is simultaneously a missing element in the study of policy processes, unacknowledged in its more subtle and informal forms within studies of the local politics of natural resource management, and the basis for contestation within such management. I adopt a Foucauldian perspective of power, arguing that discursive processes of persuasion, negotiation and contestation are central to the prescribed governance system set out in policy and to the reality (and differences therein) of its performance. 2.4 Politics of Scales I have framed my approach in this thesis as political ecology (see 2.5). Bryant & Bailey (1997) argue that political ecology is characterised by two central themes: the first is power, as discussed in the previous section; the second is scale, which is the subject of this section. Power and scale are very closely linked as the latter are the product of the compartmentalisation of socially-produced space according to power systems (Zulu, 2009). If we consider the complexities of community conservation to be socio-political, which need to be investigated through an examination of power, the socio-spatial aspects of power are critical to this. Here I discuss some of the fundamental aspects of spatial concerns within human geography, and how these are of relevance to a political ecology study of CBNRM. My interest in the area is in the application of a ‘politics of scales’ approach to investigating the governance of CBNRM projects, and how this is constructed and manipulated by actors in conflicts over access and power. Adopting a politics of scales approach facilitates further investigation of the processes of negotiations and struggle described in 2.3, therefore. Below I trace the development of the term, its application and debates surrounding its usage. 51 2.4.1 Geography and Scale “What gives geographers their disciplinary identity is their explicit consideration for spatial relationships” (Gibson et al., 2000: 226). Geography, in its many guises, is fundamentally concerned with place and space. Geography’s history in exploration and tradition of regional study clearly show this, and contemporary geography retains these foci, particularly in the expanding fields of Geographic Information Systems and remote sensing. Since the early 1990s, human geographers interested in social theory, urban contexts and economic geography have paid increasing attention to its theorisation and study (Brenner, 2001). Scalar theories, whilst multiple, all come back to the essential argument that space is socially constructed. Attempting to distance themselves from the 1960s geography as a spatial science, authors such as Massey (1978) began to discuss space as a social phenomenon, a theme that has been widely adopted in the social sciences, both within and beyond human geography. Studies in the 1970s and 80s focused heavily upon the production of space, territory, locality and place (see Harvey, 1989). The social construction of space and place was an idea famously discussed by Henri Lefebvre (1974), who argued that spatial configurations within society or ‘social superstructures’ are both the driver and product of the development of capitalism. His work raised ‘the scale question’ (Brenner, 2001), by positing that the social compartmentalisation of space produces ‘scales’, which has been the inspiration for substantial literatures in human geography. The centre of these areas of study has been studies of capitalism and particularly “the uneven development of capital and the geography of industrial location; ...the changing geographies of state power, political regulation and socio-political identity; and...the organizational structures and strategies of labour unions, political parties and social movements” (Brenner, 2001: 592). The study of scale, its rise to prominence in geographical enquiry in the 1980s, and the debates this entailed are a product of context, and particularly the issue of globalisation at the time (Sayre, 2005). This period in history is conceptualised as involving especially large scalar transformations of organisation, associated with the growth of neoliberal models (Brenner, 2001). Brenner (2001: 594) described this shift as an “aggressive attempt to forge new global, national, regional and local scalar hierarchies in which unrestricted capital mobility, unfettered market relations, intensified commodification and a logic of ‘beggar-thy- neighbour’ competition are to be permanently institutionalized (sic)”. 52 According to Marston (2000: 221), this period of studies exploring social constructionist ideas of scale have provided three widely-agreed essential features; firstly that scale is not an external fact or ontological given, but “a way of framing conceptions of reality”; second that the construction of these scalar frames is not a rhetorical act, but is “tangible and have material consequence” in everyday life and social structures; finally it is widely agreed that these framings of reality are not accepted and stable, but actively contested, often contradictory and under constant re-organisation. Employing the idea of scales as frames of reality, it is vital not to assume hierarchy or a top-down, nested view of scales (Bulkeley, 2005; Marston et al., 2005; Purcell and Brown, 2005). In such views, sometimes referred to as the Russian doll or ladder analogies, scales are seen to start with the local and move up through regional, national and international levels, in which the latter rules and the former accommodates (Leitner and Miller, 2007). Scales are better conceptualised as networks or ‘tangled hierarchies’, emphasising the “the shifting organizational (sic), strategic, discursive and symbolic relationships between a range of intertwined geographical scales and on the ramifications of such interscalar transformations for the representations, meanings, functions and organizational structures of those scales” (Bulkeley, 2005: 884). Instead of hierarchy, scalar theorists must focus upon verticality, the inter-relatedness between scales and the politics both within and between them (Rangan and Kull, 2009). The last of Marston’s (2000) three features is the basis for the second24 fundamental theoretical underpinning in the development of the concept of politics of scales: the differentiation of space is infused with power relations and a process of political struggle (Zulu, 2009). Scales are not fixed, therefore, but are spaces of constant conflict and re- shaping and power relations are central to the scalar topography, but also the fulcrum around which struggles emerge and take place. This was a central theme in Lefebvre’s work (1974), in which he saw power relations as ‘sociospatial’ (Leitner and Miller, 2007: 199). As scale debates began to focus on the processes of scalar transformation and struggle, Smith (1992: 74) made a crucial contribution to the theorisation of scale when he argued that “the scale of struggle and the struggle over scale are two sides of the same coin”, calling for a structuration element to scalar theory in which structure and agency are mutually constitutive “with agents enacting and transforming structures through their actions and structures enabling and constraining human actions” (Leitner and Miller, 2007: 118). The fluid nature of scales and their constant construction and reorganisation has become a major theme and point of critique 24 The first being that space is socially constructed and compartmentalised into scales through power. 53 of early studies (Marston et al., 2005). These scalar struggles are conceptualised as “an integral part of social strategies to combat and defend control over limited resources and/or struggle for empowerment” (Swyngedouw and Heynen, 2003: 913). Scales should be seen not as things in themselves with inherent qualities, but rather as strategies that are pursued by (and benefit) social groups with particular spatial and environmental agendas” (Purcell and Brown, 2005: 279). Space and spatiality, therefore are intrinsically linked to understandings of poststructuralist power, as seen in the work of Foucault (Allen, 1999). Elaborating on his ideas about the transformation of scales, Smith (1993) discussed the idea of scale ‘jumping’. The term refers to the transformation of scalar formations through actors being able to use their social power to jump into another scale, thus resisting hegemonic structures (Smith, 1993). In a later publication (2004: 193), Smith extended his arguments further to differentiate between scale jumping and scale ‘bending’, where the latter refers to the upsetting of “entrenched assumptions about what kinds of social activities fit properly at which scales”. Using the example of the Mayor of New York City25 Rudy Giuliani ,addressing the United Nations on the topic of the city’s role in the world in the wake of the September 11th terrorist attacks. Smith argued that his speech to the United Nations was an example of scale bending in which an actor contravenes the cognitively-comfortable organisation of scales. Smith (2004: 205) differentiated between scale jumping and bending as the latter involving a “fragmentation of pre-existing scales”, whereas scale jumping is less transformative. 2.4.2 Politics of Scales: Transformations In the second edition of Uneven Development, Smith (1990) pulled together his work on the socially-constructed nature of scales and their essentially contested and fluid nature to coin the term ‘politics of scales26’. Politics of scales can be defined as “situations whereby actors, directly or indirectly, attempt to shift the levels of study, assessment, deliberation and decision-making authority to the level and scale which most suits them, that is, where they can exercise power more effectively” (Lebel et al., 2008: 129). The focus is, therefore, on processes among and between scales (Brenner, 2001). 25 Not a nation state, which is the scale usually associated with the construction of foreign policy. 26This is distinguished from a politics of scale approach which denotes “the production, reconfiguration or contestation of some aspect of sociospatial organization within a relatively bounded geographical area – usually labelled the local, the urban, the regional, the national and so forth” (Brenner, 2001: 599, emphasis in original). 54 It incorporates the ideas, not only of transformation, but of a strategic re-organisation of scales, and as resistance to the distribution of power. These two elements are discussed by Edwards et al., (2001: 291-2) in their study of partnerships as a new scale of rural governance in the UK, which they argue may be approached in two ways; “in terms of the extent to which power and responsibility have been shifted between the existing scales of governance; or ... in terms of creating entirely new scales of governance not concurrent with the territories of any existing institutions”. In this study of the emergence of new forms of governance and its scalar implications they argue that new forms do not develop on a scalar blank canvas, “rather they are constructed across a landscape already imprinted with a scale and territorial division of the state” (Edwards et al., 2001: 291). They make a forceful case for the adoption of what has become known as a process-based approach to scale (Swyngedouw, 1997), in which both construction and transformation of scales are key. This approach to scale “focuses attention on the mechanisms of scale transformation through social conflict and political struggle” (Swyngedouw and Heynen, 2003: 913). I employ this approach in chapter 6 in an investigation of the scalar landscape surrounding CBNRM in Tanzania, the scales of power prescribed in policy and their integration with existing and newly constructed scales in the performance of policy. In 6.3.4 I focus specifically on how these scalar processes are taking place through conflict within the case studies. In my investigation of scalar transformations I focus specifically on two aspects; the discursive construction of scale and scalecraft. 2.4.2.1 The Discursive Construction of Scale & Scalecraft A politics of scales approach, which plays close attention to the role of power in scalar struggles, allows for the integration of another stream within scalar studies, that of the discursive construction of scale (see Agnew, 1997). As discussed above, Marston (2000) claimed that one of the greatest contributions of the attention paid to scale in the 1990s has been the conceptualisation of scales as frames of reality. In combination with a process- based approach, scale is seen as discursively produced through the employment of these framings, but also continually contested by alternative framings and discursive strategies in the battle for power over the scalar landscape. Discursive strategies for scalar organisation within natural resource management privilege a particular level as appropriate (Lebel et al., 2008). For example, Lebel et al. (2005), in an earlier study of river basin management, argued that discursive strategies re-scaled management to the national level through arguing that this was both legitimate and in the 55 national interest, and through arguing that alternative levels such as the local or watershed were inappropriate due to reduced accountability. Lebel et al., (2005) also argued that these strategic relations of actors in space in the construction and utilisation of discourses for scalar purposes is present not only in the narratives and discourses employed in such processes, but also in the control and manipulation of information, which empowers some and disempowers others. On this point, there is clearly a strong link to the discussion of power in 2.3. The control of information and knowledge is a form of political technology by which power relations are exercised, and actors engage in such processes within the power struggles that represent politics of scales. Fraser (2010: 335) considers the strategies employed by actors in terms of ‘scalecraft’, which he defines as “the skills in negotiating spaces of engagement ... Scalecraft draws our attention to actors’ skills and agency amidst the structures of opportunity and constraint that constitute the politics of scale[s]”. This concept provides an important link between the perspectives of structure and agency described in section 2.3.1, which is critical in the integration of the role of power in scalar processes. This concept makes room for discursive strategies and scalar processes within and around the existing scalar structures and opens up debate about whether these structures form the backdrop around which agency is organised. In turn, this raises further questions about power, and whether the politics of scales processes, whilst attempting to transform the scalar landscape, simultaneously strengthen and reify systems of power through resistance to a recognised power system, as described by Haugaard in the first form of power (1997; see also Lukes, 2005 and 2.3.1). Power, policy processes and politics of scales are inter-related in the socio-politics of CBNRM. The process of policy formulation is a discursive process imbued with power relations which results in the construction of a scalar landscape, as represented in the prescribed governance structure. This scalar topography of power is not static however, and actors are continually engaged in negotiations and reconfigurations to position themselves to benefit from the power system in place, and to alter it for their own benefit. This they do through discursive strategies. In line with the discussion of structure and agency in 2.3.1, the scalar topography and politics of scales taking place around CBNRM represent structural systems of governance as prescribed in policy and their adaptation through the agency of individuals. Similarly, the performed governance structure is comprised of alternative power structures originating from the communities in which CBNRM is being implemented, and these are amalgamated with the prescribed governance structure and their adaptations in a 56 complex web of shifting power relations and political struggle. I bring these literatures and theoretical insights from policy processes, power and politics of scales together in a study that I define as adopting a political ecology approach. As I will discuss in the next section, power and scale have been central themes within such an approach, which has a long history of overlap with politics of scales literature in its consideration of questions concerning “the various ways and forms in which one actor seeks to exert control over the environment of other actors” (Bryant and Bailey, 1997: 39). 2.5 A Political Ecology of Nature Conservation In this thesis, I employ a political ecology approach to investigate the socio-politics of CBNRM projects in the Tanzanian wildlife and forestry sectors. In this section, I explore the field of political ecology, its roots and development and ask what it can give to such a study, why it is appropriate and how it has framed my research questions and fieldwork. Problematically, it is not quite clear what political ecology is: some describe it as a research agenda (Bryant, 1998), some as an approach (Zimmerer and Bassett, 2003), others as a perspective (see (Neumann, 2005) and others as a discipline (Blaikie, 2011). In addition, authors may define their particular approach within political ecology as radical, critical, feminist, urban, poststructural, Marxist, third world and first world (Neumann, 2005), whilst others prefer the term ‘liberation ecology’ (Peet and Watts), and some have rejected it entirely in favour of ‘environmental politics’ (Vayda and Walters, 1999). This diversity within political ecology stems partly from its interdisciplinary nature, and partly from its history of grounded studies rather than theoretically-focused research (Bryant and Bailey, 1997). The approach taken within this study is one that focuses on the politics of environmental issues and I follow the definition provided by Jones (2006: 483) that political ecology is “centrally concerned with the politics of struggles over the control of, and access to natural resources”. I use the phrase political ecology approach, as I feel this best describes its use as both a perspective and framing of an issue, and an agenda for how this can be studied. To explore this exciting but contested approach, I discuss key aspects of its history and a political ecology of biodiversity conservation specifically. I examine some of the key debates in political ecology and the multiple divisions within it, and outline the approach adopted in this study. A more detailed history and discussion of the complex roots of political ecology can be found in Neumann (2005), Robbins (2004) and Peet et al. (2011). 57 If we accept natural resources, policy and management of such resources as intrinsically political, it is clear that action to conserve natural resources, whether through PAs or CBNRM cannot be other than political (Adams and Hutton, 2007). This is a core concept of a political ecology of nature conservation, which aims to integrate politics into an analysis of conservation, both through recognition that the biophysical condition of nature is dependent upon socio-political processes, but also that the understanding of that biophysical state is a political process in itself (Adams and Hutton, 2007). Much political ecology has considered, therefore, the politics of knowledge and science, how we understand the world we live in, our impacts upon it, and how certain ways of understanding these things, particularly Western scientific knowledge, become privileged over others (Blaikie and Brookfield, 1987; Roe, 1991; Adams and Hutton, 2007). Understanding these complex outcomes and the socio-politics driving them has become a focus of political ecology, drawing on an access model inspired by the work of Amartya Sen and the concepts of entitlements (see 2.1.2) and capabilities (Forsyth, 2008). The ways in which political ecology has considered conservation and natural resources has shifted over time, however, starting with a perspective drawn from Marxist political economy and later being dominated by post-structural accounts. In the following section I discuss the core characteristics of both of these phases within political ecology’s history, before outlining, in 2.5.2 political ecology’s contribution to nature conservation debates and, in 2.5.3, the post- structuralist perspective adopted in my own research that brings together power, scale and policy processes under the umbrella of a political ecology of community-based forest and wildlife management in Tanzania. 2.5.1 Political Ecology Perspectives Natural resources and human-environment relations have long formed the core of geographical studies (Bryant, 1998). Cultural ecology grew out of this long tradition and was a key precursor to political ecology, which attempted to integrate the former’s focus on the ways in which humans shape their environment, alongside the role of external systems, particularly the global capitalist economy, to the phenomena it studied (Neumann, 2005). Political ecology emerged as the study of the influence of external forces and systems upon human-environment relationships in the local context (Forsyth, 2003). This is exemplified in Michael Watts (1983) famous study of drought and famine in a Hausa peasant community in Nigeria. Watts (1983) argued that these phenomena must be recognised as rooted in a 58 process of the erosion of rural poor communities’ adaptive capabilities through the colonial implementation of capitalist modes of production, criticising the Cartesian dualism of humans and nature as separate and insisting on replacing it with a new dialectic of society and nature (Neumann, 2005). Watts (1983) conceptualised this dialectic as giving “greater attention to the social and historical contingencies of knowledge and political economic structure”, setting out his case for the inclusion of social theory and engagement with Marxist political economy (Neumann, 2005: 22). The relationship between political ecology and political economy is a crucial one, and one of the approach’s most famous definitions argues that political ecology combines “the concerns of ecology and broadly defined political economy” (Blaikie and Brookfield, 1987: 17). This political economy focus dominated in the 1970s and 80s, producing some of the most famous and foundational texts in political ecology. In (1985), Piers Blaikie published The Political Economy of Soil Erosion in Developing Countries, in which he addressed the failure of policies and projects to address the issue. He argued that environmental degradation is inherently and concurrently an ecological, economic, social and political problem whose investigation requires three essential features; place-based analysis to examine where degradation is taking place; non-place-based analysis to incorporate socio-economic factors such as state-society relations and mechanisms of production; and finally he opened up the issue to incorporate “an assessment of the perceptions and ‘rationality’ of not just the local land users, but also of the government officials, conservationists, and scientists” (Neumann, 2005: 31). He therefore introduced some of the key characteristics of political ecology in multiple-scale enquiries and the integration of social and biophysical analyses (Rocheleau, 2008) and later questions of power and knowledge (Bryant, 1998). This first phase of political ecology is closely tied to both political economy and the examination of biophysical processes alongside social and economic factors (Bryant and Bailey, 1997). Numerous studies emerged under this new framework (see Bassett, 1988; Peluso, 1992; Bryant, 1998), “stressing the social relations of production influencing land users' choices and access to environmental resources” (Moore, 1993: 381). Furthermore, this period saw the consolidation of historical analysis and multiple method investigations as characteristic of the approach (Rocheleau, 2008). Political ecology came under increasing criticism however, for its reification of structural forces, and a lack of serious attention to political factors (Walker, 2005). Michael Watts (1990 in Walker, 2005: 75) described the need for political ecology to engage with the “rough and tumble” of environmental politics, 59 whilst Moore (1993: 381) called for political ecologists to turn their attention to the “micropolitics of peasant struggles over access to resources and...the symbolic contestations that constitute those struggles”. In the 1990s political ecology made an important shift towards post-structural perspectives. Poststructural political ecology turned towards the politics of political ecology, and particularly the discursive dimension of power. This was not simply a reaction to the deficiencies of structuralist explanations and alternative approaches in environmental research, but a theoretical framework that conceptualises the environment as inherently ‘politicised’ (Bryant and Bailey, 1997). The concept of the politicised environment draws upon three key assumptions (Bryant and Bailey, 1997: 28): that environmental change creates unequal patterns of costs and benefits for society; that this inequality reinforces or reduces existing socio-economic inequalities; and finally, that these inequalities have political implications through altered power relations. Peet & Watts’s (2004) book Liberation Ecologies is an example of these developments in the approach, positioning itself as a poststructuralist, discursive and social theory based study that is rooted in the social movements literature to examine nature-society relations and access and control over resources in new forms of global governance. These foci within political ecology are very compatible with a study of the socio-politics of CBNRM. Whilst the acknowledgement of differential outcomes, the imposition of a new form of governance upon existing power structures and the importance of socio-political factors are all important developments within CBNRM literature, political ecology provides an approach for the investigation of these factors and processes. Drawing upon the access model, issues such as inequalities in the impacts of CBNRM, both positive and negative need to be explored through their relationships to local power relations alongside broader socio-economic processes (Birkenholtz, 2011). Fairhead & Leach’s (1996) text Misreading the African Landscape became another key publication that marks the changes in the approach during the 1990s. In line with poststructuralist theory, they highlight the importance of seeing multiple readings and understandings of the landscape and environment in their study of vegetation change in Kissidougou, West Africa. Their study focused on a pluralistic ecology approach that incorporated local knowledge to deconstruct crisis narratives of deforestation seen in forest islands in the area and argue that these narratives can be understood through attention to 60 politics and power as state technologies to justify nationalised control over land (Fairhead and Leach, 1996). They also drew on literature on agrarian change to provide insights into how social arrangements condition resource use and management, the everyday struggles these involve, and their resolution (Fairhead and Leach, 1996). Their study is an example of how political ecology began to challenge concepts of a universal truth in knowledge, and open up questions about different types of knowledge (Zimmerer, 2006). The socially- constructed nature of knowledge became a major theme, alongside the nature and power of narratives and discourse (Forsyth, 2003). This has been described as political ecology’s ‘turn to discourse’ (Peet and Watts, 2004) and produced large volumes of work that consider the ‘politics of knowledge’, including gendered bases of environmental knowledge (Shiva, 1991; 1993), the dominance of Western scientific knowledge over other forms (see Bryant, 1998) and the discursive construction of environmental narratives, including degradation (Neumann, 2005). This phase of political ecology studies is characterised by increased attention to “local level studies of environmental movements, discursive and symbolic politics and the institutional nexus of power, knowledge and practice” (Walker, 2005: 75). 2.5.2 Nature Conservation in Political Ecology Nature conservation has been a large focus in political ecology, especially discussions of the development of PAs and the history of conservation, and political ecology perspectives have been active in many of the debates surrounding protectionist conservation, community conservation and the social justice movement (see 1.1; Adams and Hutton, 2007). PAs have been particularly popular with political ecologists because of their clear spatial boundaries and the definite control of access to natural resources within and around them (Zimmerer and Bassett, 2003). The roots of this model for conservation and the underlying environmentalism it involves has been one of extensive discussion within political ecology (see Anderson and Grove, 1987; Grove, 1987; Hulme and Murphree, 2001b; Adams and Hutton, 2007). Its development is inherently linked to processes of imperial trade and the desire to map, delimit and claim (for the nation-state) valuable natural resources, particularly timber (Grove, 1987). PAs act as enclosures which, Neumann (2004b) argued, are a political project and an important aspect of statecraft and colonial control, incorporating the twin motives of ‘civilising’ the colonies and bringing land and resources under the authority of the state. Similarly Peluso (1993) argued that as states hold a vested interest in centralising control over natural resources (and in avoiding the devolution of power) and they may, therefore, use conservation discourses to retain authority over these resources. She goes on to describe the 61 discursive construction of identities for the state as guardians of these resources through conservation, and, through this, the legitimisation of violence to achieve conservation objectives (Peluso, 1993). Political ecology discussions of the politics of power devolution and state-society relations can usefully be incorporated with the discussion of the politics of power devolution in 2.1.1 and the mixed results from CBNRM in 1.3. In a clear link to the issue of recentralisation, Peluso (1993) also argued that the incentives for the state to retain control over valuable natural resources has been an important driver in the balance between conservation and development within community conservation, leading to a common criticism that it consists of tweaking coercive conservation to include a development component alongside militaristic management of PAs. This kind of study, focused on the social impacts of PAs and the political economy of conservation benefits through, for example ecotourism and CBNRM, has been common amongst political ecologists. These issues continue to attract attention as the conservation fashion has moved on to ‘ecosystem services’ and concepts such as Reduced Emissions from Deforestation and Degradation27. The way that the relationships between conservation, biodiversity, communities and poverty have been conceptualised by political ecologists have been highly influenced by the tying together of conservation and development objectives and the recognition that outcomes from such attempts are often a set of trade-offs between two set of objectives, rather than the predicted win-win scenarios, and that the social costs of conservation are often not recognised and rarely addressed (Adams and Hutton, 2007). As discussed in 2.5.1, poststructural political ecology is characterised by increasing interest in multiple actors, and identities, situated knowledges and cultural politics with “complexity and contingency in social and ecological relations of power” (Rocheleau, 2008: 722). This represents the addition of network theories to the conceptualisation of inter-related nature of structure and agency in poststructuralist theory (Peet et al., 2011; see also 2.3.1). There are, therefore, important links between these areas of study in political ecology and views of the policy process with respect to policy network views and the argumentative turn (see 2.2.2), and to the politics of scales literature discussed in 2.4 with respect to the discursive construction of scale and the politics of scalar struggles. In the following section I address the 27 See Blom et al., (2010) and Harvey et al., (2009). 62 integration of policy, power and scale under the title of a political ecology approach within my study. 2.5.3 Political Ecology, CBNRM, Power and Politics of Scales “The complexity of human-environment interactions demands an approach that encompasses the contribution of different geographical scales and hierarchies of socioeconomic organisations” (Neumann, 2009: 398) Human geography is centrally concerned with space, place and scale, and despite complex debates surrounding the role of scalar analysis in human geography, many of the themes that weave together a political ecology approach are essentially scalar in their nature (Neumann, 2009). This is especially clear in the inter-relatedness of the social and the natural that political ecology argues for (see Zimmerer, 2000), and also in the central role that political ecology places on the role of power relations in shaping access to and control over environmental resources and space (Neumann, 2009). Political ecology has been criticised however for an under theorisation of scale and reification of the local level (Brown and Purcell, 2005; Purcell and Brown, 2005; Neumann, 2009). Earlier structure-focused political ecology studies (see 2.5.1), whilst crucial to attempts to integrate politics into human ecology, have been particularly criticised for lacking appreciation of the sociospatial, presenting scales as hierarchical and ontologically given (Rangan and Kull, 2009). Later political ecology and the adoption of poststructural perspectives have engaged with the scale literature more comprehensively, however; the developments in consideration of networks and scales and the ‘politics of scales’ in political ecology are key examples of this (Rangan and Kull, 2009; see also 2.5). Political ecology is well-positioned to contribute to the scale literature, particularly through the emphasis such an approach places on deconstructing the local trap and investigating the socio-political complexities of the local level (Neumann, 2009). Lisa Campbell’s (2007) study of sea turtle conservation in Costa Rica is a good example of how political ecology and scalar perspectives can be combined whilst avoiding the pitfalls described above. She argues that power, seen both in the use of natural resources (material practice) and power over ideas (discourse), is crucial to understanding the case studies, but also that these power practices can be investigated through a scalar perspective. She emphasises the discursive construction and manipulation of scales, and the multi-level nature of actors operating within the conservation system around sea turtles, exemplifying the 63 approach described above to conceptualise scales as dispersed interscalar networks and tangled hierarchies (Bulkeley, 2005). The crucial nexus between the approaches of politics of scales and political ecology, to create a political ecology of scales, is found in the theme of power. In a politics of scales approach, especially a process-based approach, the socio-spatial processes of transformation and conflict between scales are conceptualised as strategic acts initiated for control over nature, and which benefit some, whilst disadvantaging others (Swyngedouw and Heynen, 2003). This argument forms the core of my discussion in chapter 6, which traces the scalar configurations of power within Tanzanian CBNRM, and the politics of scales taking place around these as actors and groups attempt to benefit from the management of natural resources and the revenue this generates. I use this political ecology of scales perspective to address the research questions for chapter 6 (see Table 1.1) and the over-arching questions in my research concerning the transformation of the prescribed governance system into that which is performed, and the ways in which some people benefit disproportionately from CBNRM. In his examination of CBFM in southern Malawi, Charles Zulu (2009) employed a similar approach and produced an excellent example of how a politics of scales approach can be combined with a political ecology focus on power relations to add insight into the socio- political characteristics and struggles ongoing in such projects. Zulu (2009: 697) focuses on “spatial and scalar configurations of CBFM, forest boundary demarcation and allocation, rule formation and scaling, and scalar dynamics of external facilitation”. He uses explanations of how scalar arrangements are expressed and the struggles by individuals to gain from these, including scale jumping, to interrogate the mechanisms underlying corruption, elite capture, decreased project accountability and resistance to the project and its perceived structural inequalities (Zulu, 2009). He concludes that a containerised vision of CBFM set out in policy that ignored these politics of scales created strong patterns of winners and losers, disrupted livelihoods and was responsible for creating tensions between modern and traditional institutions, which ultimately can be detrimental to the objectives of CBFM (Zulu, 2009) A political ecology of scales addresses three key themes that overlap with the diverse literatures discussed in this chapter; the interactions of power, agency and scale; socioecological processes and scaling; and scaled networks (Neumann, 2009). The first of these themes is seen in studies of the state and its re-organisation of scales to consolidate, regain or retain power over natural resources (Swyngedouw, 2000; Keil and Debbané, 2005), and will be examined through a policy processes perspective of reform to introduce CBNRM 64 in Tanzania. The second follows Swyngedouw’s discussion of how the production of nature is an essential part of the production of scale, and will be examined in the scalar topographies of power set out in the prescribed governance system, and their re-configuration into the performed governance system. The third theme “highlights the way networks of actors (human and non-human) transcend single spatial scales to produce new relational socioenvironmental spatialities” (Neumann, 2009: 404). This will be examined in the empirical chapters with respect to the politics of scales taking place within Tanzanian CBNRM and in the adaptation and integration of the prescribed governance system into the performed governance system. 2.6 Synthesis A socio-political research agenda for CBNRM needs to acknowledge that natural resources and the environment are inherently political, as are their management, the policies that set out their management and the people that implement such policies. The political ecology approach outlined in this chapter is centred around these ideas, and places power as the unifying theme for investigating these politics. The study of power within Tanzanian CBNRM must focus on understanding how people negotiate access to and benefits from natural resources and their management within CBNRM. Such a theorisation of power draws on poststructuralist perspectives and relationships between structure and agency outlined in this chapter to investigate the different types of power that are at work and being constructed within the implementation of CBNRM. I argue that the theme of power runs throughout the literatures on policy processes, politics of scales and political ecology, and that a postructuralist perspective of power that emphasises its discursive nature, and how such tactics are employed within both policy processes and politics of scales, draws these literatures together to add depth and breadth to the investigation of the socio-political processes within CBNRM. In the following empirical chapters I examine these socio-political realities and complexities of Tanzanian CBNRM through an examination of what I have termed the prescribed and performed governance systems. These terms draw upon the recognition of policy as being intrinsically political and process-based to address the issue of how CBNRM has been adopted within the Tanzanian policy context (chapter 5), the system that is set out in official policy discourse (the prescribed governance system), and how this has been taken on by people and put into practice (the performed governance system; chapters 6 and 7). I draw on 65 scalar theory and specifically the politics of scales literature to consider how this performance reflects struggles over power and the re-configuration of scales as part of these struggles. In the final empirical chapter I develop this idea of the performance of policy to address questions about the elements of the governance systems operational within Tanzanian CBNRM that are hidden from view, not prescribed in policy, and operate outside of the official discourse. There are several different aspects of power under consideration within my research therefore, which are each informed by the different literatures discussed in this chapter and brought together through this unifying theme of power, which is at the core of a political ecology approach (Jones, 2006). Before tackling these themes in the empirical chapters, I first outline the context of my research in Tanzania, and the methodology employed in my research. Chapter 3: Tanzanian Context 66 3.1 Introduction The focus of this thesis is on Community-Based Natural Resource Management (CBNRM) in the forest and wildlife sectors of the United Republic of Tanzania (Tanzania). Tanzania is a large28, coastal country in sub-Saharan East Africa (see Fig. 3.1). The research presented in this thesis was carried out in the region of Iringa, which is located in the southern highlands of Tanzania and forms part of the Great Rufiji river basin (National Bureau of Statistics, 2007). The locations of research are discussed further in 4.2.1. 28 Tanzania covers an area of 945,000km2 and the national census of 2002 recorded a mainland population of 33,667,659 (National Bureau of Statistics, 2007). Fig. 3.1: Map of the United Republic of Tanzania showing the Administrative Region of Iringa. Tanzania is located at 60 001S and 350 001E. 67 In this chapter I aim to introduce the context in which CBNRM emerged in Tanzania. I focus specifically on the initiation of devolved environmental management within the political system and the changing relationships between the state and local levels over the country’s history. In section 3.2, I discuss the wider political developments that took place in Tanzania during the colonial and post-independence socialist periods and later neoliberal reforms. Insight into the contemporary milieu of institutions involved in natural resource management and the political-economic and sociological factors at work are an expression of this historical context (Hurst, 2004). I discuss these political changes with respect to the role of the local level in political administration and the development of state policies of decentralisation (see 2.1.1). In section 3.3, I then discuss the historical context of the policies specific to the wildlife and forestry sectors, upon which my research is focused. The development of CBNRM in Tanzania exemplifies changing relationships between the state, society and international actors through neoliberal reforms, policies of decentralisation and the devolution of responsibility for natural resources to the local level. This was the outcome of multiple factors that contribute to the political culture in Tanzania, including the changing political context discussed in 3.2. 3.2 Devolved Environmental Management The role of the village and local29 levels in both natural resource management and wider politics in Tanzania has been a topic of discussion for decades (see Ingle, 1972; Mushi, 2001). Hyden (1980) argues that the importance of the village level throughout Tanzania’s history is based upon its autonomy, supported by its control of the subsistence base, and the maintenance of an ‘economy of affection’ (see 2.1.3), despite the state administration of a market economy. The role of the local level in political culture is more complex however, and I discuss changing state-society relations and the role of the local level in two sections below; the first relating to the colonial and socialist periods; and the second to the implementation of neoliberal reform and decentralisation. 29 Defined as the district and village levels within the political-economic administrative structures of Tanzanian government. 68 3.2.1 Devolved Environmental Management: Colonial Appropriation of Resources, Socialism and Ujamaa The period of colonial rule in Tanzania played an important role in the shaping of Tanzania’s social and political systems (Iliffe, 1979; Hyden, 1980). Tanzania’s colonial episode began in 1884 when Dr Carl Peters formed the Gesellschaft für Deutsche Kolonisation (Association for German Colonisation), under which he secured rights to over 140,000km2 of territory, which became a protectorate a year later (Lovett, 2003). The period of German colonial rule introduced the first declaration of ‘crown land’, bringing all areas not under specific ownership under the control of the state (Lovett, 2003). The transition to British colonial rule took place in the aftermath of the First World War. Germany ceded its East African territory as part of the Treaty of Versailles in 1919, and a League of Nations Mandate in 1922 placed Tanganyika (now mainland Tanzania) under British administration (Iliffe, 1979). In 1946 this relationship was altered to that of Trusteeship, until independence in 1961 (Iliffe, 1979). Both the British administration and the newly-independent government retained the German colonial system of public lands (Lovett, 2003). The colonial period in Tanzania is very strongly associated with the appropriation of land, therefore. The annexation of land was accompanied by specific forms of state-local relations in which the role of the local level was highly restricted. The role of the local level in political administration was not completely removed, but was retained through the implementation of a policy of indirect rule during the British colonial period, in which the existing local structures of power and authority in chiefs and native authorities were used to legitimise local political systems and concurrently extend the power of the newly formed state into the local level (Mniwasa and Shauri, 2001). As an African state, post-independence, political culture in Tanzania represents the blending and hybridisation of both social norms and bureaucratic forms from indigenous and Western sources (Hurst, 2004). In relation to the conservation of natural resources, particularly in the inheritance and maintenance of colonial systems of Protected Areas (PAs; see 3.3), and the role of external organisations in policy processes, these hybrid cultures, structures and processes comprised important aspects of the relationship between the state and local people, and are relevant to the conflicts and power struggles that still take place over natural resources (see 3.3). 69 Tanganyika gained independence from the British Empire on the 9th December 1961, under its first president, Julius Nyerere30, and in 1964 joined with Zanzibar to form the nation of Tanzania (Coulson, 1982; Boiesen and Lund, 2003). The newly independent nation retained many of the administrative systems put in place during the British colonial period, including and assembly of representatives overseeing the executive and responsibilities separated into sectors with corresponding line departments (Hurst, 2004). On the 29th January 1967 Tanzania passed the Arusha Declaration, which heralded the start of socialism in the country (Coulson, 1982). Tanzania has often been studied as a key example of the role of the local level in politics, due to its commitment to a socialist ideology and the focus of the Nyerere government upon rural development to achieve social transformation (Hyden, 1980). The Arusha Declaration was implemented under the slogan ‘socialism and self-reliance’, and led Tanzania to become the most committed socialist country in Africa (Coulson, 1982). The socialist period is famous for its ‘villagisation’ policy, which began in 1967 and by 1975 had brought about the relocation of more than 75% of the population into ‘ujamaa’31 villages, with the intention of creating communal production units (Boiesen and Lund, 2003). Whilst the policy began as voluntary, between 1973-5 it became compulsory (Coulson, 1982). The colonial era marked a period of state appropriation of resources which continued through to the independent administration, but the socialist period marked further important transitions in state-society relations. This is discussed below with respect to tenure arrangements at the local level and legislation aimed at increasing public participation in development. Alongside the resettlement schemes, the socialist period marked the streamlining of government administrative structures, and restriction of the role of the local level in politics through the replacement of locally-elected governments with centrally appointed civil servants (Coulson, 1982; Boiesen and Lund, 2003). The socialist period’s focus upon rural development was also accompanied by the abolition of local authorities (District Councils) and creation of District and Regional Development Councils, through the Decentralisation of Government Administration Act 1972 (Hyden, 1980). The aim of these changes was to increase public participation in development and thereby to accelerate it, but ultimately resulted in the deterioration of service delivery to rural areas and decline of urban 30 Nyerere ruled for 24 years, and in 1965 he declared the country to be a one-party state, which was recognized in law in a constitutional amendment of 1975, governed by the Tanganyika African National Union (Hurst, 2004). This remained in place until the amendment of the constitution to form a multi-party democracy in 1995 (Boiesen & Lund, 2003). 31 The concept of ‘ujamaa’, which translates from Kiswahili as ‘family ties’, is based on three assumptions: respect, common property and obligation to work (Hyden, 1980). 70 infrastructure (Ngwilizi, 2002). Whilst heralded as decentralisation, this period in fact constituted the consolidation of central power at lower levels, and is better termed de- concentration (Mniwasa and Shauri, 2001; Ngwilizi, 2002). However, through the ujamaa policy, the shifting role of the local level and restrictions described above were accompanied by the first recognition of legal rights of the village to be responsible for lands and natural resources not directly under the control of the central government32 (Nshala, 2002). The colonial and newly independent eras in Tanzania both represent complex shifts in state-local relations, marked by state appropriation of land and natural resources, the prominence of the village level, but its restricted role in political administration. In the following section I trace the shifts in the role of the local level that occurred later in Tanzania and the introduction of devolved environmental management that was facilitated by these shifts. 3.2.2 Devolved Environmental Management: Neoliberalism, Decentralisation and Local Government Reform An important step in the political history or Tanzania and the eventual devolution of rights to manage natural resources at the local level emerged in the 1980s when large shifts in Tanzania’s economic and political policy took place and policies of decentralisation and later local government reform were implemented (Mniwasa and Shauri, 2001). These changes formed part of the introduction of neoliberal reform in Tanzania, partly in response to the poor economic situation of the 1970s and 1980s33 (Mniwasa and Shauri, 2001). This economic situation drove a number of important changes, including a new emphasis on external investment through donor-government partnerships (Nelson, 2007), which doubled in real terms per capita between 1974 and 1980 (Hyden and Karlstrom, 1993). The economic situation also led to reduced funding in the natural resources sectors, where funding for wildlife, forests and fisheries received just 1.2% of the development budget between 1976 and 1981 (Kideghesho, 2006). When Mwinyi succeeded Nyerere as President in 1985, he accepted the International Monetary Fund’s Structural Adjustment Programme for induced privatization (Neumann, 1998). The implementation of neoliberal reforms began reversing the changes made through the Arusha declaration and reducing the size and roles of the state (Hurst, 2004). In 1983 the 32 The 1975 Villages Act underlies this shift (Hyden, 1980). 33 Tanzania, alongside many African countries, experienced a deep economic recession during the 1970s and 80s, contributed to by plummeting coffee prices, oil shocks and rising international interest rates creating a higher burden of debt (Hyden & Karlstrom, 1993). The domestic situation was heightened by the extremely costly war with Uganda 1978-9 (Nelson, 2007). 71 government began a policy of decentralisation, introduced in the Local Government (District Authorities) Act (LGA) 1982, and associated acts covering government administration, urban authorities, local government finances and services (Mniwasa and Shauri, 2001). The biggest change was marked by the amendment of the constitution in 1984 which sanctioned the existence of local government authorities (Mniwasa and Shauri, 2001). Mniwasa & Shauri (2001) argue that these changes must be seen in a wider context of the trends in political culture, both within the country and globally, towards increased space for human rights, rule of law, political transparency and good governance. The increased role afforded to local authorities under decentralisation was part of the twin reforms of the civil service in central government (as desired by the World Bank in 1981) and local authorities, which involved the transfer of rights and responsibilities form the centre to the local arena (Seppälä, 1998; Mniwasa and Shauri, 2001). Under the LGA (District Authorities) 1982, legal rights to hold management responsibilities for natural resources were devolved from the Ministry of Natural Resources and Tourism (MNRT) to the local (district and village) level (Boiesen and Lund, 2003). Under the same act village authorities were established or registered as the institution responsible for matters within village boundaries (Boiesen and Lund, 2003). The institutional structure implemented in this case heralded the Village Assembly as the supreme authority, and the organisation through which the Village Council is elected to oversee the day-to-day management of the village (Boiesen and Lund, 2003). The act established two tiers of local government as both “democratically representative corporate bodies with mandates to provide services and enact and enforce by-laws” (Lund, 2007: 2). Crucially, Tanzania began a process of Local Government Reform in 1996, which became policy in 1998 (Lund, 2007). This set out a government priority to improve the delivery of service to the public through decentralisation on the understanding that it would bring about greater efficiency and capacity for effecting change lies at the local level (Ngware and Chale, 2002). The reforms covered political, administrative and financial decentralisation and the redefinition of the relationship between rural districts, urban council and lower level local authorities (Ngware and Chale, 2002). Within the programme, the role of the central government was confined to one of facilitation and enabling of service provision, the development and management of the policy framework, monitoring of accountability of local authorities, financial and performance auditing and the provisions of adequate grants (Ngware and Chale, 2002). Local authorities on the other hand took on new roles and 72 responsibilities of facilitating participation, planning and executing development programmes and fostering partnerships with civic groups (Ngware and Chale, 2002). The relationship between the centre and local arenas was thus changed with the central’s role restricted to a policy-making body with supportive duties, monitoring responsibilities (Ngware and Chale, 2002). The regional administration was restructured within the programme so that the regional development directorate was replaced by regional secretariats, which were given a back-stopping role for local governments (Ngwilizi, 2002). The changing roles and reporting lines of the different levels within Tanzanian government are shown in Table 3.1 and Fig. 3.2 (Blomley, 2006). The changes involved not only reporting lines, but also financial mechanisms, whereby the government was required to make conditional and unconditional block grants to local government authorities to provide increased autonomy in financial planning and budgeting, to be in tune with local needs (Walsh, 2012). To effect these changes, amendments were made to the LGA in 1999 (Act 6) to give legal effect to the principles guiding local government reform focusing on good governance, enhanced transparency and accountability (Ngwilizi, 2002). These changes have demonstrated political commitment to the policy of decentralisation and process of devolution through strengthening the authority and responsibilities of the local level and restricting the role of the state. 73 Administrative/ Political Level Functions Number on Mainland Tanzania Village Government • Overseeing development activities at the local level • Ensuring local law and order • Enforcement of local bye-laws • Co-ordination of local planning • Overseeing land use planning and application 10,571 (registered) Ward • Co-ordinating and supporting village planning • Supervising service delivery • Ensuring integration of priorities into district plans and budgets 1,756 District Council • Maintaining law and order and good governance • Ensuring equitable and effective delivery of services to people in their areas • Raising, receiving and distributing funds in line with local development priorities 97 Regional Administrative Secretariat • Linking local governments to central ministries • Advising local governments on planning, financial management and service delivery • Monitoring and reporting local governments activities to central government 21 Table 3.1: Local Government Structure and Functions in Mainland Tanzania (adapted from Blomley, 2006). 74 The administrative rights and responsibilities of the local level under decentralisation and local government reform were supported by changes to land tenure arrangements in Tanzania as part of the Land Act 1999 and Village Land Act 1999, which came into effect in 2001 (Nshala, 2002). These established three categories of land (general, reserved and village34) in Tanzania. Village authorities, through the policy changes discussed in this section now hold 34 Village lands are classified under four circumstances; land registration under section 22 of the LGA no 7 of 1982 as within the boundaries of a village; lands designated under the Land Tenure Village Settlements Act 1965; clearly demarcated village boundaries; and land that is not reserved land and has been used by the villagers for at least 12 years before the enactment of the Village Land Act 1999 (Nshala, 2002). Fig. 3.2: Local Government Reporting Structures in Tanzania using Participatory Forest Management as an Example (Blomley, 2006). Legend for Fig. 3.3 Impact and Output Reporting Financial and Progress Reporting Ministry of Finance Ministry of Natural Resource and Tourism (Forestry and Beekeeping Division) Prime Minister’s Office-Regional Administration and Local Government Regional Secretariat Regional Catchment Forest Office District Catchment Forest Office District Council (District Forest Officer) 75 the legal right to manage natural resources on such lands, which are classified (de jure) as the property of the village. Neoliberal reforms from the 1980s, therefore, resulted in broad shifts in the relations between state and society and marked a definite increase in the role of the village within political administration, the roll-back of the state with respect to responsibilities and financial arrangements and legal security of the rights of the Village Council to manage areas of village land. In the next section I discuss this increasing role of the local level with respect to the policies for natural resource management and the eventual introduction of CBNRM in national policy from 1998. 3.3 Nature Conservation Paradigms, Policy, and Protected Areas in Tanzania The history of forest protection and wildlife conservation in Tanzania is one that has been recorded since the annexation of Tanganyika by Germany in 1891. Prior to this, little is written or known with certainty about natural resource use within the country (Ylhäisi, 2003). The long history of formal PAs began during the colonial period, which marked their growth to cover 200,000km2 (around 20% of the land within the country; Neumann, 1998; Siege, 2001). Whilst the history of forest and wildlife policies and PAs has been very separate in Tanzania, there are common themes of colonial appropriation of resources for utilisation and preservation through the creation of PAs, the exclusion of local people from natural resource management and tension between the local and state levels with regards to natural resource management (Nelson et al., 2007). In this section I discuss the development of PAs in both sectors, using the terms wildlife conservation and forestry protection to distinguish between the different objectives in these sectors; forest PAs were introduced to secure the production of timber for the state, not initially for conservation (Ylhäisi, 2003; Sunseri, 2005), whereas the wildlife sector has included an interesting mix of both wildlife utilisation and strict protection under the banner of conservation. 3.3.1 Colonial Protected Areas: Strict Protection, Wildlife Hunting & Forest Reserves The justifications for the creation of PAs in both sectors in Tanzania during the colonial era were often based on the morality of preserving these resources for the future, threats posed by disease epidemics and an increasing commercial ivory trade, the need to manage hunting 76 resources for their long-term use, and the inability of local populations to manage such resources (Neumann, 1998; Neumann, 2001; Nelson et al., 2007). However, Neumann (2004b) argues that conservation enclosures also served the twin objectives of the colonial powers to ‘civilise’ the local population and gain control over natural resources. He argues that these twin objectives were part of the larger process of ‘statecraft’ in which colonial state policies of social control and spatial segregation utilised PA enclosures as a way of ‘enframing’, fixing the boundaries between nature and culture and ultimately society and the state (Neumann, 2004b). The legitimacy of the nation-state relies upon clear boundaries and total sovereignty through the re-ordering of territory, driving the demarcation of land and natural resources, placing them and society into areas defined by different rules and relationships between man and nature (Neumann, 2004b). Thus Tanzania’s colonial history played a key role in the context of the evolution of conservation practice and policy within the country, leading Nelson et al., (2009: 301) to conclude that “the history of wildlife conservation in Tanzania is a story of increasing central control over wildlife resources across the colonial, post-independence socialist, and post-structural adjustment periods”. Like Neumann, Nelson et al. (2009) argue that the creation of PAs and the centralisation of control over natural resources, whilst arguably for conservation reasons, also served to bring rural people under state-managed systems for ‘development’. By 1896 wildlife PAs had emerged in colonial Tanzania in the form of Game Reserves (Siege, 2001), and laws were put in place requiring all hunting within the territory to be sanctioned by license. By 1913 the German colonial administration had created fifteen PAs to address fears of over-hunting (Baldus, 2001b). Similarly in the forestry sector, the advent of colonialism heralded the exclusion of local communities from resources, through the 1904 gazettement of 0.75 million hectares of Forest Reserves, where all uses required state authority sanction (Ylhäisi, 2003). Wildlife hunting has a long history in Tanzania and was one of the driving forces behind the history of wildlife conservation throughout the twentieth century (Adams, 2004). As PAs and regulations for the preservation of game were increasingly introduced, it was suggested that licences to hunt game would be a valuable way of funding conservation; nature could pay for its own conservation (Adams, 2004). The British colonial administration introduced a Game Department in 1921 to manage the country’s Game Reserves, enforce the hunting regulations and deal with problems of human-wildlife conflict (Leader-Williams, 2000). The concept of aristocratic hunting also developed rapidly at this time in East Africa, championed 77 by famous hunters such as Theodore Roosevelt and, by the 1920s, the concept of the hunting safari became very fashionable (Adams, 2004). As one of Africa’s richest countries in terms of wildlife resources, tourist hunting became an important economic use of Tanzania’s Game Reserves, bringing in fees for the government through the wildlife authorities (Leader- Williams, 2000). In the 1950s the Game Department introduced a fee structure for different species rather than a licence covering all species (Baldus and Cauldwell, 2004). Tourist hunting generally took place within Game Control Areas, although just prior to independence, Tanzania introduced Africa’s first system of hunting blocks within its Game Reserves35, which were leased to safari outfitters (see 3.3.2; Leader-Williams et al., 2009). Prior to independence, the 1940s witnessed increasing legislation of natural resource use in the Game Ordinance 1940 and National parks Ordinance of 1948 (Nelson et al., 2009). The mid 20th century saw the expansion of protected areas in Tanzania to include the gazettement of the Serengeti National Park in 1958, the expansion of the Selous and Rungwa Game Reserves, the latter of which was subsequently upgraded to a National Park (Ruaha) in 1964, and further expanded in recent years to incorporate the Usangu Game Reserve, making it the largest National Park in the country (see also 4.2.1; Nelson et al., 2009; Walsh, 2012). 3.3.2 Post-independence Continuation: Commitment to Conservation In the natural resources sectors, this period marks the start of a sharp increase in external involvement through bilateral aid agreements centered around both conservation and development, especially the agreement between the Tanzanian and German governments to start the Selous Conservation Project in the late 1980s (Nelson, 2007). This period also included large shifts in the hunting sector, where private investments increased massively between 1984 and 1985 (Majamba, 2001). The role of the natural resources sector began to change through these processes, but at the local level, more visible changes resulted from the policies for decentralisation and the introduction of local government. The commitment of the Tanzanian state to the conservation of its natural fauna and flora and the PA model to achieve this was reinforced post-independence: From 1960, the international community, through the International Union for the Conservation of Nature (IUCN), focused global attention towards the need for conservation in Africa, arguing that the destruction of its flora, fauna and habitats was the most pressing conservation problem of the time, and launching the ‘African Special Project’ to ensure conservation success 35 Previously closed to all hunting. 78 (Neumann, 1998). This led directly to the IUCN Symposium on the Conservation of Nature and Natural Resources in Modern African States (1961), where 21 African nations were represented alongside 5 international organisations (Neumann, 1998). It was at the conference that Nyerere made his famous ‘Arusha Manifesto’ pledging Tanganyika’s commitment to conservation, and inviting international experts and NGOs to assist them with this project (Neumann, 1998). This commitment to protectionist conservation in the wildlife sector was demonstrated in the publication of the Wildlife Conservation Act (WCA) in 1974, which represented one of the most crucial developments in Tanzanian wildlife policy through setting out the different categories of wildlife PA in Tanzania and their management (Nshala, 1999). The WCA 1974 replaced the Fauna and Flora Conservation Ordinance Cap. 302 (1940) and, until 1998, served as the primary legislation for wildlife (Nshala, 2002). Importantly the act vested the Director of Wildlife with the powers to oversee the management of the country’s wildlife, placed him/her in charge of all the PAs, and clearly defined nature as a ‘ward of the state’ (Nshala, 1999; Neumann, 2004a). The centralised control of forest resources was also apparent in the post-colonial period, where policy remained focused on production. The Forestry Policy of Tanzania published in 1963 aimed to ensure that Forest Reserves produced sufficient forestry products to meet domestic needs and be competitive in the global timber economy and to protect water catchments (Ylhäisi, 2003). Similarly in 1989 a new Forestry Action Plan was published, which again focused upon the creation of Forest Reserves, and whilst the participation of local communities was mentioned, it is not compulsory or detailed, and cannot be considered as community conservation (Ylhäisi, 2003). The six classifications of PAs for wildlife and forestry within Tanzania vary considerably in terms of the levels of protection and permitted uses within their boundaries. National Parks, Game Reserves and the Ngorongoro Conservation Area are the most strictly protected, whilst Game Control Areas (GCA), Partial Game Reserves and Forest Reserves are less restrictive in terms of settlement and consumptive use (Mkumbukwa, 2008). National parks are used for non-consumptive uses only, specifically tourist-related activities such as photographic tourism and accommodation whilst, within its network of Game Reserves and GCAs, Tanzania has maintained a policy of wildlife utilisation, except for during the period 1973-78 when by government notice 210 of 1973 all hunting was banned (Nshala, 1999). When 79 hunting activities resumed, the parastatal Tanzania Wildlife Corporation (TAWICO) was granted monopoly over the industry, with all powers to grant bocks, licenses etc. (Nshala, 1999). This was revoked in 1984, as part of the liberalization of the Tanzanian economy and political system, and power was handed back to the Wildlife Division and its Director (Nshala, 1999), who remains in charge of hunting activities in Tanzania to this day (see Chapter 5). Hunting takes place within a system of over 140 hunting block concessions36 allocated by the Wildlife Division (Baldus and Cauldwell, 2004), covering an area of over 250 000km2 (Nelson et al., 2009). Hunting is a multi-million dollar industry in Tanzania, which generated a gross income of $27.6 million from around 1,400 clients in 2001 (Sachedina, 2008). Tourist hunting generates the vast majority of revenue from hunting, with average income to the Wildlife Division per hunting client of approximately US $520 per day, totalling US $7,000, the majority of which is made up of a trophy fee (Baldus and Cauldwell, 2004). Resident hunting operates through a system of permits issued by District Game Officers for each administrative district. Over time the hunting safari has merged into safari tourism, with photographic tourism now accounting for a much larger number of visitors (Adams, 2004), although the large sums involved means that tourist hunting still generates larger revenues for the Wildlife Division (Sachedina, 2008). Hunting has also taken on increased significance in conservation since the introduction of CBNRM in Tanzania, particularly with the potential to channel associated revenues to local communities (see 3.3.2; Adams, 2004). The centralised history of natural resource management through the colonial periods and up to the 1980s is summarised by Baldus (2001a) as ‘conservation against the people’. The development of ‘conservation by the people’, and eventually CBNRM, is discussed in the following section, and further in Chapter 5. It is important to note that the history of both wildlife conservation and forest protection, and the restrictions placed on local communities through the enclosure of land and appropriation of natural resources at the state level had important implications for state-society relations. Within both the wildlife and forestry sectors, the colonial annexation of land and creation of PAs is associated with a prodigious number of displacements of local communities (Neumann, 1998; Ylhäisi, 2003). Conservation-related displacements are comprised of two 36 To increase revenue, the number of hunting blocks was almost doubled when the Wildlife Division took over management of wildlife hunting in 1988 (Leader-Williams et al., 2009). 80 processes: the forced removal (eviction) of people from their homes; and economic displacement, relating to the prevention of access to areas vital to livelihood strategies, and thereby enforcing a shift in economic activity (Brockington and Igoe, 2006). In many cases evictions and resettlements were the result of colonial efforts to control sleeping sickness incidence within the country, as well as to re-organise the population into ordered settlements, integrate them into the tax system and improve agricultural and labour systems (Nelson et al., 2009). For example, the creation of the Selous Game Reserve in 1922, underpinned by all these processes, as well as conservation concerns (see Neumann, 2001), led to the eviction of more than 40,000 people (Nelson et al., 2007). Similarly the history of Mkomazi National Park involved the eviction of thousands of pastoralists37, justified in the name of wildlife conservation, and which had long-term impacts upon their livelihood strategies and security (Brockington, 2002). Sachedina (2008) details the impacts of the implementation of community-based conservation in the Tarangire ecosystem of Northern Tanzania on both poverty alleviation and wildlife conservation. He argues that throughout Tanzania, the extension of state control over wildlife resources and the evictions of pastoral communities38 from the Serengeti, Amboseli and Mkomazi National Parks, together with the Maasai Mara National Reserve and Usangu Game Reserve have contributed to the deterioration of relations between local people and the state, creating wildlife wars and a “militant state of environmentalism” (Sachedina, 2008: 121). The introduction of CBNRM in Tanzania, as discussed in Chapter 5, followed many of the arguments concerning social justice and the social impacts of conservation set out in section 1.1. CBNRM also represents new forms of environmental governance in Tanzania supported by the neoliberal reforms discussed in 3.2.2. Thus, in the 1980s, both Community-Based Forest and Wildlife Management appeared in Tanzania not as a result of legislative change, but through the implementation of pilot projects, often donor-funded (Gillingham, 1998). In the wildlife sector, the Serengeti Regional Conservation Strategy (SRCS), funded by Norwegian bilateral aid from 1985 was the first example of these, followed by the Deutsche Gesellschaft für Technische Zusammenarbeit (GTZ) funded Selous Conservation Programme 37 People whose livelihoods are based at least partly around the herding of cattle (Nelson et al., 2009). 38 Attention has often been focused on the displacement of pastoral communities through PA and community conservation programme establishment. This is especially well-documented in the North of Tanzania where pastoral groups and their uses of land for grazing have been continually marginalised and largely ignored in policy processes (Goldman, 2003; Nelson et al., 2009). Furthermore, their livelihoods are being significantly altered as they are placed in competition with both conservation and agricultural land uses (Nelson et al., 2009; Goldman, 2003; Goldman, 2011). 81 (SCP). In the forestry sector, the first examples of community forestry began in Swedish International Development Cooperation Agency (SIDA) funded pilot projects in Babati and Singida districts, known as the Land Management Programme (LAMP) projects (see also chapter 5, Wily, 1997; Wily, 2000; Wily and Dewees, 2001b). By the late 1980s, several attempts at community conservation were being piloted and implemented in Tanzania, drawing upon experiments ongoing in other parts of sub-saharan Africa and the favourable environmental for Transnational Conservation Organisations and development agencies to initiate projects in Tanzania (see 3.2.2; Nelson et al., 2007). Currently, Tanzanian state- implemented community conservation initiatives relating to forest and wildlife consist of two approaches, which have grown out of the developments discussed above, but vary considerably in their approach. The first of these is the Community Conservation Service (CCS), which began in 1988 as part of the Tanzanian National Parks Authority (TANAPA) benefit-sharing programme (Walsh, 2000). It is a park outreach programme that provides funds for community development projects with the intention of increasing local support for National Parks, and decreasing infringements upon their rules (Bergin and Dembe, 1996). Despite its name, the initiative clearly involves no power devolution to local communities, and cannot be considered CBNRM. The second group represent Tanzanian CBNRM policies, consisting of Community-Based Forest Management (CBFM), which forms part of a larger policy of Participatory Forest Management, and Wildlife Management Areas (WMAs). 3.4 The Introduction of CBNRM CBNRM has been integrated within this existing hierarchy of PAs. Within the wildlife sector, the creation of WMAs has been facilitated through the conversion of GCAs on village land (Zacharia and Kaihula, 2001), whilst Forest Reserves have been delineated into those owned by the government and those located upon village land for the creation of forestry CBNRM projects (see Table 3.2). Within both these sectors the state retains ownership of all land, although a village holds tenurial rights to manage natural resources on that land. In terms of the extent of power devolution (see 2.1.1 and Appendix 1), CBNRM in Tanzania involves bundles of rights for participating communities that include withdrawal, management and exclusion (but see also 5.2), but communities do not hold full alienation rights (Shauri, 1999). 82 Forest Type Management System or Details National Forest Reserves Managed by central government for: • Protection (e.g. water catchments) • Production (e.g. mangroves, plantations) • Nature reserves • Forests of general lands managed by the government Local Authority Forest reserves Managed by Local Authority for: • Protection (e.g. water catchments) • Production (e.g. plantation and natural forests) Village Land Forest reserve • Areas of forest on village land where management is vested in the Village Council • Includes Community Forest Reserves and Village Land Forest Reserves Private Forests • Forests on village land held by one or more individuals or under customary right of occupancy • Forests on general land or village land for which the rights of occupancy or management have been leased or given to one or more individuals or a partnership for the purpose of managing the forest 3.4.1 Participatory Forest Management in Tanzania Within the forestry sector, two approaches have been developed, as part of Participatory Forest Management; Community-Based Forest Management and Joint Forest Management (Meshack et al., 2006). Both resulted from the National Forest Policy (NFP) 1998 and the Forest Act of 2002 (Blomley et al., 2008b). The development of Participatory Forest Management in Tanzania follows several years of pilot activities since the mid-1990s (Blomley et al., 2008a). The aims of CBFM are three-fold (Blomley et al., 2008a); to improve forest quality through sustainable management; to improve local livelihoods using increased forest revenues; and to improve forest governance through effective and accountable natural resource management institutions. The NFP 1998, replaced the 1963 version, and heralded substantial changes in the management of forest resources (Ministry of Natural Resources and Tourism, 2009). The main policy instrument in the NFP is the creation of Village Land Forest Reserves (VLFRs; Government of the United Republic of Tanzania, 1998b; Boiesen and Lund, 2003). The policy places clear emphasis on participation through the establishment of VLFRs, making communities both owners and Table 3.2: Forest Reserve Categories in Tanzania according to the Forest Act 2002 (adapted from Ministry of Natural Resources and Tourism, 2009). 83 managers of these forests, alongside agreements for Joint Forest Management (Ministry of Natural Resources and Tourism, 2009). The NFP 1998 also makes it clear that the changes made in forestry policy in this respect are due to the recognition of the large areas of forest that lie outside of reserved areas, on village land, over which the Forest and Beekeeping Division has no mandate for management, thus the revised strategy is a mechanism for expanding the principles of sustainable use to a much wider area (Ministry of Natural Resources and Tourism, 2009). The Forest Act of 2002 is important in that it provided the legal basis for communities to own, manage or co-manage and collect their own revenue from forest uses within their VLFR on village lands, and the state holds no entitlement to this revenue (Lund, 2007). The Forest Act and National Forest Policy were followed by the publication of Community-Based Forest Management Guidelines (2001a). 3.4.2 Wildlife Management Areas in Tanzania In 1998 the Government of the United Republic of Tanzania published the Wildlife Policy for Tanzania (WPT; Government of the United Republic of Tanzania, 1998a) 39, which set out a new category of Protected Area called a Wildlife Management Area (Government of the United Republic of Tanzania, 1998a). A WMA is defined as “an area set aside by a village government for the purpose of biological and natural resource conservation” (Government of the United Republic of Tanzania, 2002a). A WMA is therefore an area of devolved management responsibility where local communities are awarded user rights and management responsibilities for natural resources within the WMA area (Nelson, 2007). In 2003 a list of 16 pilot WMAs sanctioned by the MNRT was announced, which operated until 2006 (Nelson et al., 2007). The viability of the WMA concept was premised upon the fact that through investment agreements for the consumptive and non-consumptive use of natural resources (most commonly through tourist and resident hunting and photographic tourism), local communities would receive economic benefits, making wildlife management a competitive form of land use, and thereby incentivising the sustainable management of resources (Nelson, 2007). The stated objectives of the WPT (1998) provide the first legal mechanism “to allow rural communities and private land holders to manage wildlife on their land for their own benefit” (Government of the United Republic of Tanzania, 1998a: 14). WMAs were created as a mechanism to achieve both conservation and poverty reduction targets therefore, and were 39 Since updated in the WPT 2007. 84 strategically located in buffer and corridor zones, usually on areas previously designated as GCAs (Nelson, 2007). Following the publication of the WPT (1998), the institutional and governance systems for WMAs, the applications and gazettement procedures, rights of different stakeholders within the WMA, and how the costs and benefits from the WMA are to be distributed amongst these stakeholders were first outlined in the WMA Regulations (Government of the United Republic of Tanzania, 2002a; Government of the United Republic of Tanzania, 2005), WMA Guidelines (Wildlife Division, 2001), and the WMA Reference Manual for Implementing Guidelines (Wildlife Division, 2003). Since the publication of these policies setting out the institutional and governance structures to implement WMAs on village land, a process of recentralisation in the wildlife sector has been widely reported (see Benjaminsen & Bryceson, 2012). This trend began as early as 2000 when the Ministry of Natural Resources and Tourism (2000b; and see 2002) published an amendment to tourist hunting regulations that outlawed any investments within PAs (including WMAs) without written permission from the Director of Wildlife, including pre- existing arrangements. In later years the publication of a new Wildlife Policy for Tanzania (URT, 2007) marked an important shift in the tone of community involvement in Tanzanian Wildlife PAs and conservation. The new Wildlife Policy for Tanzania makes little reference to participation at the local level, and whilst it still endorses the gazettement of WMAs on village land, the role of the public in conservation is described as “to support the government efforts in the conservation, management, development and sustainable utilisation of wildlife resources...In addition, local communities living on village lands with viable populations of wildlife have a role in protecting and benefitting from wildlife ...by setting aside wildlife conservation areas on their land” (Government of the United Republic of Tanzania, 2007: 18). This represents a definite shift in tone from the 1998 WPT which stated that communities participating in WMAs would have “full mandate of managing and benefitting from their conservation efforts” (Government of the United Republic of Tanzania, 1998a, Statement 34). Furthermore, in the discussion about the importance of Game Control Areas as wildlife corridors, the statement that WMAs should be established “in order to secure habitat for wildlife” (Government of the United Republic of Tanzania, 2007: 24) strongly indicates that the emphasis in this new policy is not on the inclusion of local communities in wildlife conservation, but on securing wildlife populations (see also Benjaminsen & Bryceson, 2012). 85 A further major development in wildlife policy and example of recentralisation occurred when new regulations covering non-consumptive uses of wildlife resources were published in 2008 (Government of the United Republic of Tanzania, 2008), which altered revenue sharing arrangements concerning WMAs. Not only do all investment contracts related to WMA and village lands require sanction by the Director, but the revenue sharing arrangements were altered so that the AA no longer receives all revenue generated, but retains just 65%, and the remaining 35% is retained by the Regional Authorities (Government of the United Republic of Tanzania, 2008). Benjaminsen et al (in press) argue that this arrangement is further complicated by the fact that there is no clear way of knowing the total revenue generated, and members of the AA or the Village Government are unable to hold the district authorities to account with respect to this revenue sharing. Under these new revenue-sharing arrangements, communities no longer retain all revenues generated at the local level, as described above, but this process of recentralisation concerns not just revenue, but also the restriction of community autonomy in decision-making and interference of the state at the local level. In 2009 the Wildlife Conservation Act was updated (Government of the United Republic of Tanzania, 2009) and provides further evidence of this recentralisation taking place within the wildlife sector. The new act strengthens state control of wildlife and increases the authority of the Wildlife Division to interfere in management of wildlife resources on village land (Benjaminsen & Bryceson, 2012). This occurs principally through prohibition of hunting and grazing within a GCA without permission from the Director of Wildlife, and the involvement of the Wildlife Division and District Authorities in the negotiation and signing of investment contracts for a WMA (Government of the United Republic of Tanzania, 2009). The policy environment today for WMAs is evidently very different from that published in 1998, and the recentralisation discussed above clearly has important implications of the way power is currently devolved in WMAs (see also 5.2.4). Benjaminsen & Bryceson (2012) and Benjaminsen & Svarstad (2010) discuss this recentralisation as a result of the changing roles of external organisations and NGOs in the wildlife sector. Whilst such organisations were critical to the development of WMAs and the policies that first set them out (see 5.3), their involvement in wildlife policy has reduced markedly in the intervening period, partly due to accusations of corruption within the wildlife sector (e.g. Sachedina, 2008, Nelson, 2009, Nelson, 2010) and the perceived failures in the implementation of the 1998 WPT (Benjaminsen et al., in press). Without the pressure from these donors to maximise power 86 devolution, the state has sought opportunities to re-capture and expand its power and revenues sources associated with the management of wildlife resources, resulting in the policy amendments described above. 3.5 Summary The political context in Tanzania with respect to the control and management of natural resources has undergone large shifts over the past 120 years. Tanzania has undergone major political changes during this period, with corresponding impacts on the relationship between the state and the local levels and the role of external agents in both conservation and development areas. With respect to natural resources Tanzania has gone through transitions of centralised control to devolved environmental management and in conservation there has been a shift from strict protection towards community conservation, and specifically CBNRM. The emergence of CBNRM in Tanzania is a product of this political history, and specifically the pathway to devolved environmental management is a result of the adoption of neoliberal reforms in Tanzania, the roll-back of the state and political decentralisation that has occurred since the 1990s. Chapter 4: Methodology 87 4.1 Ontology, Epistemology & Methodology In this thesis I adopt a political ecology approach to the study of socio-political processes in the development of, and ongoing in the governance of two case study Community-Based Natural Resource Management (CBNRM) projects in Tanzania (see 1.4 and 2.6). “If you can measure it, that ain’t it” Kaplan (1964) in Berg (2009: 2). My empirical research is qualitative, an approach which can be typified as concerned with an “intersubjective understanding of knowledge, [an] in-depth approach, [a] focus on positionality and power relations, [and] contextual and interpretive understandings” (Dwyer and Limb, 2001: 6). Qualitative approaches have risen to prominence in geographical studies, and political ecology specifically, as part of broader changes in the discipline. This followed the critique of spatial science and the subsequent foci, firstly on Marxist enquiry and “the constraining effect of social, economic and political structures” and, secondly, on humanist approaches which study the “mediating influence (on spatial patterns and processes) of human agency” (Cloke et al., 1991: 15). These different approaches have been part of wider epistemological, ontological and methodological debates concerning the inter- twining of structure and agency (see 2.3.1), which in turn have been key factors in the development of political economy and, subsequently, political ecology (see 2.5; Kitchin and Tate, 2000; Neumann, 2005). Since the 1990s poststructuralist approaches to political ecology have sought to “destabilise both the overarching structuralist explanation of Marxist geography and the fiction of the integrated humanist subject that denies multiple different subjectivities” (Dwyer and Limb, 2001: 5), and to focus upon “discursive and symbolic politics and the institutional nexus of power, knowledge and practice” (Walker, 2005: 75). Themes central to this thesis, especially the situated nature of knowledge, multiple perceptions of environmental issues and the primacy of power in understanding access and distribution to natural resources, and the discursive construction and reinforcement of these power structures have become central themes within research that describes itself as political ecology (Brockington, 2002; Watts and Peet, 2004). Qualitative procedures “provide a means of accessing unquantifiable facts about the actual people researchers observe and talk to...Researchers using qualitative techniques examine 88 how people learn about and make sense of themselves and others” (Berg, 2009: 8). Ontologically, qualitative research “does not start with the assumption that there is a pre- existing world that can be known, or measured, but instead sees the social world...as intersections of cultural, social, political and economic processes” (Dwyer and Limb, 2001: 6) and the researcher “seek[s] subjective understanding of social reality rather than statistical description or generalizable predictions” (Dwyer and Limb, 2001: 6). The adoption of a qualitative methodology was deemed appropriate to my research as I aim to provide an in- depth understanding of socio-political features and processes that are shaping the governance systems and outcomes of the case study projects; characteristics which I did not feel could, or should, be measured quantitatively. I began fieldwork with a list of research questions and themes to explore, but no hypotheses to test. As such, the research project has been designed to be inductive, aiming for abstractive theory generation and grounded research, rather than theory verification. Grounded theory argues for a research strategy that is “open to what the site has to tell us, and slowly evolving a coherent framework rather than imposing one from the start” (Glaser and Strauss, 1967; Miles, 1979: 117; see also Punch, 1998). I began this research project with little knowledge of Community-Based Natural Resource Management (CBNRM) in Tanzania and no prior knowledge of the case study projects, and by the start of my fieldwork phases my knowledge of the policy framework and case study projects had developed through extensive reading and interviews conducted with conservation practitioners who had worked in Tanzania, and also on the case study projects. My research questions, therefore, had been guided by the theoretical framework and contextual knowledge developed prior to fieldwork, but the emergence of new information and research themes whilst in the field was both a product of the ideological and methodological approaches adopted, and critical to the findings of the project. In this chapter I outline the approach taken in my research and the specific methods employed. I begin, in 4.2, with an introduction to the research area and the case study CBNRM projects. In 4.3 I then discuss the specific research techniques used and in the final section I discuss methodological issues and how these were addressed in my research. 4.2 Research Design 4.2.1 Case Study Selection The case study approach has become a common aspect of qualitative research and is valuable in the overlapping epistemology that is “orientated towards analysing concrete cases in their 89 temporal and local particularity, and starting from people’s expressions and activities in their local contexts” (Flick, 2002: 13). The approach is often criticised for its ability to produce in- depth information about a specific locality, but its lack of generalisable findings. Whilst this is inevitably true, the value of the research undertaken in this project lies not in the development of a widely applicable theory or model of governance processes, but in the recognition of the critical role of the socio-political processes and struggles inherent to the governance of CBNRM projects. I argue that it is the articulation of local context within the wider policy framework which holds the issues of interest for understanding these socio- politics of governance, and this necessitates investigation of these issues at the case study level in order to access the day-to-day management of these projects within the villages that run and participate in them. I carried out research into the twin policies of Community-Based Forest Management (CBFM) and Wildlife Management Areas (WMAs) in Tanzania across multiple levels spanning the national, ministerial, policy-making level and its associated institutions at regional and district levels. I then conducted in-depth research in two case studies; one CBFM project, and one WMA project. Both of these case studies are located in Iringa region, in the Southern Highlands of Tanzania (see Fig. 4.1). Iringa region is the 7th largest region in the country (National Bureau of Statistics, 2007) and has long been recognised for ecological significance as a transition zone between the Sudanian Acacia-Commiphora vegetation community of East Africa and the Zambezian Brachystegia vegetation of Southern Africa40, and Ruaha National Park, which falls within the region also contains Tanzania’s second largest elephant population (Hartley, 1997). The region is comprised of seven administrative districts (Iringa Rural, Kilolo, Makete, Mafindi, Njombe, Ludewa and Iringa Urban). The research presented in this thesis was carried out within the district of Iringa Rural which is made up of six administrative divisions, 20 wards and 119 villages (National Bureau of Statistics, 2007). Iringa rural is the largest of these districts and, in 2002, had a recorded population of 245,033 (National Bureau of Statistics, 2007). 40 The region falls within the tropical and sub-tropical grasslands, savannas, shrublands and woodlands biome (Burgess et al., 2004). 90 Fig. 4.1: Map of Iringa Region, Tanzania. The two case study areas are indicated in further detail in Figs 4.2, 4.3 & 6.4. 91 The region of Iringa was named from a Hehe word ‘lilinga’ meaning fortress, and refers to the past struggles between the people of the region and other powers, including the colonial administrations (Institute of Resource Assessment, 2007). At the regional level, Iringa has the second highest level of income per capita (after Dar es Salaam), and is primarily an agricultural region, with commercial production of tobacco, tea, coffee and dairy products, although the majority of production within the region consists of small-scale subsistence and local market farming (HIMA, 2000). The Iringa Rural district economy is also predominantly agricultural, with 95% of the population estimated to be engaged in subsistence agriculture (Health and Development International Consultants (HDIC), 2010). The principal market crops are rice, maize and potatoes (Health and Development International Consultants (HDIC), 2010). In terms of natural resources, the region contains significant resources in both the wildlife and forestry sectors. Almost half of the region is covered in woodland41 (National Bureau of Statistics, 2007) and there are 20 catchment and local authority forest reserves, predominantly located in Kilolo and Mufindi districts and covering over 1,700km2 (Lovett and Pócs, 1993)42. The district of Iringa Rural is dominated by Ruaha National Park, now the largest in the country and originally created from the colonial Saba River Game Reserve (first designated in 1910) and gazetted as a National Park in 1951(Neumann, 1998). Iringa Rural also hosts the Lunda Game Reserve and the MBOMIPA Wildlife Management Area (Institute of Resource Assessment, 2007; National Bureau of Statistics, 2007). Ethnically, the district is dominated by the indigenous Hehe and Gogo communities, with smaller numbers of Bena, Kinga, Wanji, Safwa, Kimbu, Nyamwezi and Ngoni present (Ministry of Natural Resources and Tourism, 2003). The Kosisamba group is made up of communities that were removed from within Ruaha National Park and is distinct to the district (Walsh, 1995). There are significant pastoral populations (predominantly Il-Parakuyu Maasai), particularly in the lowland zones surrounding the Little Ruaha River (in Pawaga division), who have been resident in the area from the 1950s, followed by smaller numbers of Barabaig and Sukuma livestock herders (Walsh, 1995). Many of these pastoral groups have adopted cultivation alongside livestock keeping (Walsh, 1995). 41 Miombo and acacia woodland with varying canopy cover, and including natural grasslands (National Bureau of Statistics, 2007). 42 Although the largest of these forest reserves (Udzungwa Scarp and West Kilombero Scarp) are shared with Kilombero district, which is in the region of Morogoro (Lovett & Pócs 1993). 92 The case studies used in my research were selected for the opportunity to work on two case studies within the same administrative district (Iringa Rural) and under the same district authorities’ office, similar timescales of project histories and development, and a scarcity of previous research in this geographical area. Within Iringa region, there is currently one WMA project (MBOMIPA) that has reached a stage in the application process that permits a study of this kind. At the initiation of the research project there were 15 villages in the district that had reached a feasible stage in the CBFM application process, although the first phase of fieldwork revealed that only five of these were active43. The research carried out in the case study projects developed and was refined throughout the course of the research project, drawing upon existing literature and policy documents to identify the institutions involved and socio-economic characteristics of the communities A preliminary qualitative survey of all 21 WMA villages and 5 CBFM villages was completed and the potential villages for further in-depth research were reduced for logistical and safety reasons and final selection was made on the basis of socio-economic comparability, project revenue comparability and the reputation and status of the villages within the CBFM and WMA networks at the district, region and national levels. The case studies and selected villages are introduced below. 4.2.1.1 WMA Case Study The case study WMA project is located within the district of Iringa Rural within the site of the former Lunda-Mkwambi Game Control Area44 (GCA, see Fig. 4.2), which is located along the eastern border of Ruaha National Park around 130km from the town of Iringa (Health and Development International Consultants (HDIC), 2010). The project began as a British Overseas Development Association bi-lateral aid programme named the Ruaha Ecosystem Wildlife Management Project (REWMP), which has been described as a “classic integrated conservation-development project” (Hartley, 1997: 2). REWMP’s mission comprised two major foci; the first was to assist in the planning for the adjacent Ruaha National Park, for which the project villages provided a buffer in the previously designated Lunda-Mkwambi GCA (Hartley, 1997). A smaller focus was the development of a sustainable wildlife utilization scheme aimed at providing benefits for communities 43 The remaining ten were not producing revenue at the time as they were closed for regeneration of the forest. 44 The southern portion of this Game Control Area has been re-gazetted as a WMA. The Northern section remains under government authority. 93 surrounding the park (Hartley, 1997). REWMP was not the first project of this kind in Tanzania, but followed the models applied in both the Serengeti Regional Conservation Strategy (SRCS) and Selous Conservation Programme (SCP; Hartley, 1997). 94 Fig.4.2: REWMP/MBOMIPA Case Study Area. This map does not show the final two villages to join the WMA (Kitisi and Magombwe), which were both formerly sub-villages of Idodi and Isele respectively. 95 REWMP ran from 1992 to 1996, and Hartley’s (1997) review of the project shows that whilst the project is widely considered as a success, there were several periods of tension and difficulty, relating to the new arrangements for hunting in the area, the exposure of corruption and institutional weaknesses and the setting and sale of hunting quotas (Hartley, 1997). REWMP officially ended in mid-1996, and the project was transformed into MBOMIPA45 from October 1997, just six months before the Wildlife Policy for Tanzania (WPT) was published (Walsh, 1998). MBOMIPA began as a collaboration with both the Wildlife Division and the Tanzanian National Parks Authority (TANAPA) with financial support continuing to come from the Overseas Development Association, now re-named as the Department for International Development (Walsh, 1998). International funding ceased in 2002, with the project operating as an unofficial pilot of the new WMA policy (Interview P74). From 2003 the Worldwide Fund for Nature (WWF) became a facilitator for implementation of the WPT nationally, and began to assist MBOMIPA with the long procedure of securing final gazettement of the WMA (Walsh, 2003). WWF acted as technical advisors to MBOMIPA and negotiators between the project and the government authorities, through their Tanzania Policy Implementation Programme Team (Walsh, 2003). In January 2002 MBOMIPA was the first WMA pilot project to register its Authorised Association (AA; Walsh, 2003), although official completion of the application procedure took place in 2007, when the WMA was gazetted and the MBOMIPA AA received its user rights. In a 2007 assessment of the WMA pilot projects (Institute of Resource Assessment, 2007), MBOMIPA was listed as a gazetted WMA with no allocated user rights. Fieldwork carried out in the WMA villages during 2010 showed that these have since been granted by the Wildlife Division, and the project has now secured several investment agreements with both hunting and photographic tourism outfitters. In the last financial year (2009-2010) MBOMIPA accounts show that the project generated over Tz Sh. 160,000,000 (around $110,000), which provided approximately $1,590 for each of the 21 participating villages to use as they wish. MBOMIPA is well- known for its record of generating income, initially from the sale of a wildlife quota to resident hunters under REWMP and subsequently MBOMIPA (Hartley, 1997; Walsh, 2003; Institute of Resource Assessment, 2007), and later greatly increased by the 45 ‘Matumizi Bora Maliasili Pawaga na Idodi’ meaning sustainable use of natural resources in Pawaga and Idodi. 96 incorporation of revenues from the Wildlife Division tourist hunting block within the previous GCA. MBOMIPA WMA is an association of 21 villages, which are split between the administrative divisions of Idodi and Pawaga, as shown in Fig. 4.2 and Table 4.1. Administrative Division Village Name Idodi Mahuninga Makifu Tungamalenga Mapogoro Idodi Malinzanga Mafuluto Nyamahana Kitisi Pawaga Isele Magombwe Mbuyuni Kimande Kisanga Kinyika Mboliboli Itunundu Mkombilenga Ilolo Mpya Magozi Luganga MBOMIPA has been selected as the case study project due to its long history as a CBNRM project, its inclusion within the pioneer group of such projects (URT, 2007) and its presence during the critical period of policy formation in the late 1990s. It has also been selected because it has stood the test of time, and remains a celebrated example of the WMA policy, due in part to the income the project is generating, the significant area covered by the WMA (776.65km2) and its proximity to what is now the country’s largest national park, the potential for managing wildlife for community benefits in the area (Taylor, 1995; Walsh, 1995) and the large number of villages involved (21). The village selected for further research within MBOMIPA was Makifu. This village was selected according to the findings of the Table 4.1: Villages Participating in MBOMIPA by Administrative Division. 97 preliminary qualitative survey, which highlighted Makifu’s strategic position within the political struggles taking place within the WMA at the time, its average population size amongst the participating villages, alongside logistical reasons for creating a base for research. The preliminary qualitative survey also led to the decision to expand the fieldwork undertaken during the second phase to include further interviews, alongside the in-depth research undertaken in Makifu, in five other MBOMIPA villages. These villages were Tungamalenga, Nyamahana, Luganga, Mbuyuni and Itunundu. The justification for this decision and the selection of these particular villages was based on the findings of the preliminary qualitative survey, the supra-village nature of the governance system implemented in WMAs, as discussed in 5.2.3, and the power struggles taking place within the MBOMIPA WMA, as described in 6.3.3.2. 4.2.1.2 CBFM Case Study Participatory Forest Management began in Iringa Rural under the Danish International Development Agency (DANIDA)-funded MEMA (Matumizi Endelevu ya Misitu ya Asili46) project, which began officially in June 1999 (MEMA, 2001). The overall project consisted of two jointly-managed projects; the Udzungwa Mountains Forest Management and Biodiversity Conservation Project and the Community-Based Natural Woodlands Management Project, both of which were located within Iringa Region (MEMA, 2001). The objective of MEMA was to “develop community-based sustainable management of natural forest and woodland in Iringa district, and through this effort to conserve valuable biological diversity and improve the welfare of rural poor communities” (MEMA, 2003: iii). This was to be carried out through two immediate objectives; to “develop, test and implement widely applicable Participatory Forest Management models for environmentally sustainable production, use, management and protection of natural forests and woodlands in the pilot areas”; and secondly to “support capacity development in natural forests, woodlands and biodiversity resource management in Iringa District” (MEMA, 2003: 1). MEMA targeted the forests and forest users of 23 villages, and its strategy was to establish and utilize Village Natural Resources Committees (VNRCs) to prepare forest management plans, working alongside MEMAs five zonal planning teams (MEMA, 2003). The project aimed to establish Joint Forest Management within central and local government forest reserves, and CBFM on village lands, which would then become declared Village Forest Reserves (MEMA, 2003). The Danish government has a long history of involvement in biodiversity conservation work 46 Meaning ‘sustainable use of natural forests’. 98 in the Iringa region, for example through academic research carried out between the University of Dar es Salaam and the University of Copenhagen (MEMA, 2001). The funding was supplied by the government’s Environment, Peace and Sustainability Facility (MEMA, 2003). Implementation was carried out with the Forestry and Beekeeping Division and Lands and Natural Resources Office of Iringa District Council (MEMA, 2003). A map of the case study area for CBFM, which is located within the former Natural Woodlands Management Project, is shown in Fig. 4.3. The process leading to MEMA began in 1992, when the Royal Danish Embassy fielded a project preparation mission to the area of the Udzungwas, thereby initiating support for the project (MEMA, 2001). This was followed by a feasibility report in 1993 (MEMA, 2001). In 1997, a request to DANIDA was received from the Regional Commissioner’s Office in Iringa, with support from a previously ongoing project called HIMA (Hifadhi ya Mazingira47), which asked for a revised project (MEMA, 2001). The following project identification mission in March 1998 recommended numerous forestry-related projects in the region, including foci on the Udzungwas and natural woodlands areas (MEMA, 2001). The rationale for such projects was identified as the pressure being exerted on these forest areas by the neighbouring Iringa town, and the need for design and implementation of sustainable use models (MEMA, 2001). 47 Meaning ‘conservation of the environment’. 99 Although both the Udzungwa Mountains Forest Project and the Natural Woodlands Managment Project were guided by the same principles, their specific aims differed. The Udzungwa concentrated upon conservation, whilst the natural woodlands project worked in forests that were “meant for production”, and therefore included harvesting forest products (Massao, 2003: 3). MEMA’s first phase ran from June 1999 to May 2002, and concentrated on the preparation of JFM and CBFM management plans, which were completed on the 19th June, 2002 (Massao, 2003). The second phase ran from June 2002 to May 2003, and its focus was to test the management plans and make necessary revisions (Massao, 2003). The project was then extended further to the end of 2003 to assist the transition from MEMA to Fig.4.3: CBFM Case study area. The villages of Kiwere, Mfyome, Kitapilimwa, Kinywang’ang’a and Itagutwa and Ikengeza form part of the previous MEMA Natural Woodlands Management Project and are currently implementing CBFM on village lands. 100 national policy-aligned Participatory Forest Management projects, and the seven month extension activities included revision of the management plans, continued monitoring and evaluation and the hand-over of project assets to the Forestry and Beekeeping Division and Iringa District Council (Massao, 2003). The initiation of MEMA in 1999 took place in between the publication of the National Forest Policy of 1998, which promoted substantial change in forest management, introducing Village Land Forest Reserves (VLFRs), National Forest Reserves and Local Authority Forest Reserves (Ministry of Natural Resources and Tourism, 2009) and the Forest Act of 2002, which provided the legal basis for villages to own, manage or co-manage VLFRs and to create forest management by-laws through the village forest committee (known as the VNRC or Village Environmental Committee; Ministry of Natural Resources and Tourism, 2009). The authority of these village-level institutions, and their responsibility to manage lands designated as belonging to the village, is laid down in the local government reforms of 1982 and the subsequent land act of 1999 and village land act of 1999 (Ministry of Natural Resources and Tourism, 2009). The CBFM village selected for further research was Kiwere. This village was selected based on the project income and its comparability to the revenues associated with MBOMIPA at the individual village level, and also due to the location of previous research within the five potential villages, in order to avoid research fatigue amongst the participants. The village is located approximately 20km from Iringa town, within the division of Kalenga. The forest has an area of 4904Ha, and the village completed the application procedure and its forest, called Kidundakiyave, was officially gazetted as a VLFR in 2002 (Village of Kiwere, 2002) Notable natural resources in the forest are wood for firewood and timber (plus other uses such as poles for house construction), stones and wild animals (Village of Kiwere, 2002). Discussions with the village authorities confirmed that the forest is divided into several zones within its management plan, including zones for harvesting, and those closed for protection of water catchments and sensitive areas on steep hills or near water sources (Focus Group P195). The management plan for the forest sets out the rules for use of the forest, including a timetable for access for firewood collection and other activities (such as mushroom collecting), and prices for permits to carry out harvesting activities such as firewood collection for tobacco curing, charcoal making and timber felling (Village of Kiwere, 2002). 101 The management of the forest is overseen by the Village Natural Resources Committee48, which is part of the Village Council49 structure. 4.2.2 Fieldwork Timetable The research project was split into two major periods of fieldwork. The first began in March 2010 for a period of six months, concentrating upon research at the national, regional and district levels and also included the preliminary qualitative survey of the villages involved in CBNRM in the region. The second period of fieldwork was five months long, beginning in January 2011, and focused particularly upon the village level and repeat and extended rounds of interviews at the district and national levels. 4.2.3 Research Outside of Tanzania During the course of my research, data collection took place in the UK, Denmark and Tanzania. In order to undertake research in Tanzania, I made contact with the Sokoine University of Agriculture in Morogoro, Tanzania and became a Research Associate in the Department of Wildlife Management for the duration of my fieldwork under the guidance of Professor Alex Songorwa. Through this connection to the university, I used a snowball approach to identify and carry out further interviews with academics and conservation practitioners at the national and international levels. I also travelled to Denmark and worked as a visiting scholar at the centre for Forest and Landscape at the University of Copenhagen. 4.2.4 Areas of study The collection of data was split according to the levels of analysis and according to different areas of study. Across the levels of analysis four main areas of study were investigated. The first involved the history and development of the policies underlying the CBFM and WMA projects under study. This theme also included the history and development of the case study projects themselves, and their roles in the formation of national policies in the forest and wildlife sectors. The second theme addressed the institutional and power structures in place in these policies, as seen through policy documents and in the implementation of these policies in the case studies. The third area of study focused on unofficial or ‘hidden’ characteristics of the governance structures. These were studied through examination of the outcomes, problems and conflicts within the case study projects, as perceived across the 48 Kamati ya Maliasili 49 Serikali ya Kijiji 102 multiple levels of analysis, alongside cultural aspects of power within the projects. Focusing on issues of outcomes, problems and conflicts, within this area of study allowed me to address the areas set out in the objectives of the research, whilst also addressing the sensitive issues that came out of the research in a non-confrontational manner. Finally, to investigate the internalisation of power and the nature of power as a political technology, particular attention was paid to the discourses employed by participants across all levels. 4.3 Methods employed Multiple methods were used within and across these areas of study and during the two periods of international fieldwork. This began with a series of repeat semi-structured interviews with international conservation practitioners based both in the UK and Denmark, who had worked on the case study projects during their phases as donor-funded projects (see 4.2.1.1. and 4.2.2.2). These initial interviews also served as the sources for collation of project documents and reports for the case study projects, national policy documents and relevant literature (see 4.3.1). I used a snowball approach to identify further respondents of relevance to the history of CBNRM policies in Tanzania and the case study projects. During the first phase of fieldwork a preliminary qualitative survey was carried out at the village level and semi-structured interviews were carried out across all institutional levels. During the second phase of fieldwork, further semi-structured interviews were carried out across all institutional levels, focus groups and participatory activities with different stakeholder groups and participant observation were carried out at the village level, and a series of local assistant reports were collected in the CBFM research village (see 4.3.7). Interviews at the national, regional and district levels and with NGOs were arranged in person or by telephone and confirmed the day before the scheduled meeting. Interviews were usually conducted in the office of the participant, and lasted between 45 minutes and 90 minutes. Data collected at the village level were organised through initial introductions and explanations of the research at the village offices. Interviews and focus groups with village officials were completed first, and used to identify the first groups and individuals within the community of interest to the study. Subsequent interviews and focus groups were organised in advance in person. The length of focus group discussions did not exceed 1 hour 45 minutes and interviews at the village level ranged between 25 and 90 minutes in length. A 103 summary of the data that was collected is shown in Tables 4.2 and 4.3 and a schedule of this data is included in Appendix 2. Level Semi-Structured Interviews Preliminary Qualitative Surveys Focus Groups International 8 National 5 NGO 7 District 4 MBOMIPA 5 Village Makifu Village Authorities Makifu Village Residents 5 16 21 2 4 Tungamalenga Village Authorities 2 Tungamalenga Village Residents 2 Nyamahana Village Authorities 2 Luganga Village Authorities 2 Mbuyuni Village Authorities 3 Itunundu Village Authorities 3 Total 64 21 6 Table 4.2: Semi-Structured Interviews, Qualitative Surveys and Focus Groups Carried out in the Wildlife Sector. 104 Level Semi-Structured Interviews Preliminary Qualitative Surveys Focus Groups International 4 National 5 NGO 2 Regional 2 District 3 Village Kiwere Village Authorities Kiwere Village Residents 8 20 5 3 8 Total 44 5 11 4.3.1 Document Analysis Documents used to further understanding of the policies underlying CBNRM in Tanzania and their implementation across the country were collated from academics at the Sokoine University of Agriculture, Government of Tanzania (Ministry of Natural Resources and Tourism), conservation practitioners, Non-Governmental Organisations (NGOs) and the case study projects themselves. The documents used in the research project are shown in Appendix 3. These documents were used to further understanding and to facilitate the design of the research project, and where relevant they were analysed in Atlas.ti for textual data and discourse purposes. REWMP/MBOMIPA and MEMA project documents that were written during the donor-funded periods of the projects were analysed to build up a picture of the history of the projects and the context from which the current project emerged. Conference proceedings and policy documents were analysed for the evolution of policy over time and the discourses surrounding this. Contemporary reports written by NGOs, government ministries and from the projects themselves (including accounts and investment agreements) were used for specific data they contained regarding the projects and the discourses employed within them. 4.3.2 Village Preliminary Survey During the first phase of fieldwork, following the interviews carried out at the national, regional and district levels, a qualitative survey of 26 villages (21 WMA villages and 5 Table 4.3: Semi-Structured Interviews, Qualitative Surveys and Focus Groups Carried out in the Forestry Sector. 105 CBFM villages) was carried out. The purpose of this survey was to introduce the research project to all villages involved in these projects, assess the feasibility of further study in each of these villages and to collect data on village socio-economic characteristics, project management at the village level and project performance. The format for the survey was a group interview, the participants in which varied between the villages but included representatives from the Village Council, Village Natural Resources Committee (VNRC; for CBFM villages) and MBOMIPA village representatives (WMA villages). The interview schedule included a list of topics of interest and specific questions relating to socio-economic characteristics of the village, but no pre-prepared question structures or topic order were used (Kitchin and Tate, 2000; Berg, 2009). 4.3.3 Focus Groups Focus groups bring together a small group of people to discuss a topic, although their characteristics vary widely (Morgan and Krueger, 1993). The focus group was chosen as an appropriate method in this study to explore the social meanings of these policies and to provide a source of information on the ways people construct and debate knowledge relating to these issues (Hoggart et al., 2002). The group context is an important feature of social processes (Kitzinger, 1994), and focus groups provide a socially sanctioned space to examine and induce the articulation of generally assumed and unstated norms and values (Bloor et al., 2002). Focus groups were used in the two villages selected for in-depth study (Kiwere and Makifu). This process began with a focus group with representatives from the village authorities, with the objective of collecting basic information about the project and to facilitate the identification of stakeholder groups within the projects. These stakeholder groups were then used as break characteristics for focus group segments (Morgan, 1998). Inter-group heterogeneity and intra-group homogeneity were implemented in order to limit the impact of intra-group power relations as much as possible (Morgan and Spanish, 1984; Cameron, 2005), with the objective of facilitating discussion (Knodel, 1993). The segments used for the focus groups were: CBFM research village: • Resource user groups (charcoal, tobacco, beekeepers) • Pastoral groups using the forest • Gender • Official role in the project (village natural resources committee leaders, village natural resources committee members , village forest scouts) 106 WMA research village: • Official role in the project (village natural resources committee leaders, village game scouts) • Pastoral groups using the forest • Gender • Sub-villages within the entire village Focus groups were repeated with these characteristics until data saturation was reached. Particular attention was paid to the participants involved in these groups, their socio- economic status and the area of the village they came from to ensure geographic and social- group coverage to as wide an extent as possible. Focus groups were held within the village in a public yet quiet space that was deliberately chosen to be far from the village offices and homes of the leaders of the VNRC, as it was felt that these could influence the spatial aspects of participation, including participants’ perceptions of the study and the appropriateness of people’s responses (Kesby, 2007). In order to encourage discussion in the focus groups I used an invitational approach to participant recruitment, whereby we identified one member of the target stakeholder group and asked them to find between five and seven further members of the same stakeholder group who would be willing to participate in the focus group discussion. This was a particularly successful method of recruiting participants but had two implications that needed careful attention; firstly this meant that the respondents were highly likely to know each other, which is usually seen as an undesirable characteristic of the method, but in the context of the research villages it was clear that this could not realistically be avoided; secondly the invitational approach meant that the participants were more likely to have close links to the individual that we first contacted, and were more likely to come from similar socio-economic groups and to live in close proximity to one another. To deal with this, we had to take careful notes of the participants’ socio-economic status and which sub- village they came from, using a map of the village, and following up the discussions with visits to the participants’ houses to qualitatively assess their socio-economic status (according to the wealth-ranking activity that was carried out in the village) to ensure that a illustrative sample had been created (McClintock et al., 1979; Flowerdew and Martin, 1997). Focus groups followed a semi-structured format using a list of topics of interest. As with all other forms of discussion-based data collected, the format of the focus groups depended upon the participant group, and the planned topics to be discussed had to be altered according to different groups’ uses of the natural resources governed by the projects, and the level of understanding of different groups. I assessed this level of knowledge using broad, ‘throw- 107 away’ questions (Berg, 2009) at the beginning of the interview. For example, during introductions and explanations of my research I would ask general questions about the objective of the CBNRM project and how they were involved in and affected by the project. The semi-structured approach was very effective for tailoring the discussion to the knowledge base of the participants and allowed new themes and topics to emerge from the data. The one major obstacle I faced with using focus group discussions was that I had to alter their format in line with cultural norms in the villages that I worked in. It became very clear early on in the research that a traditional style focus group where participants discuss amongst themselves was an inappropriate expectation in the context of my research and mine and my assistant’s roles had to be altered along the lines of more active facilitation in order to create socially-acceptable channels to express different opinions. This was partly because cultural norms often prevented debate or disagreement (see 7.4). I had to pay careful attention to the subtle ways in which disagreements were voiced during such discussions therefore, and to carefully select the break characteristics in order to facilitate discussion as much as possible within the groups. Where I felt that individuals were restrained within the group discussion, or where interesting comments had been made but could not be explored fully because of these social norms, individual interviews were arranged with respondents to discuss issues in further detail. It also became clear that the respondents were happier with a group interview session, rather than a discussion, and a compromise had to be made between my own objectives and the wishes of the participants. In forging this compromise, my assistants and I paid close attention to the ways in which we could build trust and facilitate discussion with respondents, particularly by not taking lots of notes during the focus groups and by spending time at the beginning and end of the discussion engaging in more general conversation with the group. 4.3.4 Semi Structured Interviews Within the study, I used semi-structured interviews in three ways: firstly they were used across the village, district, regional, national and international levels as a primary method to collect data from key individuals involved with the case study projects and policies. These interviews took places in several stages, organised according to levels within the governance structures of the policies, and were repeated whenever further information was required or new themes discovered of relevance to the participant. Secondly, interviews were used at the village level to investigate themes of interest that arose from the focus group and interview 108 data in more detail. In some cases this was carried out when a comment of interest was made in a focus group, but which was deemed inappropriate for further discussion within the group. Semi-structured interviews were also used when further individuals of interest to the study were revealed through the course of fieldwork. The interviews followed both a purposive and snowball sampling strategy, therefore, to create an illustrative sample (Flowerdew and Martin, 1997). All interviews were semi-structured, using a topic-guide, as with the focus group and qualitative survey methods employed. Through reflection and practice, attention was paid to reducing the instances of leading and complex questions, and to alter the phrasing of questions in line with discoveries of socio-cultural meanings and level of knowledge of the participants (Singer et al., 1983; Smith, 1987; Foddy, 1993; Smith, 1995; Parfitt, 2004). I had to develop my skills as an interviewer to ensure that the interviews remained semi- structured, allowing new ideas to be contributed to the project and new topics explored as they arose, but also to ensure that the discussion remained focused upon topics of relevance to the research (Kitchin and Tate, 2000). I also had to pay careful attention to managing the discussion to achieve depth of insight, particularly through the use of prompting participants in a manner that encouraged elaboration and discussion without being leading or pressuring (Krueger, 1998; Berg, 2009). Finally, semi-structured interviews were carried out by a team of local assistants in the CBFM research village as a means to access further opinions in the village, particularly regarding sensitive issues (see 4.3.7). I employed six local villagers who each completed assignments to talk to local villagers and write reports on a specific issue related to my research. I employed this method to attempt to provide an extra level of depth to the data that I was collecting personally. The reports that were generated provided a medium for local villagers to contribute to the project in a totally anonymous fashion and were especially valuable for providing a means of communication about topics that were sensitive, particularly corruption mechanisms within the committee structure of the projects and villagers’ complaints about the project leaders themselves. 4.3.5 Participatory Activities In conjunction with the focus groups, several participatory activities were carried out by the respondents. A list of the participatory activities carried out is shown in Table 4.4. 109 These activities were implemented not for emancipatory objectives (Cooke and Kothari, 2001; Hickey and Mohan, 2004a), but as a tool to facilitate discussion and group dynamics, and to provide a meaningful way for people to articulate knowledge. In the case of semi- structured interviews carried out with the MBOMIPA Representatives in Nyamahana for example, a village map was created during the interview as a tool to facilitate discussion. This interview had started off very hesitantly, and the map drawing exercise was carried out as a means to help the participants relax and begin talking more freely with myself and my research assistant. The use of these techniques, for example the participatory wealth ranking in the communities, was also a valuable source of information about the village which helped to set other information in context, and as described earlier, was also a vital source of validation of the range of participants that had been included in the project. The participatory activities also assisted in the clarification of bureaucratic procedures through a diagrammatic representation of procedures and responsibilities. 4.3.6 Participant Observation Within the research villages I carried out participant observation as a tool to situate data gathered in focus group and interview settings (Agar and MacDonald, 2003), and as a stand- alone method for use during meetings and daily activities in the communities. The value of this data came from both the things that were said in these situations, but also in the greater Forestry Sector Wildlife Sector Village Natural Resources Committee annual calendar of duties Community-Based Forest Management financial system Village natural resources map Village governance structures Beekeeping System Charcoal System Tobacco system Scout and Patrol System Wealth Ranking Chart Village natural resources map (Makifu and Nyamahana) Farming calendar Scout and patrol system Wealth ranking Table 4.4: Participatory Activities Carried out During Fieldwork. All activities were carried out within the villages selected for in-depth research, unless stated otherwise 110 insight it provided into the power relations within the communities, social norms and groups within the communities. During the fieldwork periods I participated in 3 village meetings (one in Kiwere and two in Makifu), which are quarterly meetings at which all issues of importance to the village are discussed. I also attended one meeting of the zonal committee for Participatory Forest Management, which is held to report on the activities (financial and protection) of all Participatory Forest Management projects in the zone (consisting of five villages, including Kiwere; see Fig. 4.3) and provides a forum for advice and discussion between the leaders of the project and district authority representatives. 4.3.7 Local Assistant Reports Within the CBFM research village, in addition to working with a full-time research assistant, I employed ten local assistants to carry out further discussion-based research and to write reports on specific topics. I employed this technique because it became clear that within the small village community, discussing sensitive issues and creating space for people to discuss their complaints about the forest and its management by the VNRC was a difficult task. The local assistants were each given a topic to research by talking to people within the community in an informal way, and were asked to create a written report within a week. This technique proved especially useful for providing a means for people to express their opinions on such sensitive topics in an anonymous way, as the local assistants were instructed not to record the identities of the individuals they talked with. This technique proved very successful in gathering further information about sensitive issues and contextual information about the research village. The local assistants lived across the sub-villages within the research village and were all young and male and had completed some degree of secondary schooling. It was not possible to recruit female local assistants due to the lack of secondary education completed by most young girls within the village and the domestic responsibilities placed upon them. Despite the gender issue, the positionality of the local assistants within the village was largely beneficial as they were able to speak to many different social groups within the community and held no positions of authority or close personal connections to those involved in the management of the VLFR. 4.4 Triangulation Multiple methods were employed to investigate each of the areas of study and research themes. Triangulation was not used principally for information verification and replication, 111 as I chose to rely upon reaching data saturation within each of the methods. Triangulation was valuable, however in providing multiple ‘lines of sight’ upon the research questions (Campbell, 2001; Berg, 2009). These ‘lines of sight’ are combined to provide the researcher with “a better, more substantive picture of reality; a richer, more complete array of symbols and theoretical concepts; and a means of verifying many of these elements” (Berg, 2009, p.5). Triangulation in this sense provides a “means of refining, broadening and strengthening conceptual linkages” (Berg, 2009: 7). As discussed above, therefore, multiple methods were used to address each of the research questions. 4.5 Language Although English is an official language of the United Republic of Tanzania, it was necessary to conduct research in Kiswahili in all cases except for when dealing with international conservation practitioners and researchers and a small number of national level respondents. I carried out three months of intensive language training in total for the purpose of fieldwork. I passed the final examination at the level of ‘advanced high’. I carried out my language studies at the Riverside Language and Cross-cultural School, based just outside Iringa town. This location also allowed me to begin project setup and to initiate contact with the regional and district authorities and several conservation and development Non-Governmental Organisations (NGOs), many of which became important contacts within the CBNRM community and research participants. Whilst Tanzania is an ethnically diverse country, Kiswahili is a universally spoken language that unites different ethnic groups. The villages in which I was carrying out in depth research included residents that spoke several different languages, including Kihehe, Kibena, Kigogo and Maa amongst many others. Choosing to learn Kiswahili enabled me to conduct research with all of the different ethnic groups present within the villages without the need for a translator. Learning to speak Kiswahili was a vital skill that facilitated my research in several ways; being able to carry out all of my interviews in one language not only reduced the time taken to cover issues (and thus facilitating wider coverage of issues in interviews), but also reduced the level of artificiality that would have been associated with translation through an interpreter (Frey and Fontana, 1991; Evans et al., 1994). During the course of my research it became clear that the use if Kiswahili was beneficial to recruitment of participants, as they expressed relief at being able to conduct the interview in one language and not having to be concerned with how my research assistants would translate their responses. To avoid 112 misunderstandings I was accompanied by a research assistant to all pre-arranged interviews and focus groups. In response to the concerns of participants over translation, I chose to also use Kiswahili for cases where I required clarification of meaning during interviews. 4.5 Recording and Note-taking Strategies A field diary was kept throughout the fieldwork periods. This was used to keep a written record of all research carried out and topics or themes that arose from these. The diary also served as a tool for reflexivity by providing a medium to express thoughts concerning motivations, decision-making and positionality (Dwyer and Limb, 2001). All interviews at the international, national, regional, district and village authority levels were recorded with the permission of the participants using an Olympus WS-331M Digital Voice Recorder and noise cancelling microphone (when required). Interviews at the village level (non village authority) were not recorded, as it was clear that respondents found this intimidating and were suspicious of what the machine did and how the data would be used. However, I did record focus group discussions that took place in the villages in order to facilitate revision of the discussion. During all interviews notes were taken by myself and my research assistant and were written up in the fieldwork diaries within 24 hours. Discussing the day’s work and writing up notes from the activities carried out each day was a vital tool for reflexivity and also allowed freedom for myself and my research assistants to take minimal notes during discussions. 4.5 Transcription and Analysis The data outputs entailed notes taken during interviews and focus groups, participant observation, participatory activities, local assistant reports and document sources. The recordings of interviews and focus groups were transcribed in cases where this was deemed necessary for further understanding, or where the interview or focus group proved particularly rich in information or discourse (see Appendix 1). This resulted in 22 cases of interviews and focus groups being transcribed in full, and partial transcripts being taken for nearly all of the interviews and focus groups. Transcription was carried out by my research assistant and the data was transcribed into Kiswahili, not translated into English. These sources were put into the textual analysis software Atlas.ti for analysis. Analysis involved the application of thematic codes to the text data. These codes were used to draw out instances where participants discussed themes of previously identified importance and to 113 attribute meaning and theories (Crang, 1997). Secondly the coding process enabled identification of emic codes, used by the informants themselves, to highlight participants’ understandings (Crang, 1997). Coding enabled organisation of the data, the drawing out of important themes and the process of coding proved highly useful to beginning to analyse the data and order thoughts and theories that emerged from them. The codes created in Atlas.ti were used to create a conceptual hierarchy and network of codes to assist with analysis. Coding also assisted with comparisons between the interviews within each sector and across the wildlife and forestry sectors. 4.6 Constraints and Reflections 4.6.1 Positionality “The version somebody presents in an interview does not necessarily correspond to the version he or she would have formulated at the moment when the reported event happened. It does not necessarily correspond to the version he or she would have given to a different researcher with a different research question. The researcher, who interprets this interview and presents it as part of his or her findings, produces a new version of the whole. Different readers of the book, article or report interpret the researcher’s version differently, so that further versions of the event emerge.” Flick (2002: 9) Since the mid 1980s, when qualitative research endured a ‘crisis of representation’, the fallibility of these kinds of data has been thoroughly debated. It has become clear that in the context of interviews and focus groups, Haberrmas’s concept of an ‘ideal speech situation’ is deeply flawed (Puri and Sahay, 2003). Qualitative methodologies involve social encounters, or dramaturgical performance (Berg, 2009) that takes place within the influences if myriad social processes and relationships (Mishler, 1986). The information presented in interviews should always be recognised as a performance of discourse, power relations, social norms, the expectations of the informant, and their understandings of the researcher’s intentions (Briggs, 1986). This was particularly apparent in the focus group discussions and the modifications that had to be made to these (see 4.3.3). Considering these limitations, it is recognised that the value-neutrality assertions of research are false, research “cannot but be political” (Hammersley, 1995: viii) and through a post-structuralist lens, it must be accepted that the knowledge presented in data collection methods is situated. The situation of knowledge occurs in terms of the context of the discussion, whereby both interviewer and 114 participant arrive with desires and agendas and unique understandings of the matter to be discussed, and the discussion itself (Kitchin and Tate, 2000). Secondly, knowledge is situated in cultural ideologies, or “general, unwritten laws concerning what is permissible within society” (Kitchin and Tate, 2000: 24). “Play fool to catch wise” Jamaican Slave Proverb (Scott: 3) The knowledge presented in the data collection methods employed in this study is situated according to the positionality of myself as a researcher also. Positionality refers to the social roles assigned within research. These roles are an amalgamation of the researcher’s cultural background, socio-economic status, race, gender and age, and positionalities affect both the researcher’s approach to the research problem, and also drive the roles and positions assigned to the researcher by the people being researched, and in turn the relationships that can be built between researcher and participant (Howard, 1994; Madge, 1994). This placed constraints on me in relation to my gender in a patriarchal culture, nationality in a country previously under colonial rule and status as an outsider and non-native speaker of the language (Skelton, 2001; Mercer et al., 2003), However, I discovered that these roles could be useful in terms of my research. Robson & Willis (1994) argue that these kinds of positionalities can be manoeuvred to create opportunities for participants to feel as if they are educating the researcher. This I found to be very true, and beneficial in several ways; accepting my status as an outsider allowed me to project an image as a student in search of knowledge, an un- threatening persona, and opened up opportunities to address sensitive issues, such as corruption, in a less-threatening way. I also accept that my positionality as a foreigner, particularly as a researcher from the University of Cambridge, but also through my links to the Sokoine University of Agriculture, were key factors in gaining access to documents and being granted permission to carry out interviews. 4.6.2 Sensitive Issues In addressing sensitive issues, the methodological approach was cautious and iterative. I found that the most effective way to discuss issues such as corruption mechanisms within the projects was to open opportunities for the participant(s) to contribute to these issues, but not press participants to take these opportunities. In short, I tried to construct a space for people to reveal ‘hidden transcripts’, or “critique[s] of power spoken behind the back of the dominant”(Scott, 1990: xii). I had to be sensitive to other people involved in the discussion, 115 who may have been inhibited in making disclosures, and to remain perceptive of people who might wish to discuss issues privately. However, I was also acutely aware that the motivations of different actors within the projects made separating truth from rumour and gossip extremely complicated. Here I had to rely on my background knowledge of the projects and its governance systems to assess whether certain accusations were feasible, and through triangulation I was able to increasingly judge the validity of many, but not all, claims. Instead of doggedly seeking the truth in the performance given to me by research participants, I adopted an approach of seeing rumour, jokes and gestures as an expression of the hidden transcript, “albeit in disguised form” (Scott, 1990: xiii). 4.6.3 Ethics Research in the past has been heavily criticised for its extractive, exploitative nature on participants (Chambers, 1997; Kitchin and Tate, 2000). I attempted to address this issue by building in aspects of reciprocity into my research design. I compensated participants for their time, using food rather than cash, whenever possible, and made a conscious effort to assist in the community, particularly through the service of my car to help people transporting goods to market, people to the clinic or elderly people to the next village, for example (Robson, 1994). I thanked people who had been particularly helpful in the logistics and management of my research using culturally appropriate gifts, such as traditional gifts at particular national holidays and religious celebrations. I always ensured that participants were informed of my research and consented to participate. Due to the invitational recruitment technique used for the focus groups, these specifically applied an ‘opt in’ basis. I obtained permission to record the interviews where the dictaphone was used and where permission was not granted (in two instances), I did not use the dictaphone and relied upon the notes that were taken during the interview. Confidentiality and anonymity were assured in the project, data were stored securely and primary data were never shared. For the purposes of anonymity, the names of research participants, or instances where they referred to individuals in the projects have been removed whenever this is not critical to the narrative or to provide context to the statements made. 116 4.6.4 Reflexivity Embracing the political nature of my research, I found it important to be reflective during the research process. This is defined as “self critical sympathetic introspection and the self conscious analytical scrutiny of self as researcher” (England, 1994: 82). In accordance with Flick (2002), I found that reflection helped to highlight some of the intricacies of the research project, and became a significant analytical and interpretive tool and data in their own right. To help reflect on the research, the participants and the findings, I used field diaries and discussion with my research assistant. Reflection was particularly useful in addressing issues of cross-cultural research and attempting to situate findings within a cultural context. I was able to discuss with my research assistants instances that puzzled me from the discussions that we had participated in. The field diaries were also useful for keeping a record of the discussions and I was able to return to these notes at a later date and to add extra insight into the discussions from the things that I had learnt and experienced over the course of the fieldwork periods. I found reflection also to be invaluable in considering the themes that I was drawing out of my research findings and how these were shaping my further research and my understandings within further data analysis. I found it important to keep a record of these developments within my thinking and to discuss the general themes that I was drawing from the research with my research assistants, who would sometimes have different ideas and contradictory beliefs regarding the research findings. I found these discussions incredibly useful in refining my research strategy, identifying my own beliefs, pre-conceptions and misunderstandings within the research. I used the outcomes of my own reflection and the discussions with my research assistants to continually assess the trajectory of the research being undertaken and to refine the discussions that were taking place and my own understandings of these. Chapter 5: Policy Processes in Tanzanian Community-Based Natural Resource Management 117 5.1 Introduction In this chapter I investigate the policy context in which Wildlife Management Areas (WMAs) and Community-Based Forest Management (CBFM), as outlined in chapter 3, are currently being implemented. I aim to explore the evolution of these policies, and identify the factors that played key roles in the emergence of the Community-Based Natural Resource Management (CBNRM) agenda in Tanzania as a whole, and those shaping the distinct policy pathways that were adopted in the forestry and wildlife sectors. I will attempt to deconstruct the processes within policy reform and to highlight the role that pilot projects in Tanzania, both prior to and during national policy development, played in this. Following the framework set out in 2.2.3, I focus on policy reform, conceptualising this as the amalgamation of agenda setting, policy development processes and policy implementation (see also 2.2.2). Each of these represent complex interrelated aspects of structured interests, actors’ agency, and power-knowledge relationships. In the first section of the chapter I outline the key features of the institutional governance systems prescribed in policy for both WMAs and CBFM. I utilise these structures and the application procedures to create a WMA and Village Land Forest Reserve (VLFR), as set out in policy, to draw attention to the differences between these two approaches to CBNRM in Tanzania. Next I address the discursive shift to CBNRM in Tanzania and the role of external influences upon this. In section 5.4 I draw upon ideas of a political economy of natural resources to add depth of understanding to the policy pathways chosen in the forestry and wildlife sectors, and explain the differences these represent. In this section I also address the discursive and political struggles that took place within the process of policy reform, and the ways in which this relates to the structured political-economic interests of actors driving policy options. In section 5.5, I adopt a network view of policy processes to consider the policy communities and discourse coalitions involved in the development of CBNRM policy, with particular attention to the role that donor-funded pilot projects played in these processes. This chapter draws both upon previous research carried out, and upon primary research findings from my own fieldwork. The development and emergence of CBNRM at the policy level, and the shift from protectionist, fortress conservation strategies to the inclusion of local communities in Tanzania has become a popular topic for research (Songorwa et al., 2000; Brockington, 2002; Songorwa, 2004; Nelson, 2007; Nelson et al., 2007; Nelson and Blomley, 2010), and I draw upon this background to add extra depth to the area of study and address 118 questions that have been raised in previous research. From my own research into these issues, I draw on policy documents, alongside project documents for both the Ruaha Ecosystem Wildlife Management (REWMP) and later MBOMIPA projects and the MEMA forestry project (see 4.2.1.1 & 4.2.1.2 and Appendix 3) and research findings from the semi- structured interviews carried out. 5.2 Prescribed Governance Systems In Tanzania, policy reform in the forestry and wildlife sectors has produced a discourse of CBNRM, but is implemented in policy through WMAs and CBFM projects in two starkly different governance systems. As described in 3.4, the objectives of WMA and CBFM projects are essentially the same, and follow a discourse of CBNRM to achieve sustainable management of natural resources through the tying together of conservation and development goals implemented in the devolution of power to the local level. This highly sectoral nature of CBNRM policy in Tanzania is identified by Blomley & Iddi (2009: 16) as a major weakness with the policies resulting in “divergent ideas about how to devolve management to the village level”. In this section I utilise an analysis of the policy documents underlying both WMAs and CBFM, alongside records held by the Village Natural Resources Committee (VNRC) of Kiwere and Authorised Association of MBOMIPA to compare the institutional structures prescribed in policy, the revenue systems within both WMAs and CBFM and the application procedures involved in gazetting a WMA and a VLFR. 5.2.1 Community-Based Forest Management Institutions As a result of the development of CBFM, the role of the government in natural resource management is being both restricted and redefined, resulting in an approach that was described by MEMA as “eyes on, hands off” (MEMA, 2001: 3). The Forestry and Beekeeping Division of the Tanzanian Ministry of Natural Resources and Tourism does not implement CBFM directly, but its role consists of policy guidance, quality control and to offer guidance, capacity-building and training (Blomley, 2006). The principal institution in CBFM has become the Village Natural Resources Committee (VNRC), which acts as the institution responsible for day to day management of the VLFR. Table 5.1 outlines the main institutions involved in CBFM and the roles set out for them according to policy. 119 Institution Operating Level Principal Responsibilities Ministry of Natural Resources & Tourism (Forestry and Beekeeping Division) National Sanctioning of land use plans and officially gazette the VLFR Policy guidance Advice Capacity-building and training provision District Council District Collect taxes from the VLFR (5%) Facilitate and approve the land use plan for the VLFR Declaration of a VLFR alongside the Village Council Monitor the VNRC bank account Village Council Village To prepare the management plan for the forest land as part of the gazettement process Control revenue collection and spending of the VNRC To decide with the VNRC on expenses for the committee Authority to apprehend anyone in contravention of the management plan Village Natural Resources Committee Village/ inter- village Implement forest management plan for the VLFR To hold monthly meetings and prepare monthly reports on revenue, patrols and resource monitoring Carry out monthly perception interviews in the community To prepare quarterly accounts and reports to be read aloud at village assembly meetings Enforce rules of use and village bye-laws Resource monitoring Conflict resolution Enforcement of sanctions for illegal activities Collection of revenue from the VLFR NGOs/ development organisations All Facilitate application and gazettement of the VLFR Private Sector All Enter into investment agreements with the VNRC (through purchase of permits) Assist in protection of natural resources Adhere to the terms and conditions of the VLFR land use plan and investment agreement Table 5.1: Principal Roles and Responsibilities of Institutions within Community-Based Forest Management Policy (adapted from Boiesen and Lund, 2003; MEMA, 2003; Ministry of Natural Resources and Tourism, 2009). The roles set out in this table were also particularly informed by Semi Structured Interviews P1, P9, P11, P12, P28 and Focus Group P196. 120 The institution charged with the chief management role, the VNRC, is a unique component of the village government system in Tanzania, as it is the only sub-committee of the Village Council that cannot consist of Village Council members, but must be elected through the village assembly (Boiesen and Lund, 2003). Its members are elected every five years (Lund, 2007). Within the CBFM system, the VNRC is the executive authority, with power to arrest defenders and collect revenue (Boiesen and Lund, 2003). It is important to note that in VLFRs the VNRC, through the creation of a management plan, is able to set its own permission system, prices, quotas and fine levels and sanction systems (Koppers et al., 2004). No other body is able to issue permits for use within the VLFR (Koppers et al., 2004). The VNRC constitutes a standard committee structure, which must represent all the sub-villages within the village and consider gender representation (Boiesen and Lund, 2003; MEMA, 2003). The Chairman distributes all reports to the district offices, Village Council and assembly, prepares all the VNRC meetings and chairs them. The Secretary is responsible for keeping records of all VNRC meetings and other meetings that have discussed the VNRC and for filling in all monthly forms and keeping copies of them. The Treasurer records all permits, receipts and payment vouchers, collects all revenue and prepares the financial report monthly for the VNRC meetings. There is also an Interview Chairman, who is responsible for carrying out interviews with the local community to record uses of the forest, perceptions of the state of natural resources and the governance of the project. There are 5-7 ordinary members who assist in decision-making processes and monitor the performance of the VNRC and its members. The Patrol Commander ensures that patrols are carried out and that patrol forms are delivered to the Secretary and finally, 3 patrol guards carry out the actual patrols and fill in the patrol forms (Boiesen and Lund, 2003). According to Boiesen & Lund (2003) this institutional system leaves the top members of the committee with responsibilities of linking with other organisations and for the collection and distribution of information and material. They also argue that the patrol guards hold a special position on the VNRC, as they are not full members, having been “consciously removed from the VNRC, as it is believed that their role as ‘policemen’ will inflict on the possibility of the other VNRC members to build more friendly relations with the villagers” (Boiesen and Lund, 2003: 79). The VNRC is responsible for monitoring the resources within the VLFR, which it does through three methods; revenue accounts, perception interviews and monitoring the patrolling of the forest (Ministry of Natural Resources and Tourism, 2009). The perception interviews carried out must number at least five per month from people in different sub-villages, and should ask 121 interviewees to describe changes and developments in the forest resource, and to what they attribute these developments (Boiesen and Lund, 2003). The VNRC must produce a monthly report, which includes a summary of the committee’s activities, including meetings, general assemblies, workshops, a summary of permits, fines, revenue and expenditure, a summary of patrol observations, and a summary of perception interviews carried out (MEMA, 2003). This report is kept by the VNRC and copies are sent each month to the District Natural Resources Office, and to the Ward Executive Officer (MEMA, 2003). 5.2.2 CBFM Revenue Systems VLFR harvesting is conducted according to the management plan created by the village government, and approved by the district authorities (Government of the United Republic of Tanzania, 2002b; Ministry of Natural Resources and Tourism, 2009). In the research villages, interviews with the committee members and villagers confirmed that harvesting centred around timber, firewood for charcoal and tobacco processing, building poles and stones, livestock grazing and fodder collection and medicinal plant collection (Interviews P34, P35, P38, P47-P56). Revenue from these activities is collected through payments for permits purchased by investors from the VNRC (Ministry of Natural Resources and Tourism, 2009). According to the Forest Act Regulations (Part XII) “timber harvested, sold and transported from village forest reserves shall be marked with a “registered mark” (hammer) which shall be issued by Local Authority for that respective village, and registered and gazetted by the Director of Forestry”. Since the political reforms of 2006 (see 3.2.2), financing and reporting lines for CBFM have altered; previously financing from development partners went from the Ministry of Natural Resources and Tourism (MNRT) to district accountants, and District Forest Officers’ reported activities and expenditures, based on approved budgets, back to the MNRT (Blomley, 2006). In the current system funding is channelled directly from the Ministry of Finance to the district level, without passing through the MNRT, and routine reporting is to the District Council, which is collected by the Prime Minister’s Office Regional Administration and Local Government, which compiles financial reports and activity summaries across district for the Forestry and Beekeeping Division (Blomley, 2006). District governments still report to the Forestry and Beekeeping Division concerning impact and output monitoring, as its staff retain responsibility for monitoring policy implementation. 122 Under CBFM, communities are exempt from regulation governing the harvesting of ‘reserved tree’ species, and are not obliged to share royalties with the district or national government (Blomley, 2006). The villages do pay a 5% cess (tax) to the District Council, however (Boiesen and Lund, 2003). Cess is a tax which the District Council retains for discretionary spending, as it does not have to be shared with central government (Koppers et al., 2004). Having paid their fee to the VNRC to receive a permit, producers or wholesalers are then exempt from paying the regular district cess (Boiesen and Lund, 2003). Revenue collected by the VNRC must be used to finance expenditures related to forest management, and all remaining funds must be used for public services (Lund, 2007). 5.2.3 Wildlife Management Area Institutions In WMAs the prescribed governance structure places management responsibility at the level of the Authorised Association, which represents the villages participating in the WMA. The institutions involved in WMAs are outlined and discussed with reference to their roles in Table 5.2. 123 Institution Operating Level Principal Responsibilities Ministry of Natural Resources & Tourism (Wildlife Division) National Authorise Community-Based Organisations to become Authorised Associations and grant user rights Declare WMA officially Sanction investment agreements Allocate revenue to lower operating levels Allocate hunting block and set hunting quota Determine memorandum of understanding on relationship between the Authorised Association and Wildlife Division concerning management of the WMA Tanzanian National Parks Authority National Act as the agent of the Wildlife Division in the WMA Participate in the District Natural Resources Advisory Board Assist AAs with anti-poaching activities and assist in anti-poaching and problem animal control District Council District Facilitate application procedure for gazettement of the WMA Form link between the AA and the Wildlife Division Approve land use plans for the WMA Approve village bye-laws Monitor all potential and existing investment agreements between private sector and the AA Issue resident hunting licenses to the AA District Game Officer District Responsible for managing problem animal control Organises patrols and activities of the District Game Scouts District Natural Resources Advisory Board District Arbitration and resolution of conflicts Provide technical and legal advice to the AA Co-ordinate with the Wildlife Division to set the hunting quota Verify AA investment agreements and advise Wildlife Division as to their suitability Board of Trustees District Approves annual plans, budgets and financial statements of the AA Advisory role concerning all aspects of the WMA and its management Authorised Association Inter-village Complete requirements for official gazettement of the WMA Implementation of the land use plan (including conflict resolution, sanction enforcement, problem animal control, resource protection) Resource monitoring Protect biodiversity and wildlife resources in the WMA Report and seek authorisation for investment agreements Recruit, train and oversee Game Scouts Carry out financial management of the AA, including paying fees to relevant authorities Issue permits for resource use within the WMA Report to village assembly Monitor wildlife resources in the WMA Village Council Village Co-ordination of natural resource management activities at the village level Preparation of the land use plan Creation of village bye-laws for approval by the District Council Monitor the AA and report to the District Council Enter into agreement with AA on management of the WMA Table 5.2: Principal Roles and Responsibilities of Institutions within Wildlife Management Areas (adapted from Walsh, 1998; MNRT, 2000b; Zacharia and Kaihula, 2001; MEMA, 2003). The roles set out in this table were also particularly informed by semi structured interviews P73, P74, P77, P87, P88, P93, P97, P99, P100. 124 Village Natural Resources Committee Village Strengthen relationship between the village and village government, surrounding villages, Ruaha National Park and the Natural Resources and Community Development institutions at the ward and district levels Prepare Natural Resource Use Plans and ensure these become village plans To oversee the implementation of natural resource use plans Advise village government on the formulation of village bye-laws for natural resource and environmental protection Oversee implementation of bye-laws Co-ordinate Village Game Scouts’ patrols and reporting systems Receive confiscated exhibits/trophies and hand them to the District Officers Monitor and evaluate natural resource use by identifying and reporting issues with VNRC and Village Game Scouts To keep records of all activities related to the environment To develop and oversee Income Generating Projects for honey, fishing and forest products Provide environmental education through public meetings, community groups and primary schools Prepare and submit monthly income and expenditure reports for natural resource use and submit them to the Village Council, District Natural Resources Office and MBOMIPA offices Ensure budget for natural resource use is properly implemented Village Assembly Village Approve formation of the WMA Elect village representatives for the AA Approve village land-use plan and bye-laws Approve uses of village revenue from WMA Authorise AA to enter into investment agreements Non- Governmental Organisations/ Development Organisations All Facilitate application and gazettement of the WMA Provide technical advice to the AA Participate in the District Natural Resources Advisory Board upon request Private Sector All Enter into investment agreements within the WMA Participate in District Natural Resources Advisory Body upon request Assist in protection of natural resources Adhere to the terms and conditions of the WMA land use plan and investment agreement 125 According to the constitution of MBOMIPA (2002), the institutional structure of the Authorised Association (AA), the principal managing institution, consists of the Secretariat, Executive Committee, Sub-committees of Finance, Security and Discipline. The Secretariat is formed by the two elected members from each participating village, and from this the Chairperson, General Secretary and Treasurer, their three deputies or assistants and the members of the further committees are elected. The sub-committees of discipline and planning have five members each, whilst the security sub-committee is made up of six members (MBOMIPA, 2002). The chair and secretary of each sub-committee become members of the executive committee, alongside the AA’s chair, secretary and treasurer and their three assistants (MBOMIPA, 2002). The executive committee meets monthly, and is responsible for preparing the expenditure and income reports, annual programme of activities and the AA budget (MBOMIPA, 2002). Committee members hold their position for five years, and may re-apply for their post (MBOMIPA, 2002). The chairperson of the AA is responsible for overseeing all activities, chairs all AA meetings, and is the designated spokesperson for the AA (MBOMIPA, 2002). The Secretary General of the AA is responsible for all AA documentation, is Chief Executive and coordinates all activities, programmes and plans, also reports to the village members of the organisation, and handles all applications for membership and reports them to the Executive Committee (MBOMIPA, 2002). The Treasurer of the AA is the chief financial and accounting officer, who is responsible for all financial and accounting records, proposes the budget to the Executive Committee, is a signatory on the bank account of the AA, cheques and bills/promissory notes, and may sign agreements on behalf of the AA if the Chairman and Secretary are not present (MBOMIPA, 2002). The role of the Disciplinary Committee is to oversee all members of the organisation (MBOMIPA, 2002). This is the institution responsible for ensuring accountability, transparency and identifying wrong-doing. The Financial Committee’s main responsibility is to discuss the financial programmes and to make suggestions at the secretarial conference (MBOMIPA, 2002). The Security Committee’s role appears vague in the constitution, described as carrying out all security duties and recruiting security guards (MBOMIPA, 2002). The AA holds at least three meetings per year. These are called by the Secretariat and are attended by its members and any invited guests (MBOMIPA, 2002). In practice this is 126 usually at least the District Game Officer, and often the Board of Trustees also. A special meeting can be called at any time, however (MBOMIPA, 2002). The conference of the secretariat provides the opportunity for proposed changes to the constitution, provided 21 days notice of the changes has been given, and the passing vote is made up of at least three quarters of the secretariat members (MBOMIPA, 2002). The meetings also form the site of dissolution of the organisation, which can take place if two thirds of the secretariat agree, and causes all the organisation’s assets to be transferred to a humanitarian organisation with similar values to the AA (MBOMIPA, 2002). De-gazettement of a village within the WMA can take place only through application to the Director of Wildlife, which must include the certification of authorization given to the Community-Based Organisation (prior to AA status), a letter of approval from the District Council, the minutes of the village assembly showing approval of removal from the WMA (Ministry of Natural Resources and Tourism, 2000a). Disbanding of the WMA entirely can take place with two thirds agreement by the participating villages (MBOMIPA, 2002). The role of the district remains important in the WMA governance system. The District Council must endorse all investment agreements (Booth et al., 2002) which is designed to assist AAs, who are seen as not having the negotiation skills to ensure that agreements are fair, but Booth et al. (2002: 45) argue that it also “erodes the concept of freedom of contract of the AA”. They also argue that it creates decreased transparency and accountability and provides the District Council with supervisory powers with no corresponding mechanism to check this power (Booth et al., 2002). At the district level the District Natural Resources Advisory Board coordinates the administration of the WMA, and is made up of up to 12 members, including the District Commissioner (as Chairman), District Executive Director, District Game Officer (as Secretary), District Land Officer, District Forestry Officer, District Community Development Officer, District Fisheries Officer, District Planning Officer, a representative from the AA, and a representative of the Game Reserve or National Park can also be included (Interview P88, Government of the United Republic of Tanzania, 2005). Interviewees from the district authorities, MBOMIPA investors and Board of Trustees reported that the Board of Trustees is made up of eight locally respected people, usually businessmen, and often with some experience in the tourist sector and interest in conservation. Their role is principally to provide advice on any aspects of the WMA and its management, but also to review all financial plans and statements for the WMA (Interviews P88, P102, P105). 127 Since the publication of the legislation underpinning WMAs in Tanzania, a new institution has been created at the national level; the National Consortium of Authorised Associations is made up of representatives from across the country’s WMAs and is designed to represent a coherent voice to represent the WMAs to the central government (Interview P100, Interview P80). MBOMIPA AA leaders described the consortium as being first suggested by MBOMIPA, and created through their actions with support from their Non-Governmental Organisation (NGO) facilitator, the Worldwide Fund for Nature (WWF; Interview P100). Two members of the MBOMIPA AA also held roles within the national consortium committee structure. 5.2.4 WMA Investment Agreements and Revenue Arrangements The Wildlife Policy for Tanzania (WPT) 1998 outlines that WMAs are eligible to gain income through entering into investment agreements with private investors on WMA land (Booth et al., 2002). These agreements can be in several forms, including a single fee for use of the area or a percentage of gross annual income (see Booth et al., 2002). All investment agreements are limited to three-year terms, and must be endorsed by the District Council and the Wildlife Division (Government of the United Republic of Tanzania, 2005). The role of the Village Assembly in WMA investment agreements remains unclear. In the 2005 Regulations for WMAs (Government of the United Republic of Tanzania, 2005), the role of the AA in statement 22 is described as seeking authorisation for investments from the Village Assembly and reporting investment activities to them. However, in statement 63, regarding investment agreements, the regulations make no mention of the Village Assembly (Government of the United Republic of Tanzania, 2005). Research in the case study WMA indicated that the village level played no role in approving investment contracts, and even members of the Village Council were unaware of the number of investment contracts that existed, or their nature. Revenue distribution surrounding WMAs has been a topic of debate for several years (see Nelson, 2007), both concerning the distribution between the Wildlife Division and AAs and within the AA itself. The financial system operating in WMAs involves the payment for hunting block concessions to the Wildlife Division, which then provides a proportion of this revenue to the AA (25%). The WMA regulations (Government of the United Republic of Tanzania, 2005 Statement 73) state that “(1) benefit sharing in a WMA shall comply with circulars issued by the government from time to time and shall adhere to mechanisms of 128 equitable distribution of costs and benefits targeted at economic development and poverty eradication. (2) AA shall ensure that from it’s (sic) annual gross revenue – (a) not less than 15% shall be reinvested for resource development. (b) not less than 50% shall be directed to member villages forming the WMA: and (c) not less than 25% shall be used to strengthen the AA”. Whilst the distribution of funds within the AA is clearly stated, the proportion of revenue from wildlife utilisation within WMAs with designated hunting blocks, which must be handed down from the Wildlife Division to the AA is not stipulated. As early as 2000, the MNRT published an amendment to the Wildlife Conservation (Tourist Hunting) Regulations which declared that all investment contracts within any wildlife Protected Area (PA; including hunting blocks and WMAs) require written approval by the Director of Wildlife (Ministry of Natural Resources and Tourism, 2000b). This change is significant because it declares previous investment agreements for non-consumptive uses of wildlife (e.g. photographic tourism) on village land inside a WMA to be illegal, without this ministerial sanction. The amendment has caused much controversy and some have declared that it has no legal basis (see Nshala, 2002; Institute of Resource Assessment, 2007). This policy change, along with alterations to the revenue sharing arrangements which see AAs keep just 65% of non-consumptive wildlife revenues, has been further sanctioned by the state in the publication of the Non-Consumptive Wildlife Utilisation Regulations in 2008 (Government of the United Republic of Tanzania, 2008; see also 3.4.2). Nelson et al., (2009) and Nshala (2002) argue that these amendments indicate the Wildlife Division’s desire to expand state control within the WMA framework, and represent a decrease in community autonomy. Several villages with pre-existing investment agreements now find themselves accused of illegal actions if they permit such investments and revenue generating activities to continue (Nelson et al., 2009). This is argued to form part of a wider trend towards recentralisation and state control over natural resources in the wildlife sector that has been strengthened between 1998 and the recent policy updates. This argument is supported by the updated Wildlife Conservation Act (2009), but also in other sectors including marine and fishery resources (see Benjaminsen and Bryceson, 2012; see also 3.4.2 and 5.4.2). The process of recentralisation that has taken place in the wildlife sector between the publication of the WPT (1998) and its update in 2007 marks drastic changes for the local level in terms of power devolution and the authority of the AA to manage wildlife resources on village land. This is particularly clearly exemplified in the wildlife regulations that have expanded state authority to intervene in investment arrangements, land management decisions and re-capture 129 revenues from wildlife resources (both consumptive and not; see also 3.4.2). The restricted devolution of power witnessed in the WMA governance structure has undergone successive restrictions and increased state interference and the AA now holds no authority to determine investments or to collect revenues autonomously. 5.2.5 A Comparative View of the Governance Systems in Tanzanian CBNRM There are three major areas of disparity between the institutional governance systems (shown in Figs. 5.1 and 5.2) as prescribed in policy for WMAs and CBFM projects. These differences were reported by a District Official as representing different cost-benefit systems and intrinsically linked to the politicisation of these resources and their values: “Wildlife are hotcakes; the money from tourist hunting activities is very large” (Interview P28). As discussed above, both WMAs and CBFM, as set out in policy, involve an administrative system that uses a management committee to oversee the daily management of the project. The first area of divergence in the prescribed governance systems concerns this administrative structure and the devolution of power within both sectors. Within the forestry sector management responsibility is invested in the existing village government system, with principal management responsibilities residing with the VNRC. Within the wildlife sector, management of the WMA is much more hierarchical, with an increased role for national, regional and district authorities, and local level responsibilities held mainly by the supra- village AA. The disparity in complexity between Tables 5.1 and 5.2 clearly indicate this. The AA is an institution that has not previously existed in Tanzanian law and has no pre-existing equivalent for natural resource management or local government structures. The effect of this new level in the governance structure of the wildlife sector is to devolve power to a level that sits above the village level, making village governments, and VNRCs particularly, upwardly accountable to the AA. In effect the management of the WMA is scaled up to the level of the AA and there is little downward accountability to the local level (see also 6.1.1). Village Council members in participating MBOMIPA villages described how they were invited to some MBOMIPA meetings and were able to give suggestions to the AA, but they expressed frustration that the AA should be accountable to them, but in reality they feel their job is to report on the activities of the AA to the village, and they have no control over the AA or the WMA (Interviews P129, P133, P89, P91). In contrast the retention of management within the VNRC for CBFM maintains accountability of its members down to the lowest level in the local government structure. Nelson & Blomley (2010) describe these differences as representing different levels and rapidity of power devolution within the 130 policies; CBFM has followed a more rapid route of devolution of power to the local level, with greater ownership rights vested in the local level than is produced by the WMA system. A comparison of the role of the district level between WMAs and CBFM indicates a much more intense role in the wildlife sector. Whilst the district is seen as a crucial level in the governance system for both CBFM and WMAs, through creating a link between the local and national levels (Zacharia and Kaihula, 2001), the institutional structures shown in Fig 5.2 shows that in the wildlife sector additional roles are created through the incorporation of a new institution: the District Natural Resources Advisory Board. The role of the regional level is also increased in the wildlife sector through the creation of a Board of Trustees for the WMA. 131 Fig. 5.1: Institutional Governance Structure Prescribed in Policy for CBFM in Tanzania Fig. 5.2: Institutional Governance Structure Prescribed in Policy for WMAs in Tanzania 132 Both policies set out a system of communities producing management plans for the land they wish to gazette as a WMA or VLFR, which are then approved by the authorities at the district and national levels, prior to official declaration. Once the application procedure is complete, the AA or VNRC are eligible to enter into investment agreements with private sector companies to gain income from the sale of their rights to use the resources within the WMA or VLFR. These procedures set out the two further areas of disagreement between the forestry and wildlife sectors’ policies, however. Firstly, the application procedure to gazette a WMA is comparatively lengthy and complicated, which Nelson (2007) describes as motivated by the will to retain maximum district and national scale influence at all stages. The gazettement of a WMA follows the following steps (Nelson, 2007): 1. Village Assembly agrees to form a WMA based on the Village Council recommendation 2. Villages form a Community-Based Organisation (by election) and register it at the Ministry of Home Affairs 3. Community-Based Organisation prepares a strategic plan 4. Villages prepare a Land Use Plan 5. Environmental Impact Assessment of Land Use Plan 6. Villages create by-laws to support Land Use Plan 7. Community-Based Organisation prepares a Resource Management Zone Plan 8. Community-Based Organisation applies for user rights to become an Authorised Association 9. WMAs on Game Control Areas must be re-gazetted 10. Authorised Association applies to the Director of Wildlife for hunting block allocation 11. Authorised Association enters into investment agreements 12. Investments subject to environmental impact assessment According to a national review of Participatory Forest Management in Tanzania (Ministry of Natural Resources and Tourism, 2009), the formal procedure to gazette a VLFR is as follows50: 1. A management plan for the forest is created by the Village Council51. 2. Written intention to declare a VLFR to the district authorities 52 50 Prior to the creation of the management plan, the villages must establish that they have legal tenure over the land (classified as village land by the Ministry of Lands and Human Settlement), although management of the land can take place before this procedure has been completed and the village has obtained a certificate of village land. All internal and external boundaries of the forest to be included in the VLFR must also be demarcated. 51 The plan must describe how the forest will be used, managed and protected. If timber utilization is proposed, the plan must outline the areas of the forest where this will take place, and the amounts involved (in accordance with the Forest Act 2002 part III section 14). The plan must contain a map. 133 3. The management plan is presented to the village assembly for approval 4. The village elects a Forest Management Committee or VNRC 5. The village (Village Council and VNRC) prepares bye-laws that support the management plan 6. Bye-laws presented to the village assembly for approval 7. Ratification and declaration of the VLFR by the District Council53 8. Formal Gazettement by the Forestry and Beekeeping Division54 The key differences between these procedures is the complexity of the steps that must be completed to gazette a WMA; as part of this process permission must be granted from the ministerial level on four separate occasions (steps 8, 9, 10 and 12), which follow a lengthy period of complicated planning requirements for both the WMA as a whole and for individual villages participating in the WMA (a strategic plan, a land use plan, and a resource zone management plan). One academic researcher described the required process as “more complicated than what has to be done to gazette a National Park, and yet it is communities that are expected to do it” (Interview P75). In contrast, the procedure for VLFRs involves just two applications: one to the district level to declare the VLFR; and one (optional) application to the Forestry and Beekeeping Division for formal gazettement. A VLFR also only requires one management plan to be created. A WMA does not formally exist until step 9 and may not collect revenue from investment contracts until step 12 is completed for each and every investment contract. However, a VLFR becomes operational at step 7, and revenue can begin even before the final gazettement by the Forestry and Beekeeping Division, which is seen as a formality after a period of time. In CBFM projects, the village must produce a management plan and by-laws to implement this plan, but no environmental impact assessments are required, and investments do not require sanction beyond the level of the VNRC, who ensure they are in accordance with the management plan. The processes also reveal key differences in the direction of accountability within these projects and the structures within which they operate: once a village assembly has endorsed the initiation of the application procedure for a WMA, there is no formal requirement for the management 52 The Chairman of the Village Council writes to the District Executive Director and informs him of the intention to declare a VLFR. The District Executive Director and District Forest Officer may comment on the application. The plan must also be forwarded to the Director of Forestry for comments and consideration 53 Copies of the bye-laws, management plan, minutes of meetings and members of the VNRC are sent to the District Council 54 After 3 years, if the village has managed the reserve in accordance with the management plan, they may request formal gazettement by the central government, which provides a formal certificate signed by the Director of Forestry. Where villages have failed to manage the forest in accordance with the management plan the Director of Forestry can remove the rights of villagers to manage their own land (and hand over management to the District Council) 134 plans to be ratified by this level, approval is always required from levels above the AA. In the CBFM application procedure, village assembly sanction of both the management plan and the by-laws are stipulated as a requirement. The direction of accountability in the WMA governance system is consistently upwards, with few opportunities for downward accountability, whereas a mixture of accountability mechanisms is apparent in the forestry sector. Finally, the revenue sharing systems within the two policies are contrasting. In WMAs, communities enter into investment agreements (with Wildlife Division and District Authority sanction) to gain income from non-consumptive uses of the WMA, for example photographic tourism55. Hunting activities within the WMA are controlled by the Wildlife Division, however, who allocate hunting blocks, sanction all hunting quotas, and share the revenues from these activities, although in an undisclosed and insecure manner, as no confirmation of revenue sharing proportions has been made in policy (see Government of the United Republic of Tanzania, 2005; see also 5.2.4). In contrast, within CBFM consumptive and non- consumptive uses of the forest can be undertaken by the community, or income gained through selling these rights to external companies (Ministry of Natural Resources and Tourism, 2001a). Management is overseen by the VNRC, as set out in the management plan, and all revenues are retained by the community, with 5% cess paid to the District (Ministry of Natural Resources and Tourism, 2001a). The summaries above attempt to highlight some of the important differences between the two policies, and begin to indicate the origins of these distinctions and divergences. To summarise, the differences in the development of the policies has resulted in twin policies of CBNRM within Tanzania, covering distinct resources but with almost identical objectives, and which have employed a different governance systems to achieve their aims. The wildlife sector provides an opportunity to study a CBNRM policy which maintains tight national- scale control over the power devolved to participating communities and retains power, influence and benefits for the regional and national scale. In contrast, the forestry sector provides an opportunity to investigate the rapid devolution of power to local communities, and a lesser degree of regional and national scale influence and responsibility. 55 Although the negotiation and signing of these agreements is overseen by officials from the district and Wildlife Division, and communities no longer collect the revenue from these investments themselves (see 3.4.2 & 5.2.4) 135 5.3 Community-Based Natural Resource Management Policy As discussed in 3.3, Tanzania has a very strong history of conservation based on Protected Areas (PAs), both in the wildlife sector where National Parks and Game Reserves formed the dominant strategy (Nelson et al., 2009), and similarly in the forestry sector which, until the 1980s was very much focused upon production, not sustainable use or conservation (Ylhäisi, 2003, Sunseri, 2005). The emergence of a discourse of CBNRM, and the policy reform this resulted in, is part of a complex arrangement of factors and influences at the national, global and local levels. In many ways, Tanzania exemplifies the discussion in section 1.1 concerning fortress conservation and the shifts towards bottom-up development and an integration with conservation objectives in the new paradigm of sustainable development (Chambers, 1997; Adams et al., 2004; Hutton et al., 2005; Blaikie and Springate-Baginski, 2006; Adams and Hutton, 2007). This shift is clear in the policy language used in the Wildlife Policy for Tanzania (Government of the United Republic of Tanzania, 1998a) and the National Forest Policy (Government of the United Republic of Tanzania, 1998b) and Forest Act (Government of the United Republic of Tanzania, 2002b), which highlight the role and rights of communities in conservation strategies: Policy statement 5 in the NFP states that “Village institutions will be granted appropriate user rights as incentives for sustainable forest management.” (1998: 19). Similarly the Forest Act states the objectives of the act are to facilitate sustainable development and encourage the active participation of citizens through the development of community rights, and at the heart of this strategy is the objective to “delegate responsibility for management of forest resources to the lowest possible level of local management consistent with the furtherance of national policies” (Part II, Statement 3, Government of the United Republic of Tanzania, 2002b). The rights of the local level within the introduction of CBFM is clearly stated in the Forest Act, which declares that “A Village Council may by resolution..declare an area of land to be a village land forest reserve” (Part V Statement 33:1, Government of the United Republic of Tanzania, 2002b) and goes on to declare that villagers hold the rights to manage a VLFR, including rights to enter, occupy, use and harvest forest produce in accordance with the management plan for the VLFR, by-laws, rules agreements and customary practices (Part V Statement 39:12, Government of the United Republic of Tanzania, 2002b). In the wildlife sector, the WPT (Government of the United Republic of Tanzania, 1998a, Statement 34 & Section 3.3.8: 19-20) also sets out the rights of local communities to “have full mandate of managing and benefitting from their conservation efforts” within the new WMA category of PA, and sets out a strategy for 136 facilitating CBNRM in the wildlife sector through “helping the rural communities to have secure ownership/long term use rights of their land and enabling them to use the wildlife and natural resources on that land”. The discourse of CBNRM, with its foci upon the devolution of rights to the local community and the integration of community participation and benefits from conservation activities are very clearly integrated into the policies that were developed in the 1990s, therefore. I discuss this integration below. 5.3.1 External Influences in Agenda Setting This discursive shift towards CBNRM in Tanzania is described by Nelson et al., (2009) as developing on the ground prior to national policy, through the role of donor-funded projects operating in the country. The reform of natural resource management policy in Tanzania is intrinsically linked to these influences through the agency of international organisations, particularly Transnational Conservation Organisations (TNCOs), the World Bank and International Monetary Fund, and the conditionalities placed upon bilateral aid. These influences have been paramount in previous discussions of the emergence of CBNRM in the country (see Hyden and Karlstrom, 1993; Igoe and Croucher, 2007; Nelson et al., 2007). Sunseri (2005) argues that this paradigm shift to CBNRM and the policy space surrounding TNCOs and bilateral funding agencies is tied into the emergence and eventual dominance of the biodiversity paradigm at the global level, the emergence of the hotspot discourse from 1992, and the important role of the World Conservation Strategy 1980, which played a large part in the development of the sustainable use discourse. Biodiversity narratives have also been heavily adopted in Tanzania, particularly by researchers at the University of Dar es Salaam, which led directly to the designation of the first National Park in Tanzania based upon biodiversity principles (Udzungwa National Park; Sunseri, 2005). Brockington (2006) argued that whilst discourses of bottom-up community conservation projects and environmentalisms are widespread within Tanzanian society, their origins within elite scientific discourses and ideas from Western NGOs is clear. A former ministerial staff member referred to the development of CBNRM in Tanzania’s wildlife sector as explicitly externally-driven, with TNCOs playing a key role in agenda-setting, alongside the hunting and poaching problems in Tanzania “which led the government to ask for help from the TNCOs...they agreed but on the condition that it would be community-based” (Interview P80). Fieldwork at the village level also revealed a lack of bottom-up drivers of policy, with community members describing how instead of communities demanding CBFM projects, they were brought to the villages by District Officers (Interview P197). 137 The removal of the Director of the Forestry and Beekeeping Division in 1996, under international and national pressure from donors, due to his lack of support for the development of CBNRM policies is a concrete example of the agency of TNCOs and bilateral development funding in the adoption of a CBNRM agenda in Tanzania (Nelson and Blomley, 2010). The policy space to achieve this influence was constructed through several channels. Firstly, the extent of such influence stretches beyond the shift to CBNRM in Tanzania and throughout the country’s history of conservation, as is exemplified in the large areas of land set-aside for conservation purposes since the colonial era (Brockington, 2006). Lovett (2003) argues that the influence of international conservation organisations can be traced back to the colonial era in Tanzania and the establishment of the first PAs in the country under colonial rule. Tanzania’s network of PAs in both the forestry and wildlife sectors was inherited by the newly independent government and retained and expanded over time, eventually forming the basis for CBNRM strategies, as discussed in chapter 3. The political commitment to conservation practices, often discussed with reference to the Arusha declaration of 1961 (in which President Julius Nyerere gave a speech confirming this commitment) was written on his behalf by Western TNCOs, providing a good example of this long history of international influence in Tanzanian policy (Neumann, 1998). Igoe & Croucher (2007) describe this agency of TNCOs as being constructed through the provision of resources, expertise and technology, which continues to be represented today in the role prescribed in policy for NGO facilitators within the governance system of WMAs (see 5.2 and 5.5). Brockington (2006) also argues that the authority of such TNCOs is derived from the importance of nature-based tourism to the country’s economy, which is used by such organisations to discursively frame the necessity for a shift towards CBNRM and is legitimised by the privileged status afforded to the knowledge they present. International support for CBNRM was presented primarily in narratives of the requisite, rights-based incorporation of local communities in conservation, for both moral and scientific reasons. CBNRM discourse drew upon both the morality of the inclusion of local communities in conservation activities, alongside rational, scientific justifications for such an approach based on efficiency (Andersson and Ostrom, 2004; Charnley and Poe, 2007; Brockington et al., 2008). The second of these narratives was clear in the rational, scientific, biodiversity-centric accounts of the need for policy reform towards CBNRM given by interview respondents across the national, district and village levels. For example, a District officer described how “Participatory Forest Management was started in Tanzania because of a lack of government staff to run the forests, and high levels of destruction” (Interview P28). This view was 138 elaborated upon by a former Forestry and Beekeeping Division Ministerial Staff Member, who argued that sustainable development was the main issue and that CBFM was the best option to achieve this because whilst “who manages it is not the big issue... the Forestry and Beekeeping Division had inadequate resources to enforce its laws and regulations” (Interview P2). Members of both the wildlife and forestry research villages commonly explained that CBNRM had been implemented primarily for conservation reasons, because the previous systems were seen to not be working, and high levels of both wildlife poaching and forest harvesting were taking place, endangering the natural environment (Interview P42, Interview P97). 5.3.2 The Rationality of Community-Based Conservation The emergence of CBNRM in Tanzania was also driven partly by the perceived inadequacies of fortress conservation practices and fences and fines approaches to conservation, which played an important role in the apparent necessity of a new approach to end what had become a ‘war for biodiversity’ (see Neumann, 2004a). Militaristic anti-poaching operations such as Operation ‘Uhai’ (life/living) in 1989 were seen as ineffective and financially unsustainable (Hahn and Kaggi, 2001). CBNRM offered a new approach in which hearts and minds could be won at considerably less cost to the central government. Nelson et al., (2007) argue that in the wildlife sector, policy reform was a product of ongoing poaching problems, especially of rhino and elephants, alongside decreasing budgets faced by the Wildlife Division, which combined to heighten the perceived need to attract foreign donors through externally-funded projects. Many of these donors were bilateral development agencies such as the Swedish International Development Agency and the Overseas Development Agency (UK), which both funded long-running programmes in the forestry and wildlife sectors respectively (Hartley, 1997; Wily, 2000). Their development focus, alongside the paradigm shift towards participatory development and bottom-up approaches, resulted in these donors playing a key role in championing the discursive shift towards CBNRM. The agency of external organisations was a product not only of the context of conservation in Tanzania and the power of such organisations to discursively frame the policy agenda, but also of the macro-policy context and economic situation in which Tanzania was situated at the time. Igoe and Croucher (2007) describe the creation of WMAs in Tanzania as in-line with global discourses of governance and human rights, alongside the dominance of landscape-scale conservation in the global biodiversity discourse, but also as a result of the 139 influence of neoliberal models of development. The influence of these models in Tanzania is closely tied to the transition from socialism to multi-party democracy, the economic situation in the 1980s and the adoption of structural adjustment that followed (see 3.2). This wider context was identified as a factor in the move towards CBNRM by a former ministerial staff member, who described the change in forestry policy as necessary to keep in line with changing macro policy situation; Macro-economic changes were seeing the government devolving functions to the local government, and the Director of the Forestry and Beekeeping Division felt they should align with that (Interview P2). The neoliberal models of development that spread in the 1980s provided a mechanism for central government to decentralise, prioritise efficiency and re-regulate (Igoe and Croucher, 2007). They define this re-regulation as “the use of the state to commodify previously untradeable resources-those that were previously not owned, state-owned or community- owned” (Igoe and Croucher, 2007: 537). The opening up of natural resource sectors to markets through de-regulation, alongside decentralisation laid the foundation for the win-win scenarios predicted in CBNRM. In a clear link to the concept of governmentality and Agrawal’s (2005) idea of new environmentalities, Igoe & Croucher (2007) argue that the result was the creation of local communities as legal partners in business ventures and simultaneously “disciplining local people to exclude themselves from their land (Igoe and Croucher, 2007: 538). CBNRM is often discussed in terms of win-win scenarios for conservation and development objectives, but this argument highlights the nature of neoliberal models in conservation policy premised upon the devolution of power to local levels as a political technology of the state (see 2.3.1). This argument is supported by Sunseri (2005), who argues that the new forestry paradigm in Tanzanian is one where the language of CBNRM and participation is used to conceal the hegemonic discourse and interests of international conservation organisations. This creates a system of CBFM that has two sides; the appearance of devolved power over resources; and the expansion of state control over unreserved lands. In the wildlife sector, Nelson et al., (2009) argue that the model for CBNRM adopted in the WPT (1998a) and subsequent legislation can be explained in terms of these concurrent vested interests of the state and external influences, especially TNCOs. They compare the governance system operating in a pilot WMA with a competing CBNRM project focused on wildlife in Loliondo, Ngorongoro District, that had refused to follow the prescribed path to gazettment of a WMA. They conclude that the official model, whilst highly flawed in 140 comparison regarding the income being generated for local communities at the time, became the preferred option for both the state and TNCOs operating in Tanzania as it incorporated controlled devolution of power to local communities alongside a maintained reliance upon international funding and aid (Nelson et al., 2009). These studies suggest that devolution has taken place alongside recentralisation (Ribot et al., 2006) within CBNRM policy in Tanzania, and highlight the need to investigate the policy processes, and the discursive processes this entailed, which resulted in the publication of both the new wildlife and forestry policies and shaped the legal formal of WMAs and CBFM (see also 2.1). 5.4 Policy Pathways In this section I consider the process of policy reform and examine further factors that helped shape these processes and the pathways adopted in policy. The policy space created by external donors, specifically bilateral donors and TNCOs in Tanzania at this time has been discussed as a function of both the wider discursive shift towards CBNRM and the rational option that alignment with these new priorities provided for the Tanzanian state in a context of neoliberal change and conservation challenges. The setting of the CBNRM agenda is the outcome of a combination of these factors, therefore. However, the distinct differences between the CBNRM policies in the forestry and wildlife sectors require further examination. Nelson et al. (2007) argue that the role of external organisations, described above, is tempered by the interests of the policy community, where the political economy of different resources and the incentives these provide for bureaucrats, drive policy processes. In the following section, I discuss these arguments with respect to the different policy pathways adopted in the forestry and wildlife sectors in Tanzania during the late 1990s. 5.4.1 The Politics of Devolution: Political Economies of Natural Resources In their outline of policy networks, Marsh & Rhodes (1992) described the relationship between the wider policy network and the policy community as hierarchical. The moderation of external influences in policy processes can thus take place according to the interests and beliefs of the policy community, and Nelson & Blomley (2010) state that in Tanzania these interests differ between the forest and wildlife sector according to the political economy of natural resources. The politics of power devolution (see 2.1.1) with regards to Tanzanian CBNRM are distinct between these two sectors, based upon the perceived value of forest and wildlife resources and the systems of power and benefit that surround them. Nelson & 141 Blomley (2010) discuss these roles in terms of ‘hidden economies’ (or informal systems of power and benefit; see 2.1.3) of these natural resources, which provides the context for these power struggles. Tanzania’s centralized tourist hunting concession system (see 3.3.2), in the hands of the Wildlife Division since 1998, has witnessed a ten-fold growth in value of animal resources, and the Wildlife Division is now responsible for the maintenance of Africa’s largest tourist hunting land area (Nelson and Blomley, 2010). Wildlife is big business in Tanzania, therefore, and these resources have become highly politicised. Within a system of power that privileges those who have access and control over these resources (see 2.1.3), wildlife resources were described by one District Officer as “hotcakes” because of the large amounts of money involved (Interview P28). Whilst the objective of my research was not to carry out an in-depth analysis of the hunting system in Tanzania, as this has already been discussed by several researchers (see Siege, 2001; Baldus and Cauldwell, 2004; Leader-Williams et al., 2009), I do believe that the hidden economy of wildlife hunting has played an important role in shaping the policy pathway and the prescribed governance system of WMAs. The current hunting system is widely regarded as opaque and controlled by rent-seeking behaviour, which provides a clear disincentive for the devolution of power over these natural resources and authority to manage hunting investments within hunting blocks, as described in 5.2.4 (Nelson and Blomley, 2010). The hidden economy of wildlife hunting is dominated by informal networks of exchange and power in which a centralised revenue system has kept prices artificially low through the lack of an auction or tendering system, which has resulted in administratively derived rents and facilitated rent-seeking, patronage, corruption and bribery amongst the influential bureaucrats in charge of the system (Nelson and Agrawal, 2008; Leader-Williams et al., 2009). Baldus & Cauldwell (2004: 1) summarised the threat posed by the devolution of power over wildlife hunting in Tanzania for the Wildlife Division whose staff preside over “a [neopatrimonial] command system of control that favours a select group of hunting outfitters with reduced income generation and the exclusion of rural communities”. In direct contrast to the wildlife sector, the political economy of forestry resources in Tanzania incentivised the implementation of CBFM (Nelson and Blomley, 2010). Tanzania’s history of reserving forests, for high-value timber extraction and watershed conservation especially (see Lovett, 2003; Hurst, 2004), encouraged the adoption of CBFM, which, through the creation of VLFRs would, at very little cost, bring large areas of land, 142 previously unreserved, under the PA status (Nelson and Blomley, 2010). In contrast to the wildlife sector, the initiation of CBFM did not involve the devolution of power over PAs formerly under the jurisdiction of the Ministry of Natural Resources & Tourism56 (Wily, 2000). The low economic value of most of the miombo woodlands upon which CBFM has been implemented also served to limit the perceived loss of potential income for the Forestry and Beekeeping Division through the gazetting of VLFRs (Nelson and Blomley, 2010). These political economies of natural resources operating in Tanzania produced distinctly different ministerial-level divisions and policy communities in the wildlife and forestry sectors. A former development agency staff member described these differences as “two different worlds, with the Forestry and Beekeeping Division being poor but open, and the Wildlife Division being rich but more closed and rigid” (Interview P1). The respondent went on to discuss how the politicisation of wildlife resources entails much more serious and high consequences because of the large amounts of money involved and the close connections with rich hunting groups, whereas forest resources are perceived as important to wood and energy supplies, but not involving the large amounts of money and high value resources that the wildlife sector deals with (Interview P1). The differential politicisation of natural resources between the two sectors and the long history of sectoral policies has also produced little communication between the sectors. Staff working on the MBOMIPA project described the difficulty in dealing with forestry issues when their affiliation was with the Wildlife Division, who “just did not deal with forestry issues” (Interview P74). This highly separate nature of wildlife and forestry sectors at the national level creates separate bureaucratic systems to deal with forestry and wildlife issues, resulting in the same divisions being replicated at local levels, where members of the VNRC in Kiwere discussed their lack of opportunities to increase understanding of wildlife issues, due to their connection to the Forestry and Beekeeping Division, and the forestry-only focus of training they had received. They also argued that legislation in the two sectors did not fit together (Interviews P34, P42). The result of these different political-economic and historical forces on the policy pathways adopted in the development of CBNRM in Tanzania is tangible. In the following section I 56 Land under forestry reserves has usually been channelled into Joint Forest Management agreements instead of CBFM. I have not discussed Joint Forest Management as its level of community authority and participation cannot be classified as CBNRM 143 discuss how the politicisation of natural resources is linked to the ways government officials have resisted the introduction of CBNRM and the power devolution that this entails. 5.4.2 The Politics of Devolution: Resistance and Recentralisation A former ministerial staff member stated that support from Wildlife Division staff for CBNRM varied widely and that “some people using forests for illegal gains saw the transfer of forests to another organ as very threatening and destabilizing. There was some resistance from forest officials” (Interview P28). Such resistance on the part of ministerial staff was explained by one former donor-funded project staff member as directly due to the financial resources available to ministerial staff who, especially in the case of the Wildlife Division, were in receipt of lower wages compared to those working for the Tanzania National Parks Authority (TANAPA), which motivated corruption and illegal means of supplementing income (Interview P77). Ministerial resistance was also found in scepticism of the CBNRM approach. This was explained in terms of the shift in attitude and approach that it required, particularly the shift in role for ministerial staff from one of ‘enforcers controlling local populations’ to one ‘working with them as colleagues’. A former ministerial staff member described serious doubts during the period of policy formulation and implementation regarding the impact upon wildlife and forestry resources, but also about the capacity of local communities to manage such resources, and the advisability of letting them try, particularly because of the large amounts of money involved (Interview P2). The paradigm shift to CBNRM was uncomfortable for many of the Wildlife Division staff who, a former staff member explained, were generally “trained as biologists and ecologists and do not have the skills to deal with managing people in this way, only chasing and arresting them as poachers” (Interview P80). Instances of both ‘foot-dragging’ resistance to devolution (see Scott, 1985) and recentralisation were also highlighted by interview respondents with respect to both sectors. Respondents from within the Forestry and Beekeeping and Wildlife Divisions reported that whilst the ministry talked about and promoted a discourse of decentralisation and devolution, for many individuals, commitment to it was lacking (Interview P2) and “whilst they officially praised it, [they] tried to sabotage the whole operation. They would continue to act as enforcers, arresting people and wielding their power in the villages” (Interview P80). In the forestry sector, where devolution of power is rapid in comparison to the wildlife sector (see 5.2.5), the political economy of natural resources was seen as a driving factor in the type and 144 speed of devolution processes by interviewees. Under the umbrella of Participatory Forest Management, large variations in the time taken to complete the gazettement process both for CBFM and for Joint Forest Management are related to the value of timber resources within the proposed area, which acts as a disincentive to devolution, encouraging ministerial or district level resistance (Mustalahti, 2007; Mustalahti and Lund, 2009). University researchers reported that in the forestry sector district authorities can be reluctant to relinquish control over forests, or in some cases, especially once revenue begins to accumulate from CBFM projects, can be keen to manipulate the projects to increase the role of the district (Interviews P2, P7; see also 6.3.2.1). In such cases, they reported that the prescribed governance system for revenue collection and taxation may be avoided (see 7.2). This is supported by the review of Participatory Forest Management in Tanzania carried out by Blomley & Iddi (2009: 42), which found that “some District Councils have either deliberately or accidentally misinterpreted the Forestry and Beekeeping Division Forest Harvesting Guidelines and as a result were placing additional burdens, barriers or costs to villages regarding harvesting”. One respondent went on to describe a particular example of the district authorities in Suledo orchestrating authority over the sale of forest products from the VLFR, despite having no legal right to do so, as legislation stipulates that this is controlled by the management plan for the VLFR and overseen by the VNRC at the village level (Interview P2). The economic value of natural resources is clearly an important factor in the political process of devolved environmental management. Alongside the hidden economies surrounding wildlife resources, these indicate important political-economic forces that have shaped the devolution of power and the prescribed governance system set out for WMAs. This is particularly clear in the retention of ministerial roles and authority, for example over the financial arrangements for hunting block concessions (Baldus and Cauldwell, 2004), and the bureaucratic obstacles in the path of power devolution, for example through the implementation of a highly complex and lengthy application procedure for gazettement of a WMA (see also 5.2.5, Igoe and Croucher, 2007; Nelson, 2007). In comparison, the different political-economic forces shaping the policy pathways and prescribed governance system in CBFM result in an application system that is simpler and shorter to complete and a form of power devolution that is less restricted by ministerial level involvement. There are however similar forces operating in the forestry sector with regards to valuable timber resources, 145 which due to their value are often channelled into Joint Forest Management arrangements, or the process of VLFR gazettement slows down considerably (see also Mustalahti, 2007). The value of natural resources also plays a key role in their re-politicisation under CBNRM. The politics described above being played out at the national and district levels were quickly joined by similar conflicts at the local level where new perceptions of the value of natural resources initiated new struggles over the positions of power within the governance systems for both MBOMIPA WMA and Kiwere VLFR and the responsibility for managing revenues from the projects. This was supported by an academic researcher, who described how “Conflicts are seen in projects where the resource is large and producing tangible revenues. Where the resource is insignificant, there is very little conflict” (Interview P7). In section 6.3 I discuss these conflicts, and their role in the creation of the performed governance system, as scalar struggles. Finally, the ongoing nature of these struggles and the continual re-politicisation of natural resources has been shown more recently in Tanzania in response to the emergence of Reduced Emissions from Deforestation and Forest Degradation (REDD) at both the policy scale and project level. These new power struggles over the politicised environment (Bryant and Bailey, 1997) were discussed by an interview respondent as “a war going on inside the MNRT because all of the departments want control of the resources that will be under REDD and all want to have their say in its development”, resulting in the different divisions within the Ministry of Natural Resources and Tourism all vying for control and prominence within REDD policy processes and reform (Interview P3). 5.5 Policy processes Thus far, this chapter has attempted to highlight the complexities of policy processes as representing multiple competing agendas and incentive structures. Here I adopt a policy process and policy network perspective to explore the extent to which policy and CBNRM in Tanzania has been constructed through the work of diverse actors within a policy network, rather than a policy elite guided by scientific evidence in a rational manner (see 2.2). I see the myriad advantages and disadvantages that different options or pathways present as a key aspect of both the policy processes actors are participating in and network formation. I employ a conceptualisation of a policy network consisting of multiple, competing discourse coalitions (Bulkeley, 2000), and within which actors are engaged in discursive processes of 146 agenda setting, discourse institutionalisation and policy space construction (see 2.2.3; Keeley and Scoones, 2003). I examine the policy networks involved in both the formation of the Wildlife Policy for Tanzania (Government of the United Republic of Tanzania, 1998a) and the National Forestry Policy (Government of the United Republic of Tanzania, 1998b), and the shift towards CBNRM that these involved. Whilst it has already been recognised that both TNCOs and bilateral donors, alongside political-economic factors, played a key role in agenda-setting and the policy pathways adopted, I argue that an additional layer of insight can be added through an analysis of the policy networks surrounding the development of CBNRM, particularly with respect to the ways in which policy is co-constructed by actors across levels of the network, including the local level (see Keeley and Scoones, 2003). In section 5.5.1 I analyse competing narratives and discourse coalitions (see 2.2.2.2) within policy networks associated with Tanzania’s wildlife sector. Secondly, I discuss the role of policy entrepreneurs (see 2.2.2.4), highlighting the roles of key individuals at the ministerial levels (and within the policy community) in driving the policy pathways adopted in both sectors. Finally, the role of key pilot projects in both forestry and wildlife CBNRM operating within the country at the time of policy formulation and implementation are discussed. 5.5.1 Wildlife Policy Networks: Competing Narratives and Coalitions in Policy Processes The adoption of CBNRM into wildlife and forestry policy in Tanzania was part of the global shifts towards sustainable and participatory development, the twinning of conservation and development objectives and community conservation that is discussed in section 1.1. This shift in policy practice came about through a change in the dominant conservation discourse operating within both sectors in Tanzania and a focus on the role of community in conservation (see also Barrow et al., 2001). In this section, I discuss the role that external organisations and policy entrepreneurs played in facilitating this shift towards CBNRM, and the competing discourse coalitions that were active in the policy formulation process. The shift towards participatory development and community conservation by TNCOs and bilateral donors was part of the growing institutionalisation of these discourses at the global level, underpinned by both neo-malthusian and populist narratives of the need to reconcile environmental and development objectives through sustainable development (Adger,et al., 2001; Adams, 2004). The ‘win-win’ discourse of community conservation drew on ideas of the efficiency and social equity of involving communities in conservation efforts and 147 narratives that argued for the necessity of incentivising sustainable natural resource management by channelling benefits to the local level in order to foster local support (see 1.2; Adger et al., 2001). Whilst these ideas were anchors around which the discourse of community conservation developed, Adams & Hulme (2001b) argue that variety in understandings of both community participation and economic welfare57 produced a number of policy narratives, discourses of community conservation and associated approaches. TNCOs and donor-funded projects that subscribed to this ‘win-win’ discourse, and were implementing the different approaches to CBNRM at the time, were important actors within the discourse coalitions that emerged (Benjaminsen & Svarstad, 2010). The important roles of TNCOs and bilateral donors in the discourse coalitions that developed in Tanzania were the result of their policy space both to set the agenda for a shift towards community conservation in Tanzania and attempt to institutionalise their particular approach to community conservation through the use of policy narratives. This policy space was a product of their long history of involvement in donor-funded projects across the nation in both the wildlife and forestry sectors, their close connections with the policy community as a result of this (see 5.5.3 for further discussion of this), and the large amount of financial resources they brought into these sectors (see 5.3.1). The policy shift to community conservation in Tanzania was a combination of these external influences alongside ministerial advocation of the inclusion of local communities and decentralisation of natural resource management. Including communities in conservation became an increasingly rational policy option for the Wildlife Division, particularly in the context of neoliberal reform and the perception of poor results from strict PAs (see 3.2.2 and 5.3.2) that were supported by with crisis narratives within Tanzania of biodiversity loss, deforestation, and the inability of wildlife populations to survive in pockets of PAs, even if strictly protected (Adams, 2004). The shaping of policy pathways according to the influences of TNCOs and bilateral donors is an important aspect of the processes of policy reform, but the role of such external organisations in the policy networks of forestry and wildlife CBNRM emergence is not a united or simple one. The discursive shift from protectionist conservation to CBNRM took place within a context of competing narratives and discourses as to how this should be implemented. CBNRM is part of a broader collection of approaches within community 57 Based on distinct values of nature and the ways local people can and should be involved in conservation (Adams & Hulme, 2001b). 148 conservation, and within Tanzania several different approaches were being implemented in different donor-funded projects prior to national policy formation (see 3.3). Whilst CBNRM emphasises devolution, community management and sustainable use of natural resources, the issue of wildlife utilisation divided the wildlife policy network, and the donor projects into distinct discourse coalitions. A former REWMP staff member described how this divide not only split the projects in Tanzania, but these were part of a larger debate taking place across Southern and Eastern Africa and within conservation and development networks at the global level. The respondent argued that “the position of the country in Africa was hugely important for the development of community conservation: South Africa was closely associated with CAMPFIRE and Zambia’s programmes, whilst Kenya had adopted a non-utilisation policy and was strongly influenced by the African Wildlife Foundation” (Interview P77). In addition, within the policy network in Tanzania, TANAPA had initiated a different type of community conservation in its programme called the Community Conservation Service (see 3.3). There were therefore three competing discourses of community conservation operating at the time of policy agenda setting, therefore (Table 5.3). 149 Community Conservation Discourse Characteristics Community-Based Natural Resource Management Park Outreach Key Components of Discourse Devolution of rights to the local level Participation of local communities Benefit sharing Funding for community development projects Key Projects Operating Selous Conservation Programme, REWMP/MBOMIPA Serengeti Regional Conservation Strategy Community Conservation Service (nationwide) Wildlife Utilisation Yes No No Examples followed Zimbabwe’s CAMPFIRE, GTZ’s Sustainable Use Strategy African Wildlife Foundation, Kenyan model Policy Community Members Wildlife Division TANAPA Discourse Coalition Members USAID, GTZ, ODA/DfID WWF, African Wildlife Foundation Table 5.3 shows that the three competing discourses of community conservation in Tanzania fell into two broad categories; one based on ‘park outreach’ and ‘benefit sharing’; and the other based on CBNRM, although split into two distinct coalitions according to views on the issue of wildlife utilisation. Representatives of TANAPA, drawing on its experiences with its Community Conservation Service (CCS) put forward a discourse of park outreach and benefit sharing based on a policy narrative that local people’s destructive environmental practices could be reduced, alongside maintenance of strict protection of wildlife resources in PAs, by channelling funding for income generating projects to the local level and supporting livelihood alternatives that did not pose a threat to conservation objectives (Bergin & Dembe, 1996). The development of the park outreach approach within TANAPA was heavily influenced by its connection and funding provided by the African Wildlife Foundation, and particularly Patrick Bergin, who became a key member of the park outreach discourse coalition, and was instrumental to the African Wildlife Foundation’s anti-utilisation stance Table 5.3: Competing Discourse Coalitions in Tanzanian Community-Based Conservation Prior to National Policy Development (see also Bergin and Dembe, 1996; Walsh, 2000; Barrow and Murphree, 2001; Nelson et al., 2007; Roe et al., 2009). 150 (Interview P77). The leader of TANAPA at the time, who was personally very committed to initiating a form of park outreach and benefit-sharing and to building good relations between the National Parks and local people was also vital to the development of this particular discourse of community conservation in Tanzania (Bergin & Dembe, 1996, Interview P77). TANAPA had a long history of responsibility for managing the nation’s National Parks and anti-poaching duties and employed militaristic management strategies to achieve this (Bergin, 2001). Strict protection and enforcement by TANAPA staff was based on a narrative of ignorant and environmentally destructive local people, which identified those found within park boundaries as poachers. Bergin (2001) described this as a ‘de-humanising’ narrative that was used to justify exceptional action, including violence, to combat the threat posed to natural resources, and particularly wild animals. The relationship between TANAPA and communities living close to strict PAs was very tense, maintained also by local narratives of TANAPA staff that described them as savage, cruel, delighted to wield their power over normal people trying to earn a living and uninterested in understanding why people sometimes entered the park (Interview P73). As TANAPA tried to find a way to deal with these antagonistic relations with local communities, whilst at the same time maintaining its remit to protect Tanzania’s National Parks, they created the CCS around a narrative of community conservation based on reciprocity, the value of nature and ‘ujirani mwema’58. TANAPA’s CCS was based on an argument that if only communities understood the value of wildlife, they would not exploit resources indiscriminately, and that community conservation should focus on sharing benefits with local communities as a means of making these values tangible to local people (Bergin, 2001). Reciprocity was a key element to this narrative, based on the idea that villagers’ engagement in conservation-related activities would be rewarded through the provision of social services at the local level (Bergin, 2001)59. The concept of ujirani mwema was vital to the development and success of TANAPA’s CCS as it worked to re-build relations with local communities through a form of community conservation that was acceptable to TANAPA, allowing National Parks to remain under strict protection, and TANAPA to remain in charge of protecting them. The shift to community conservation threatened the status and responsibilities of TANAPA, but the concept of ujirani mwema and its application through the CCS created an identity for the organisation and its 58 Meaning ‘good neighbourliness’. 59 This was later developed further into grants for income generating activities, operating as Support for Community Income Projects (see Bergin & Dembe, 1996) 151 National Parks not as “evil, colonial relics that should be turned back over to local communities, but rather that parks should have a good and mutually beneficial relationship with surrounding people” (Bergin, 2001: 98). Within the alternative CBNRM discourses, conservation is understood less as a concern over ‘wild’ spaces and nature conservation, and more as the sustainable management of natural resources and ‘wise use’60. A central theme within the two discourse coalitions that supported CBNRM in Tanzania, that set them apart from the park outreach discourse, was the devolution of power and management rights over natural resources at the local level. These two discourse coalitions were agreed that benefit sharing and park outreach were insufficient as a form of community conservation, and that local people needed a much greater role in decision-making regarding natural resources. Both of the CBNRM discourse coalitions argued that as wildlife resources occurred on village land, and that local people often bore the costs of conservation, they should be responsible for managing those resources, not simply compensated (see 2.1). The two CBNRM coalitions were split however on the issue of wildlife utilisation through tourist hunting activities. Whilst both discourse coalitions promoted the realisation of economic benefits from wildlife management to the local level, one set of actors argued that non-consumptive uses of wildlife resources was the only form compatible with conservation objectives, the other argued that the economic benefits that could be realised from tourist hunting were too large to remove, and formed the greatest opportunity for channelling significant funds to the local level. Several of the pilot projects operating in Tanzania at the time, including the Serengeti Regional Conservation Strategy (supported by the African Wildlife Foundation), argued that a sustainable system of conservation and generation of revenues could be achieved through non-consumptive uses only, particularly photographic tourism. These members of the policy network drew on examples from Kenya to support their claims for a form of conservation that is based on intrinsic values of wildlife as a source of income to support local communities. The final competing discourse was represented by other donor-funded projects operating at the time, including MBOMIPA and the Selous Conservation Programme. This discourse argued for the conservation of nature for utilitarian reasons, based on the ‘wise use’ of natural resources to meet human needs and uses (see Adams & Hulme, 2001b) and advocated the maintenance of wildlife hunting within CBNRM. The members of this discourse coalition 60 The concept of ‘wise use’ was championed in the USA by Gifford Pinchot in the early twentieth century (Adams & Hulme, 2001) 152 drew on examples from Zimbabwe’s CAMPFIRE programme (see 1.2) and the history of wildlife hunting in Tanzania to support their argument. The competing discourses of CBNRM were played out by actors associated with these multiple organisations during the process of policy formulation as different actors and groups tried to influence the direction of community conservation. The clearest example of this is the Planning and Assessment for Wildlife Management project (PAWM). PAWM began in 1990 as a means to address the future of the wildlife sector, draw up national plans and promote sustainable economic development of the sector (Leader-Williams et al., 1996) and is seen as the first step towards the creation of the WPT (1998; Interview P76). PAWM was funded by USAID but was jointly managed by AWF and the Tanzanian office of WWF (Leader-Williams et al., 1996). This put the project at the centre of debates over utilisation as the two managing organisations were members of the distinct CBNRM discourse coalitions. Through connections with projects such as PAWM and links to the policy community, actors within both coalitions attempted to drive national policy recommendations according to their own convictions regarding wildlife utilisation within CBNRM (see also 5.5.3). They employed the narratives discussed in this section to make their arguments and contextualised them by drawing on examples from across Southern and Eastern Africa and experiences within the Tanzanian community conservation projects. Ultimately it was the pro-utilisation discourse coalition that was institutionalised as part of CBNRM in Tanzania’s wildlife sector (see 3.4.2, 5.2.4 & 5.5.3). Whilst the TNCOs and donor-funded projects played a key role in the discursive development of CBNRM in Tanzania (see also 5.5.3), a broader range of actors were associated with the different discourse coalitions; the British High Commissioner in Nairobi was very vocal in his support for a policy of non-utilisation, whereas members of the Overseas Development Agency/Department for International Development (who funded the REWMP/MBOMIPA project) were strongly in support of incorporating tourist hunting into CBNRM (Interview P77). Such actors used their political connections to raise the issue in many media, including a discussion in parliament as to the appropriateness of the British Overseas Development Agency being involved with a pilot wildlife utilisation scheme in REWMP, which a former MBOMIPA project officer described as initiated by members of the discourse coalition opposed to utilisation in an attempt to convince more members of parliament to support non-utilisation (Interview P77). The issue further divided staff working on the different pilot projects around the country. An interview with a former REWMP staff 153 member indicated that the stagnation and lack of influence of the Serengeti Regional Conservation Strategy project in policy processes was due to conflict between its staff members, who could not agree on whether it was right to introduce tourist hunting within CBNRM (Interview P77). Within the REWMP project, the eventual implementation of a utilisation component was also highly contested as the project manager was strongly against it, arguing that the focus of any form of community conservation should be to support the work of the national park around which it is based, and as any form of hunting is forbidden within a national park, it had no place within community conservation (Interview P73). Other REWMP staff members subscribed to a narrative that set out the wise use of wildlife resources, based on strict quotas for wildlife utilisation, as the best way of incentivising conservation amongst local communities because of the revenue that it could generate (Interview, P74). The narrative in favour of utilisation posited that wise use would not endanger conservation objectives, but would greatly increase the revenue generated, and would thereby improve the prospects of CBNRM successfully linking conservation and development objectives, upon which successful conservation was seen to depend (Interview P73). As the debate over utilisation continued across the country, supporters of utilisation within Iringa faced stiff opposition from actors who worked closely with Ruaha National Park and advocated park outreach rather than CBNRM, particularly the influential expatriate families who own some of the lodges within and close to the park and the Friends of Ruaha Society (a regional NGO; Interview P73). These actors argued that incentivising conservation should also be achieved through environmental education and awareness and an appreciation of the aesthetic and intrinsic values of wildlife, not purely by the generation of revenue. My Semi-structured interviews with former ministerial staff, development agency staff members, TNCO staff and academic researchers revealed that at the national and ministerial levels, the issue of wildlife utilisation became a critical juncture in the discursive development of CBNRM policy in Tanzania (Interviews P77, P77 & P80). As shown in Table 5.3, the major governmental organisations (the Wildlife Division and TANAPA) involved in the wildlife sector adopted different discourses of community conservation. Both organisations held strong interests in the development of community conservation, and especially the financial resources produced at the national level through wildlife-based tourism, but supported opposing official views regarding the incorporation of wildlife utilisation into community conservation projects. Whilst TANAPA supported a model of 154 non-utilisation, the Wildlife Division firmly defended Tanzania’s history of tourist and resident hunting in Game Control Areas and Game Reserves (see 3.3). As discussed in 5.4.3 the political nature of power devolution combined with the political pressure exerted by funding agencies and TNCOs were important shaping forces in the policy pathways adopted in Tanzanian CBNRM. The politicisation of wildlife resources, particularly as neoliberal models of conservation were influencing agenda setting and policy formulation in Tanzania led to what interview respondents reported as both the Wildlife Division and TANAPA vying for ministerial-level authority over WMAs as the process of policy formation was underway (Interview P77). Ultimately it was the Wildlife Division that emerged as the organisation responsible for developing policy and implementing community conservation in Tanzania, which shaped the resulting policies into CBNRM, rather than park outreach and institutionalised tourist hunting as a component of CBNRM. I argue that the institutionalisation of a CBNRM discourse in the wildlife sector, including wildlife utilisation, was very heavily influenced by the revenue generated by hunting in Tanzania, but also by the power of the sustainable use discourse within conservation. Whilst hunting has remained controversial throughout the history of conservation and in the development of community conservation discourses (Adams, 2004), the concept of sustainable use, underpinned by global environmental management narratives of the ability of the market to regulate utilisation and conserve resources through proper integration with the capitalist economy has become dominant (Adger et al., 2001; Adams, 2009; Buscher et al., 201. A former MBOMIPA staff member described how “hunting was at the core of the Wildlife Division” and, argued that particularly as the wages of its staff were lower than those of TANAPA staff, the creation of rent-seeking opportunities surrounding the hunting system in Tanzania had become important to the livelihoods of many of the Wildlife Division staff (Interview P73). Therefore, the of incorporation of hunting into the governance systems for WMAs in Tanzania presented an opportunity for the Wildlife Division to retain an important source of revenue (and rent-seeking, see 5.4.1 and 7.4.4) whilst aligning the development of CBNRM in Tanzania with globally institutionalised ‘win-win’ discourses of sustainable development and policy narratives in favour of the expansion of the capitalist economy to achieve this. The focus, therefore, was on finding ways for local people to benefit from the tourist hunting industry, opening up tourist hunting markets to CBNRM, rather than removing them (Nelson et al., 2007). 155 The large amounts of revenue that could be generated through wildlife utilisation were at the heart of the arguments for the maintenance and incorporation of tourist hunting into CBNRM in Tanzania (Interview P77). The value of wildlife resources for non-consumptive uses, particularly photographic tourism, was realised early on and was a driving force behind the development of conservation more generally in Tanzania, as discussed by Julius Nyerere: “I personally am not very interested in animals. I do not want to spend my holidays watching crocodiles. Nevertheless, I am entirely in favour of their survival. I believe that after diamonds and sisal, wild animals will provide Tanganyika with its greatest source of income. Thousands of Americans and Europeans have the strange urge to see these animals” (Quoted in Nelson et al., 2007: 238) The value of the wildlife-based tourism industry was dwarfed by the per capita income from tourist hunting, however (see 3.3.2), and this formed a powerful argument for incorporating utilisation into the ‘win-win’ strategies for CBNRM (Nelson et al., 2007). The utilisation discourse coalition also drew on a narrative that tourist hunting provided the means by which the largest number of local communities could benefit from CBNRM; In areas that were poorly developed for photographic tourism and lacking in infrastructure to develop these services, tourist hunting could still provide economic benefits, whereas a policy on non- utilisation would leave many rural areas without a mechanism to generate revenue from the management of wildlife resources (Nelson et al., 2007). Finally, the personal beliefs and experiences of important actors within the policy community were integral to shaping the policy pathways adopted and the CBNRM discourses that were institutionalised in Tanzania in the late 1990s. It is important to note however that discursive institutionalisation of CBNRM does not equate to unwavering support by all individuals within the discourse coalition or policy community (see also 5.4.2). The following section considers the role of key individuals in both the wildlife and forestry sectors to the institutionalisation of a CBNRM discourse in both the Forestry and Beekeeping Division and the Wildlife Division, and particularly the decision to incorporate utilisation in CBNRM in the wildlife sector. 5.5.2 Policy Entrepreneurs In the study of policy processes, the roles of key individuals and chance are often not acknowledged. Personal beliefs of these key individuals can be instrumental in pushing through policy reform that is facing resistance, hesitation or debate both within the policy 156 community or the wider policy network (Barrow et al., 2001). For example, the development of the CCS in Tanzania was the direct result of the work of the Director of TANAPA between 1989 and 1992, despite the risks involved in implementing an untested and initially unpopular idea (Barrow et al., 2001). In the mid to late 1990s the Directors of both Forestry and Beekeeping and Wildlife were individuals who advocated policy reform and the introduction of community conservation, and especially CBNRM, in Tanzania (Ylhäisi, 2003; Interviews P2, P80). In an interview, the former director of the Forestry and Beekeeping Division implied that promoting the discursive shift to CBNRM and making national policy reform were his first priorities on taking up the role (Interview P2). He described facilitating policy reform through discursive processes to recruit members of the policy community to a populist narrative of ‘forests are wealth61’, shifting perceptions of forests from a national asset to a view that accepted that “these people surrounding the forest have legitimate needs from the forest...they are the ones who know who is conducting illegal harvesting…they are the best people to manage the resource” (Interview P2). The rationality of this narrative builds on protectionist narratives of the need to safe-guard natural resources, which were familiar to the policy community, whilst at the same time shifting the argument to conceptualise local communities as best- placed to achieve this. The respondent’s description of these discursive processes clearly indicate the importance of policy entrepreneurs, particularly those in positions of authority, who are able to manipulate the policy network and discursively cultivate support for their own beliefs, priorities and policy agenda (Fischer and Forester, 1993; Hajer, 1995). His account also exemplifies the discursive nature of power through the use of policy narratives and story lines (Roe, 1991; Hajer, 1995; and see 2.2.2.2) to socially construct environmental problems, control the issues on the policy table, reduce uncertainty or alternatives and organise actors into discourse coalitions (Hajer, 1995). Adger et al., (2001) describe the process of creating discursive homogeneity when a specific discourse comes to dominate thinking and is translated into institutional arrangements. I argue that the Director of Forestry and Beekeeping was engaged in such a process of creating discursive hegemony in which he actively recruited other members of the Forestry and Beekeeping Division to support a shift to CBNRM using the social power associated with his title and discursive tactics to persuade people of the rationality and moral imperative of such a policy shift. 61 ‘Misitu ni mali’ is a well-known Kiswahili phrase 157 Similarly, in the Wildlife Division, the views and beliefs of the new Director, appointed in 1995, were crucial to the discursive institutionalisation of CBNRM at the ministerial level, and particularly to the adoption of utilisation as a component of this policy shift. The new Director of Wildlife’s beliefs concerning utilisation were shaped by his long history of employment with community conservation projects in Tanzania and across Southern Africa Former project staff who worked closely with him at the time reported that his support for utilisation came from experiences working on both the Serengeti Regional Conservation Strategy which, until 1998, employed a policy of non-utilisation, and later experience in Botswana (Interview P77). WMA pilot project staff reported that these experiences in both utilisation and non-utilisation contexts were formative in the Director’s opinions and his employment of a narrative that utilisation represented “the only way forward for a successful community conservation strategy in Tanzania” (e.g. Interview P77). The strong disincentives for the devolution of power over wildlife resources, as discussed in section 5.4, made the personal commitment of the Director to the approach a key driver in the development of a national policy and the adoption of the utilisation strategy within CBNRM in Tanzania. The key role that these entrepreneurs in strategic positions of the policy community played in the formation of policy, and the pathways adopted in this, is exemplified in the subsequent changes that took place in the wildlife sector. In 1999 the director who had presided over the publication of the WPT (1998), and thereby developed the first legislation to incorporate CBNRM in the wildlife sector was described as being forced out of his position due to the politics surrounding the management of wildlife resources, and the controversy within the policy community over WMAs (Interview P77). The new director had very different ideas about community involvement, and worked to implement changes in the policy framework that shifted the nature of WMAs. Respondents from the ministerial, academic and NGO sectors explained that the changes he attempted to put in place were significant, although subtle (Interviews P77, P80). Former Wildlife Division staff members reported that the new Director and his team “didn’t think that communities could achieve anything and they put obstacles to their achievements in the form of the WMA guidelines, by making them so complex that it stalled the process and made it very difficult for communities to go through and understand” (Interview P0). There is a strong link to these shifts within the dominant discourse of CBNRM in Tanzania, which altered with the change of director, and the politics of devolution described in 5.4.3. Whilst the official discourse of CBNRM could not be abandoned in Tanzania, the new policy 158 community held the next step of policy formulation (to produce the Regulations and Guidelines62 for the implementation of WMAs in Tanzania) within their power. In effect these policy documents provide the detailed basis, rules and systems that are implemented in WMAs. A former member of the Wildlife Division staff reported that the new Director was “very corrupt and very authoritarian” and viewed the implementation of WMAs through the WPT (1998) as highly threatening to the established channels of corruption and personal benefit within the wildlife sector, especially around hunting (Interview P80). The change in Directorship within the Wildlife Division in 1999 not only signalled a shift in the dominant discourse coalition within the policy community, but is an excellent example of the control exerted by the policy community in processes of policy reform, and the manipulation of policy processes according to individuals’ private agendas. The examples from the Forestry and Beekeeping Division and the Wildlife Division have highlighted the important roles that such policy entrepreneurs within the policy community can play in shaping policy processes, and how the ways that they do this are determined by their own personal beliefs and private interests. It is important to note that I do not argue that these influences demonstrate the elite control over decision-making that has often been conceptualised in linear models of policy reform (see Thomas and Grindle, 1990), but that they highlight the discursive nature of policy processes, and the central role of power over natural resources in shaping the discourses about their management. The Tanzanian wildlife sector is an excellent example of the politics of power over wildlife resources (see Duffy, 2000; Brockington et al., 2008; Nelson, 2010) and how the devolution of this power as envisaged in the WPT (1998) sparked a political struggle between different interest groups and discourse coalitions within the policy community. The way these struggles were played out discursively within the wildlife sector is shown in the development of the first WMA Regulations (Ministry of Natural Resources and Tourism, 2000a) and WMA Guidelines (Wildlife Division, 2001). The creation of the guidelines followed a participatory process in which representatives from pilot projects such as Selous Conservation Programme (SCP), Serengeti Regional Conservation Strategy (SRCS) and MBOMIPA were invited to discuss and contribute to their formulation. During semi- structured interviews, participants from these workshops described a change in atmosphere and relationships with the Wildlife Division at this transition of director. After the 62 MNRT (2000a; 2001a) 159 appointment of the new Director of Wildlife in 1999, respondents described a clear feeling of unwillingness to relinquish control over wildlife resources from the Wildlife Division staff they dealt with, and the outcome of the participatory workshops as unsatisfactory, especially over issues of financial resources and the application procedure to create a WMA (Interview P74). Respondents recalled feeling that these areas were beyond their control or influence, and they expressed the Wildlife Division staff’s lack of interest in their opinions and suggestions on these issues (Interview P1). Within a discourse of participation and collaboration with the policy network, the policy community utilised the policy processes in creating these documents as a means to recentralise control and power over wildlife resources (Shauri, 1999; Nelson, 2007; Nelson et al., 2007). A former REWMP staff member described how they felt that the shift towards CBNRM that was taking place at the time in Tanzania directly catalysed a “drawing back” by those who were unsure about it (Interview P77). The restrictions placed on the devolution of power to local communities, particularly with reference to ownership of wildlife resources and the financial arrangements for WMAs (see 5.2.4) are a clear example of this recentralisation of power (Schafer and Bell, 2002; Goldman, 2003; Ribot et al., 2006). 5.5.3 Pilot Projects in the Issue Network: Roles of Local Level in National Policy Development In this chapter I have shown that the policy reform processes that introduced CBNRM in Tanzania’s wildlife and forestry sectors involved a complex blend of actors and organisations and their interests, enacted through discursive processes. Here I attempt to integrate a further level of actors in the policy network into these processes of policy reform; the role of the local level and specifically pilot projects operating in the country at the time of policy reform. Previous discussions of the emergence of CBNRM in Tanzania, particularly in the wildlife sector, have focused on the national and international actors and networks within this complex story (see Igoe and Croucher, 2007; Nelson et al., 2007). The research findings presented so far in this chapter have developed their findings in many respects to show that the role of key individuals at the centre of the policy community should be integrated with those discussed as important to agenda-setting in section 5.3 and the politicisation of resources discussed in section 5.4. Here I wish to address the failure of previous studies to encompass wider areas of the policy networks involved in the forestry and wildlife sectors. Nelson et al (2007) claim that the local level has little political space in legislative procedures, and whilst they and Nelson & 160 Blomley (2010) emphasise how the development of Tanzania’s model for wildlife CBNRM was drawn from donor-funded pilot projects ongoing in the country, there is no account of the ways in which these projects were able to influence the formulation of this policy. In this section I argue that the study of the complexities of policy processes must expand to acknowledge that these intricacies are contributed to, and played out in, the local arena, which forms part of a multi-level network of coalitions which create, utilise and shape policy space. Reviews of the roles of external organisations and key pilot projects in the development of CBFM in Tanzania are well known (Wily and Haule, 1995; Wily, 1997; Wily, 2000; Wily and Dewees, 2001a; Wily, 2004; Ministry of Natural Resources and Tourism, 2009), but, to date, the roles of such projects in the wildlife sector have not been discussed. In the emergence of CBFM the role played by the Swedish International Development Cooperation Agency funded pilot projects in Babati and Singida districts are often cited as drivers of policy reform (Wily, 1997; Wily, 2000; Wily and Dewees, 2001b). These projects were the first examples in Tanzania of villages declaring a forest reserve on village land and being granted the legal rights to manage that area according to district-approved by-laws written by the community (Wily, 2000). Project staff working in Tanzania at the time attributed the key role played by these projects to the work of another policy entrepreneur working within the projects to promote the ideology of CBNRM at the national level and across Southern and Eastern Africa more widely (Interview P1). However, there has been little discussion of the processes by which these projects and the actors involved influenced the formulation of national policy. Such projects can be an important opportunity for ‘witnessing’ or grounding of experiences within policy reform (Charlton and May, 1995; Keeley and Scoones, 2003). Former Wildlife Division and Forestry and Beekeeping Division staff interview respondents described the role of such projects as key to policy reform through acting as “laboratories and field experiments” (Interview P80), which raises interesting questions about the role of this witnessing in processes of policy development and implementation (see 5.5.3.2). It has been noted that donor-funded projects have agenda-setting influence (see 5.3.1) and that project staff played a role in the formulation of both CBFM and WMA guidelines during policy implementation (see 5.5.2). Whilst the experience in the wildlife sector was described by participants in these processes as restricted and frustrating, respondents from the forestry 161 sector, including former staff of the MEMA63 projects in Iringa described having “real ownership and [a] real share” in the development of the guidelines for implementing CBFM (Interview P1). The process of formal or invited participation in policy formulation in these cases raises questions as to the limits of participation, and is connected with the discussion regarding the role of these projects in the formulation of the guidelines for implementing WMAs and CBFM projects and the changing policy community at this time (see section 5.5.2). In the MBOMIPA case study, the issues they faced in their participation regarding the drafting of the WMA guidelines relate to the beliefs and motivations of the policy community, and the structured political-economic factors driving policy pathways within the shifting policy community (and institutionalised discourse). The restricted nature of local level influence at this stage of the policy formulation and implementation reflects the tightening of the policy community’s boundaries and attempts to minimise the negative impacts of policy reform. Similarly in the forestry sector, whilst respondents from within the former project staff provided very positive feedback about the experience of participating in the development of the CBFM Guidelines, a former Forestry and Beekeeping Division staff member described their role in policy processes as “the path of policy was pre-determined. What the projects did was to support this path and provide real experiences” (Interview P2). This statement raises some important questions about policy processes and the role of the local level within these. Firstly the comment that the policy pathway was pre-determined raises the questions of by whom and how? In the context of the discussion within this chapter, the pre- determination was partly due to the external influences of TNCOs and bilateral donors, and partly due to the beliefs and priorities of the individuals within the policy community, and how these shifted over time. Secondly, the statement is interesting because it alludes to the roles of pilot projects and their participation in policy processes as potentially being strategic for the policy community. In section 5.5.3.2 I will discuss further the concept of policy being co-constructed between the national and local levels. Below I investigate policy reform in the wildlife sector, and the role of the case study REWMP/MBOMIPA project to add further depth of insight to the complexities of the role of the local level in policy network and policy processes. The discussion is organised into two sections, the first of which introduces concrete examples of aspects of the project that were 63 Matumizi Endelevu Maliasili meaning ‘sustainable use of natural resources’; see 3.3.3. 162 adopted into policy. The second section discusses the importance discourse coalitions and policy networks to understanding the policy space that permitted the project to attain such influence in policy processes. 5.5.3.1 REWMP’s Place in Policy Whilst not the first donor-funded CBNRM wildlife project in Tanzania, the REWMP project, which began in 1992, grew to become one of the most important examples of such projects in the country. By 2001 REWMP’s successor, MBOMIPA, was described as one of the top two leading CBNRM projects in Tanzania, alongside the Ukutu WMA64 (which came out of the SCP), and the only one that provided extensive experience of the sale of wildlife quotas to resident hunters (Walsh, 2000; Elliott, 2001). Its high profile meant that it was used by the Wildlife Division as an example of the development of CBNRM in Tanzania on a frequent basis, hosting visits from senior ministry staff, the Director of Wildlife, the UK Secretary of State for International Development, members of other community conservation projects around the country and students from the most prestigious college of wildlife management in Tanzania (Walsh, 1998; Walsh, 2003). This high profile at the national level and good reputation, I will argue, was carefully constructed by actors working within and around the project who, according to their own beliefs, engaged in the discursive struggle of creating policy space and influence. These actors positioned themselves and REWMP to provide significant contributions to the process of policy formulation and implementation. Below I outline three concrete impacts upon policy that came from REWMP. Firstly REWMP led the way in the design of the structures of management authority for WMAs. At a 1994 workshop of community conservation stakeholders and policy makers for the wildlife sector, a representative from REWMP raised the issue of a need to design WMAs to be managed across multiple villages, instead of individually managed sections of the WMA at the village scale (Leader-Williams et al., 1996). The latter was the model being implemented at the time in the highly influential SCP (Hartley, 1996), but the REWMP representative argued that the project had identified concerns over this model based on the socio-ecological characteristics of the project area; highly uneven patterns of human population and land rights by village drove the representative to argue that a WMA must operate across participating villages or risk producing fragmented conservation areas and excluding many villages from participating and benefitting from such a project (Hartley, 64 Usually referred to by the name of the AA for this WMA: JUKUMU. 163 1996). The inter-village model of WMA management is the one that has since been adopted into policy. Two further elements of the REWMP and subsequent MBOMIPA institutional structures have since been adopted into national model for wildlife CBNRM, as set out in policy. The first is the incorporation of a District Natural Resources Advisory Board (DNRAB) into the institutional structure of WMAs, and the second is the right of the communities involved in the project to hold decision-making rights over the use of wildlife quotas on their land (Hartley, 1997; EPIQ, 2000; Walsh, 2003). The formation of a DNRAB, then named the District Steering Committee, into REWMP came at the mid-project review in 1995 (Mackenzie, 1995), and from a community conservation workshop held in Ruaha National Park in November 1994 (Hartley, 1997). The justification for such an institution came from issues of forging links between the local, district and national levels, and because an institution for conflict arbitration was deemed necessary (Hartley, 1997). The structure of the REWMP District Steering Committee (eventually settled at its third meeting in May 1996), and these two responsibilities were adopted into the WMA guidelines from their first and second drafts (Ministry of Natural Resources and Tourism, 2000c; Ministry of Natural Resources and Tourism, 2000d) and the first draft of the WMA regulations (Ministry of Natural Resources and Tourism, 2001b). The second aspect in which a direct influence from REWMP can be identified in the WMA policies resulted from decisions about the wildlife utilisation component in the project. This was not initiated until April 1995 due to complex conflicts between stakeholders and the ideologies of the partners involved in the project (see 5.5.1, Hartley, 1997). Through the transition strategy between REWMP and MBOMIPA (1995, see Walsh, 1996), the project initiated the right of the communities involved to decide on the use of the wildlife quotas allocated to them on village land (Hartley, 1997). This gave them the right to decide whether to sell these quotas to resident hunters, or to hunt the animals themselves to provide meat for the community. REWMP was, therefore, “the first community wildlife project in Tanzania to have established the right of the communities to make choices about the resources occurring on their land” (Hartley, 1997: iv). This right is instilled in law in WMA regulation 51 (Government of the United Republic of Tanzania, 2002a; Government of the United Republic of Tanzania, 2005) and in the WMA guidelines (Wildlife Division, 2001). 164 5.5.3.2 REWMP: Networks, Relationships and the Creation of a National Example In this section, I address the question of how REWMP was constructed and positioned as a major example of wildlife CBNRM within the policy network Tanzania and I investigate the processes by which influence was wielded through the creation of policy space. Hartley (1997: 58) describes REWMP as “in the right place at the right time” because it was an emerging project just as the need for a coherent policy for CBNRM was becoming clear. A REWMP staff member described the project’s influence as “REWMP made leaps forward...mov[ing] the storyline and Selous then had to catch up” (Interview P77). This fortuitous timing that coincided with an established and growing agenda for wildlife CBNRM and calls for a national policy enabled REWMP to mould itself in line with policy and contribute to it on an ongoing basis. This key strategy was possible for REWMP (which was UK-funded) due to the lack of the funding body’s engagement in wildlife conservation projects previously, which meant that project staff were not tied to a project model set out by its funding body (Interview P77). The ability of the project’s staff to engage with this policy formulation process was forged through the connections it developed at the national level and within the policy community. Firstly, project staff had close relationships with the officers working on the PAWM project. Together PAWM and REWMP staff decided that REWMP would implement the models that were being developed for policy, rather than implement a donor-led model (Interview P77). REWMP was therefore able to create its own systems and develop alongside the national rather than a donor agenda (Interview P77). These were important steps in the construction of the discourse coalition that went on to become institutionalised in policy. Through its close relationship with PAWM, this strategy allowed the project to sit strategically at the centre of policy debates, and secured an invitation to present at a PAWM workshop on community conservation in the wildlife sector, held in February 1994 (see Leader-Williams et al., 1996). The conference not only reviewed the projects ongoing in the country at that time, but also debated the concept of a WMA and began the process of policy formation by creating a new draft of the Policy for Wildlife Conservation and Utilisation, introducing CBNRM in a formal policy capacity. It is from this point on that we can see the rise of REWMP and the creation of a national example. The February 1994 workshop placed REWMP within the leading group of high profile CBNRM projects in Tanzania and highlighted the work of the project to the policy 165 community in attendance (Interview P77). The Deputy Director of Wildlife became particularly interested in the work of REWMP and a close relationship developed between REWMP staff and the Wildlife Division as a result (Interview P74). The close connections between the project staff and the Wildlife Division were increasingly important when the Deputy Director of Wildlife took over as Director in 1995. The relationship proved to be decisive in the future of the project, as it stimulated discussions that resulted in support from the Wildlife Division, including the provision of a Game Officer, facilities, advice and high profile support in times of conflict (Hartley, 1997; Interview P76). As a consequence of this developing relationship, in November 1994 a second workshop was held at Ruaha National Park between the Wildlife Division, District officials and REWMP staff, with the purpose of discussing the future of the project (Hartley, 1997). This workshop resulted in REWMP being designated as a pilot of the WMA concept by the Wildlife Division, which signalled that the project had become a major player within the policy network (Interviews P76, P74). The personal relationships created with the policy community are therefore key to the process by which REWMP became a high profile national example of CBNRM. The personal connection to the Director of Wildlife was also an important opportunity for the construction of policy space, which providing a means for the project staff’s priorities, opinions and findings to be heard at policy making forums. As a nationally-recognised pilot of CBNRM in Tanzania, and particularly through the integration of REWMP staff into the dominant discourse coalition and the close connections with the policy community (the Director of Wildlife in particular) and other key actors and organisation within the coalition (especially PAWM), significant policy space was created around REWMP, enabling the influence upon policy that are described in 5.5.3.1. This influence was not one-way however. As mentioned in 5.5.3, the utility of such pilot projects to the policy community, their role in witnessing and their strategic use by the policy network needs to be considered. I argue that the REWMP case indicates that whilst a pilot project may be the focus for construction of policy space, policy is co-constructed between the local and national levels, and pilot projects can act as the centre for such discursive practices. Not only does a pilot project identity legitimise the position of its staff within the policy network, opening space for the inclusion of their particular views into policy processes, but the grounded identity of such projects is also beneficial to the policy community. REWMP’s malleable nature, which helped project staff to position the project within the dominant discourse coalition and align the project with developing policy, left the project open to the 166 interests of those within the coalition and the policy community. In the case of REWMP, the close personal connections between project staff and the Director of Wildlife was significant in the initiation of wildlife utilisation within the project, at the insistence of the Director of Wildlife, despite ongoing conflicts and recommendations to delay further (see 5.5.1). The decision to implement the strategy was important for the status of REWMP, maintaining its position within the dominant discourse coalition and its close support from the Wildlife Division, but was also beneficial to the Director of Wildlife and the policy community at the time, who championed wildlife utilisation within CBNRM, and were able to use the REWMP project to discursively frame, ground and witness their own policy agendas. 5.6 Chapter Summary This chapter had attempted to show how the policies that developed to introduce CBNRM in Tanzania were the result of complex, inter-related factors that represent multiple and competing agendas of different actors over different levels and scales. Whilst the objectives of CBNRM in both the wildlife and forestry sectors are clearly the same, the resulting policies and prescribed governance structures for WMAs and CBFM are a product of much more than these objectives. Stark differences in power devolution and governance systems have been identified in this chapter and discussed as resulting from a combination of political-economic, personal interest and discursive processes. The policy reform processes leading to the publication of the WPT (1998) and NFP (1998) incorporate the political- economic situation in which Tanzania was situated at the time, alongside its close relationship with TNCOs and the growth of development aid in the country, which contributed to the setting of the agenda for community conservation in national policy. As policy began to be formulated and discussed, political-economic factors relating to the politicisation of natural resources in both the forestry and wildlife sectors began to shape these policies, pulling them in different directions according to the interests of the actors within the policy community and wider policy network. The policy pathways adopted in both the wildlife and forestry sectors were clearly a result of the combination of these driving forces, which were played out discursively between different actors and groups within the policy network. The policy entrepreneur roles of the Directors of both the Wildlife Division and the Forestry and Beekeeping Division were crucial in the adoption of CBNRM strategies over competing discourses of community conservation within Tanzania. The adoption of a 167 wildlife utilisation strategy within the WMA governance system is a clear example of the discursive struggles that were taking place within the process of policy reform. I have emphasised the role of policy entrepreneurs, arguing that the Directors of Wildlife and of Forestry and Beekeeping, who fully supported the introduction of CBNRM were fundamental in the processes of policy reform. I also discussed the role of the local level, and particularly donor-funded pilot projects operating in the wildlife sector prior to policy formulation. This unexplored area of the policy network shows that in the case of the REWMP project, significant policy space was both constructed by actors at the local level, who positioned the project to become one of the leading examples of CBNRM and the new concept of WMAs, and facilitated through connections with other levels within the policy network. Through close personal relationships with the policy community and the PAWM project, the REWMP project and its policy space were legitimised through its integration into the dominant discourse coalition. The legacy of the REWMP/MBOMIPA project is tangible influences upon policy and the prescribed governance structure for WMAs. I contend, however, that the construction of policy space by such projects is not an entirely bottom-up process. Firstly, the REWMP case supports the suggestion that relationships between the policy community and the wider policy network in policy processes are hierarchical and ‘subject to power relations’ nature of (Marsh and Rhodes, 1992). The changing policy space experienced by the later MBOMIPA project with regards to participation in the formulation of WMA Guidelines clearly shows that the roles of the local level in policy processes are invited spaces that were be opened, closed and restricted by the policy community. Secondly, strategic positions within the policy network are part of processes of co-construction of policy narratives in which such projects can be utilised by the policy community in the discursive projects of policy development and implementation. Pilot projects can play an important part in shaping policy processes, but the REWMP example highlights that such projects can be shaped by the connections that enable them to do this. The role of the local level in policy processes should not be neglected, therefore, but its complexities are worthy of further consideration. The process of policy reform constitutes sub-processes of agenda-setting, policy construction and implementation (Keeley and Scoones, 2003). This chapter has dealt specifically with the first and second of these, and begun to outline the governance system prescribed in policy for both WMAs and CBFM in Tanzania. This is continued in the following two empirical 168 chapters, which further outline the governance and power systems prescribed in these policies (chapter 6), and their implementation, termed the performance of policy (chapters 6 and 7). Chapter 6: The Performance of Policy I: Power Systems in Governance and Politics of Scale 169 6.1 Introduction The previous chapter outlined the prescribed governance systems in place within Community-Based Natural Resource Management (CBNRM) projects in both the wildlife and forestry sectors in Tanzania. Through the processes of power devolution, the creation of formal institutional structures and their integration with informal institutions and local contexts of power (see 2.1.), configurations of power as set out in policy and as performed in reality are produced. According to the differences in these processes between the two sectors, as described in the previous chapter, these configurations of power are distinct between the wildlife and forestry cases studies. In the first section of this chapter, I outline these systems of power in both case study projects in further detail and discuss the scalar framings these involve. In section 6.2, I look into inequalities in the impacts of both Community-Based Forest Management (CBFM) and Wildlife Management Areas (WMAs) for different groups, addressing the issue of elite capture and how it is related to the prescribed governance system and the topographies of power it legitimises. In section 6.3, I use a scalar perspective to consider the framings that are utilised in both CBFM and WMA projects and I draw on the literature from politics of scales (see 2.4.2) to examine how the compartmentalisation of space around power relations produces scales within the projects (see Brenner, 2001). I investigate how these scalar topographies of power are under constant negotiation, struggle and manipulation (Smith, 1990; Smith, 1992; Marston, 2000; Zulu, 2009), and the discursive strategies used by actors in the struggle for empowerment and benefits from CBNRM (Swyngedouw and Heynen, 2003; see also 2.4.2.1 and 2.5.2; Campbell, 2007). Finally, in 6.4 I combine these power structures and scalar struggles together to outline the power structures in reality within both case studies 6.1.1 Devolution and Topographies of Power As discussed in chapter 5, the initiation of CBNRM in Tanzania is intrinsically connected to the idea of devolution of power away from the national, ministerial levels and towards the village level. Between the WMA and CBFM systems, the principal governing institutions and fulcrum of power have been devolved in distinct ways, and the resulting prescribed governance structures to implement CBNRM policy demonstrate different strategies to achieve this. The devolution of power within both sectors is signalled by the advisory role afforded to the ministerial levels in both forestry and wildlife and the daily management responsibilities in both CBFM and WMAs being placed in the hands of individuals elected 170 from within participating communities (see Tables 5.1 and 5.2). Within the forestry sector, the prescribed governance structure for CBFM demonstrates a rapid devolution of power to the village level and is implemented through existing governance structures. Contrastingly, the governance system in WMAs demonstrates a more cautious devolution of power, which is centred at a new institutional level created through WMAs, which sits between the local and district government levels (see 5.2.5). As pointed out to me in an interview with a former employee of a conservation Non-Governmental Organisation (NGO) in Tanzania (Interview P85), both systems represent a set of nested institutions, implemented not within traditional authority systems in Tanzania, which were gradually removed throughout the periods of colonialism and socialism in Tanzania (see Chapter 3), but within the bureaucratic national system of democratic government spanning the local, district, regional and national levels. This devolution of power from central to local is accompanied by further shifts in power resulting from the CBNRM policies. These are witnessed in the changing powers, status and responsibilities of groups within the local and district levels. Within CBFM, this is most clearly visible through the changing relationship between the Village Natural Resources Committee (VNRC) and the Village Council, whereby the implementation of CBFM empowers the VNRC and hands responsibilities for potentially large sums of revenue to this committee, which forms part of the village government and is overseen by the Village Council, although the Village Council holds supervisory powers only over revenue collection (see 5.5.1). An academic researcher in Tanzania described how this concentration of power within the VNRC, and its nested relationship within the village government, has been a source of conflict in many communities implementing CBFM to date, particularly once revenue generation from CBFM becomes significant, increasing realisation of value and politicisation of resources within the community (see 5.4), and sometimes leading to power struggles between the Village Council and VNRC “as both want to have a say in resource use” (Interview P7). This has led in some cases to “the Village Council take[ing] over control of the project because they want to collect the money themselves” (Interview P7). The interviewee also cited trust issues between the Village Council and the VNRC arising from a “lack of transparency and accountability. Both believe the other is up to something and cannot be trusted” (Interview P7). Within WMAs, the devolution of power to AAs has not only altered the relationship between the ministry and the participating communities, but as one former project officer described: “The creation of MBOMIPA marked a significant shift in local government, as power was 171 transferred away from the hunting groups” (Interview P73). These hunting groups, which formed part of the Hunting Association of Tanzania, and had previously used the Lunda- Mkwambi Game Control Area for hunting activities through the purchase of permits at the district offices, objected very vocally to the initiation of the restriction of resident hunting freedoms and the devolution of management rights to the community level within the Ruaha Ecosystem Wildlife Management Project (REWMP) and MBOMIPA project areas (Interviews P1, P74; see also Hartley, 1997). As discussed in 5.2, the devolution of power to the local level, and the shifts that this entails, are focused around distinct institutions between the wildlife and forestry policy sectors. The institutional structure of CBFM reflects the devolution of power to an existing institution, the VNRC, nested within the Village Council structure. The embedded nature of the VNRC is indicated in Fig. 6.1 and was described by village respondents in the CBFM research village, who emphasized that “the VNRC is beneath the village government, therefore there is nothing the committee does without the knowledge of the village government” (Interview P43). Committee members confirmed that the prescribed governance system meant they had to seek approval from the village government before using funds from the CBFM project, and if this permission was refused, they did not have the power to continue regardless (Interview P44). Conversely, the WMA governance system involves the injection of an entirely new institution into the management system, and simultaneously includes the creation of a new level, located across the participating villages. The roles and responsibilities of the Village Council and district authorities with regards to the area gazetted as a WMA are significantly altered by the creation of this new institution (see Table 5.2), and a parallel governance system in created. Furthermore, the institutional structure within the prescribed governance system of WMAs is more cumbersome, more top-heavy and more intricate in terms of the relationships between different individuals and institutions. The devolution of management responsibilities to the Authorised Association (AA) level is accompanied by a multifarious web of relationships between the AA and other organisations within the prescribed governance structure. The power structure within the WMA governance system can be described as simultaneously more radial and more top-down. The governance structure is clearly more top-down in terms of the retention of power and management responsibilities at the ministerial level, especially regarding hunting quotas and financial systems, and in the 172 advisory and supervisory powers invested in the District Natural Resources Advisory Board and the Board of Trustees (see 5.5). This could be described as an example of simultaneous decentralisation and recentralisation of power, a trend observed elsewhere in Africa (see Ribot et al., 2006), in which the devolution of power within such policies is limited wherever possible. This is apparent through the multiple ‘safety nets’ that surround WMAs’ governance system; for example, the revenue sharing arrangements discussed in 5.5.4 include the retention of powers at the regional and ministerial levels. Furthermore, the arrangements for signing of investment contracts between the AA and private investors include the selection of three favourable proposals by the AA, with advice provided by the Board of Trustees and potentially the NGO facilitator, which are then sent by the AA for inspection by both the District Natural Resources Advisory Board (DNRAB) and national levels (Wildlife Division), before contracts can be ratified (Interviews P86, P104, Focus Group P102). This is in direct contrast to the system in CBFM, whereby the VNRC hold the powers to allocate permits for resource harvesting (Interview P34, Focus Group P41). As discussed in section 1.1, recentralisation is an important part of the re-territorialisation of wildlife resources within a neoliberal form of statecraft (Igoe and Brockington, 2007). This re-territorialisation within a discourse of devolution takes place through the use of policy and power devolution as political technologies. As argued by Agrawal (2005), these technologies are a component of governmentality, producing new environmental localities and subjects within Tanzanian CBNRM, and represent the diffusion of state power over society (Hajer and Versteeg, 2005). The power system operating within WMAs is outlined in Fig. 6.2. 173 Fig. 6.1: Power structure prescribed in policy in CBFM in Tanzania Legend for Figs. 6.1 & 6.2 Village Governance level AA Authorised Association DNRAB District Natural Resource Advisory Board Ministry Ministry of Natural Resources and Tourism PMO-RALG Prime Minister’s Office of Regional and Local Government Regional Sec. Regional Secretariat VC Village Council VGS Village Game Scouts VNRC Village Natural Resources Committee WMA Wildlife Management Area 174 A comparison of Figs. 6.1 & 6.2 clearly shows that the system in place within WMAs is institutionally much heavier, but it also shows that there are several important relationships that are prescribed in policy between the AA and national levels. For example the AA holds direct relationships with the NGO facilitator and with investors (although this is balanced out as described above). The introduction of the AA consortium at the national level is also a further key relationship (see 6.3.2.2). This complex institutional structure within the prescribed governance system both demonstrates a lack of autonomy in decision-making and power, alongside the opening up of the new AA level to the district, regional and national levels. In terms of the bundles of rights (Agrawal and Ostrom, 2001) experienced by the village level within CBNRM, both WMAs and CBFM take place with clear village proprietorship of land and rights of withdrawal and exclusion, but the associated management rights are curtailed through the prescribed governance systems, and this takes place much Fig. 6.2: Power structure prescribed in policy in WMAs in Tanzania 175 more severely in the wildlife sector, where the rights of the AA to manage the WMA sit within a complex arrangement of safety nets and authorisations from government organisations. 6.1.2 Scalar Framings in Wildlife Management Areas and Community-Based Forest Management The prescribed governance systems set out in chapter 5 and Figs. 6.1 and 6.2 serve to produce scales within WMAs and CBFM in Tanzania. These scales are particular framings of reality that are discursively legitimised through policy processes of formulation and implementation (Marston, 2000). A discursive process of scale production that integrates the national, regional, district and village levels, plus the AA level in the wildlife sector, is evident, therefore. These scalar landscapes not only reify the prescribed governance systems and topographies of power that these involve, but serve to de-legitimise alternative framings (Hajer, 1995; Keil and Debbané, 2005; Lebel et al., 2005), levels and scales of wildlife and forestry resources and the areas of land which the WMA and Village Land Forest Reserve (VLFR) now occupy. These alternative and unrecognised scales and levels are distinct between the wildlife and forestry sectors and are described below with reference to the MBOMIPA case study. Within the participating villages of the MBOMIPA WMA, fieldwork revealed three additional levels within the scalar landscape. Firstly, the long history of pastoral use of the area, particularly in the northern sections of the WMA, and especially the transitional nature of these uses, the seasonal movement of pastoral groups throughout the year and the identity of these groups as outside of a particular village cannot be accommodated within the governance system for WMAs. In the current system, individual villages may set aside areas for pastoral uses, but grazing is forbidden within the designated WMA. Maasai groups within the WMA declared that their access to grazing sources had been drastically altered by the designation of the WMA and that they had insufficient sources to stay out of the WMA area, forcing them to break the law or watch their cattle starve (Focus Group P134). The marginalisation of pastoral groups within Tanzania is a long-recognised issue that extends beyond the issue of WMAs, having previously been highlighted with respect to National Parks, the Ngorongoro Conservation Area, Community-Based Conservation Projects and the Manyara Ranch (Goldman, 2003; Homewood et al., 2009; Nelson et al., 2009; Goldman, 2011). Pastoral claims and uses of land within the WMA have effectively been scaled out of the prescribed governance system. Secondly, within the governance system the AA sits 176 between district and village scales of governance, but does not recognise the division of the MBOMIPA villages into two administrative divisions (Idodi and Pawaga) which separate the villages administratively and geographically creating distinct spatial levels and scales. Finally the WMA governance scales mask divisions between the WMA villages between villages that have contributed village land to the area within the WMA, and those that act as boundary villages that have not contributed land (see Table 6.4). These examples of additional socio-ecological scales represent both pre-existing framings of the landscape that have been removed or not integrated into the prescribed governance system, and the construction of new scales as a result of the implementation of the WMA. These levels and scales are unrecognised within the governance system, but have become mobilising issues for scalar practices amongst the MBOMIPA villages (see 6.3.4; see Marston, 2000). 6.2 Benefit Distribution and Scalar Configurations of Power One of the clearest outcomes from the power structures operating in both case study projects was the identification of elite capture within the projects. This has been previously identified in other CBFM projects in Tanzania by Lund & Treue (2008), the MNRT (2008) and Vyamana (2009: 251), who described the inadequacies of CBFM in terms of creating opportunities for the poor to “build their own pathways out of poverty”. Leaders of the project, referring to those holding positions on the committees, and especially those holding the top positions of Chairman, Secretary and Treasurer, both in the VNRC and the AA were consistently identified as those that benefitted most from the project. This was described by local villagers within the CBFM village as resulting from several factors, including the ability of the leaders and the more wealthy groups within the community to afford to participate in the activities managed by the projects, whilst poorer groups were unable to raise the cash necessary to purchase permits (P57). Secondly, these individuals were identified as benefitting significantly from their roles on the committees through the allowances for their work: “If we had good work like them (VNRC committee), we’d build a nice house, live in a nice environment, isn’t that true?” (Focus Group P58). This situation was explained in focus groups with villagers using the saying “You can’t compare people who are in the kitchen with others who are waiting for food” (Focus Group P61) to indicate that those who are the “leaders must be at the centre of the activities, involved in everything, and will always take their share of the opportunities and benefits first” (Focus Group P59; see also 7.5). 177 The financial benefits available through representation on the committees are concentrated in the hands of the top leaders of the VNRC and the WMA Secretariat. Within the VNRC, these leaders are entitled to receive an allowance equivalent to 20% of the revenue derived from the forest and its management, which is shared between the three (Interview P35), and the financial records of the VNRC confirmed this. For the three assistant positions within the VNRC, financial allowances depend upon the work carried out, and usually only occur when one of the leaders is unable to complete a task (Interview P35, Participatory Technique P68). Within the revenue sharing system for MBOMIPA, the leaders and their assistants receive a set allowance per month, as detailed in Table 6.1. In addition, all members of the Council are eligible to claim an allowance to cover meeting attendance, but do not receive a fixed monthly amount. To put the amounts shown in Table 6.1 in context, an agricultural labourer receives a wage of around $2 per day, a full-time teacher’s annual salary is $828 and district government officers are paid around $4,956 annually. The leaders of the VNRC and AA carry out their duties in addition to their own employment and additional sources of income. The patrol and protection system within MBOMIPA has changed within the past year so that whereas previously ten scouts from each participating village worked on a rotation system for MBOMIPA, two scouts from each village are now employed by the AA and carry out official MBOMIPA patrols within the WMA area65, whilst the remaining scouts usually patrol up to the border of the WMA, although in cooperation with the MBOMIPA scouts, they sometimes patrol beyond the village areas into the MBOMIPA zone (Focus Groups P164, P165). The scouts are paid an allowance by the VNRC, as detailed in Table 6.1. In addition to the patrol allowance paid in the CBFM village, focus group discussions with the scouts revealed that they are paid an additional TSh. 1000 ($0.69) each time they escort people to the forest to monitor activities, and the same amount was often used as a reward if scouts brought offenders back to the village offices to face sanctions for breaking the rules (Focus Group P65). 65 Under this new system the MBOMIPA scouts work alternate months and receive an allowance of $62 for the months they patrol and $31 for the months they are resting (totalling $558 annually). 178 MBOMIPA WMA Kiwere CBFM 2008-9 2009-10 2008-9 2009-10 Total income ($) 90,665 110,000 2,219 2,582 Leaders’ Allowances ($) Chairman Secretary Treasurer Assistants Representatives’ travel allowance (maximum per meeting) 240 660 492 166 31 444 variable n/a 491 variable n/a Scouts’ Allowances ($ per patrol) 2 2 3.5 3.5 In addition to the financial benefits for those working closely with the management of natural resources through the committee structures, additional benefits were identified by the wider community as originating from the project. These included funds that were invested in community services, such as health clinics, improved housing for local teachers and investing in improving resources in local schools. In Kiwere this money constituted up to 25% of the annual budget, as set out in the management plan ($554 in 2008-9 and $613 in 2009-10), whereas participating MBOMIPA villages hold the rights to utilize the funds provided by the AA to each village as they see fit ($1,448 in 2008-9 and $1,517 in 2009-10). MBOMIPA also provides funding for two children from each participating village, who would otherwise not afford, it to complete secondary school education (P99, P101). Whilst the income received at the village level in MBOMIPA seems a much larger figure, the funds allocated to the villages as a proportion of the total income of the WMA in those years amounts to 34% and 29% respectively, and a significant proportion of those amounts was reported by the Village Council members, MBOMIPA Representatives and VNRC members as being used to pay the village scouts to carry out their patrols (Interviews P96, P106, Focus Group P164)66. Furthermore, the income received by the Village Council for development activities is shared between the entire population of the village, which was recorded at the last population census (2010) as 2,183 in Kiwere and 1,642 in Makifu. Revenue shared amongst the villages through development activities is comparatively little, and provides indirect benefits, whereas the leaders of the projects increase their personal financial income 66 VNRC accounts for this period were unfortunately incomplete and have not been included because of the large gaps in the data. Table 6.1: Monetary allowances derived from the case studies of CBFM and WMA 179 substantially. The disproportionate benefits of elected roles within the WMA and CBFM governance structures, and especially the leadership roles, are clear therefore. Lund & Treue’s (2008) study argued that these amounts were not inappropriate given the work that these individuals carry out. My work shows, however, that there are significant individual benefits for those in leadership roles, but also that these benefits are highly disproportionate to the rest of the village population, who benefit from CBFM or the WMA through investments and development activities that accrue at the village, rather than individual, level, as previously discussed by Blomley & Iddi (2009). The costs associated with CBNRM through protecting farmland close to the WMA and shifts in livelihood activities as a result of the initiation of CBFM (especially charcoal makers; see 6.3.3.1) are borne at the individual level, and village development activities do not compensate for these costs (Meshack et al., 2006). 6.2.1 Elite Capture through Prescribed Governance The capture of revenue and financial benefits by committee members described in the previous section is facilitated by the prescribed governance system, which sets out these allowances and also concentrates power and responsibility for the projects in the hands of these few, key individuals. This section explores how whilst legitimately taking place within the prescribed governance system, elite capture is also facilitated and maintained by the protection of these opportunities. Discussions with villagers committee members across the villages participating in the WMA and CBFM projects commonly included complaints that formal training had only been provided for the leaders of the projects and whilst wider awareness of the projects was high, understanding and engagement with it are low (e.g. Focus Group P41). Within the CBFM case study, the concentration of training within the leaders of the VNRC was corroborated by the leaders of the project who described the various training sessions they themselves had received from the Forestry and Beekeeping Division, district authorities and expert project advisors (Interviews P42, P46). They also insisted that there was an informal system for training newly elected members of the VNRC and all committee members had received training (Interview P46). The lack of ongoing formal training was explained as the result of lack of funds at the district level by both District and Village Officers (Interview P11, Local Assistant Report P20). The other members of the committee claimed not to have received adequate training however (Focus Group P41). The lack of ongoing training and focus of 180 formal training and skills to within a small group of individuals is important in two ways: firstly the same group of individuals is habitually returned to office. This was evident in both the WMA and CBFM case studies, and commonly explained by the leaders in question as due to their experience and expertise, which others in the community did not have (Interviews P35, P180); secondly this situation is advocated by both the district authorities and Village Council, who both do not wish, and claim to be unable, to bear the cost of providing extra training after each election (Local Assistant Report P20, Interview P88). This stagnation of the project with respect to the elected management committees is clearly identified in the WMA case study where, for example, since the first elections held in 2002, the current chairman previously occupied the position of Assistant Chairman and was elected in 2007 to serve as Chairman, whilst both the Secretary and Treasurer have been in place since 2002 (Interview P97). Within the CBFM research village the return of the same individuals to office within the VNRC is also facilitated by the prescribed governance system, which sets out that representatives on the VNRC are democratically elected at the sub-village level, and the positions within the committee are then settled at the first meeting of the elected representatives (Local Assistant Report P20). Whilst this system ensures that representation on the VNRC includes all the sub-villages, it also leaves each voter with rights only over the two representatives from their particular sub-village, and not the entire committee. Complaints about individuals on the VNRC were therefore voiced in anger over their continual return to the committee and their specific leadership positions (partly because of the training situation), which those outside of the individuals’ sub-village were powerless to address (Local Assistant Report P20). When asked to describe the process at the most recent election in May 2010, a VNRC member described how “about the Chairman and his Secretary: As usual, when the election took place, they came out as the representatives from their sub-villages, and when all the representatives met to select the leaders of the committee, they just got chosen again” (Interview P46). A villager from the CBFM research village showed the anger surrounding this topic in his description of the explanation given as to why certain individuals have held office for long periods of time: “other villagers don’t get a chance and the opportunity stays with those of long ago, and we are told this is because it is those people who have the knowledge and expertise of the project. The question to be asked is whether, when they first started, did they have this knowledge and experience? Didn’t they learn it through the project, and can’t new people also learn like them?” (Local Assistant 181 Report P22). The same villager went on to explain that he/she thought it was just an excuse to allow people to protect their positions in the project and the benefits they get from it (see 6.3.1). The distribution of benefits within the case study projects, and the identification of elite capture within both WMAs and CBFM are clearly linked to individuals’ positions within the governance structure, and elite capture is compounded by the concentration of power, knowledge and training with these same individuals, and further sustained through the electoral systems and lack of ongoing training. I argue that knowledge about the project is itself a form of benefit resulting from the project, a means by which to secure material benefits and exclude others from these, and also a form of power within the project. The power of knowledge enables certain individuals to inflict a form of institutional violence over others through the control of information and empowerment (see 6.3.1; Brockington, 2008). These processes produce a form of stagnation within the projects, which are also closely linked to cultural and hidden aspects of the governance system, as discussed in chapter 7. 6.3 Politics of scales So far in this chapter I have discussed the topographies of power that result from the prescribed governance structures in place within WMAs and CBFM in Tanzania. I have started to develop arguments that power, training and knowledge are concentrated within both types of CBNRM within the hands of a small group of individuals in each case. In this section, I want to expand upon these arguments to discuss how scalar configurations of power are both being reinforced by those in power and those benefiting from the projects (often the same individuals, as discussed in 6.2.1), and being contested by multiple actors and groups throughout the governance systems. These mechanisms of scalar construction and transformation are driven by the scalar configurations of power within the governance system, and can be seen as strategic attempts to gain control over both the resources managed by the projects and the projects themselves (Smith, 1992). 6.3.1 Discursive Construction and Reinforcement of Scales: The Protection of Power In their study of the river basin management in the Mekong region, Lebel et al., (2005) describe a process of discursive construction of scale through the control and manipulation of information, creating patterns of empowerment and disempowerment (see 2.4.2). Within both case study projects the concentration of training, experience and power within the hands 182 of a small number of individuals is a good example of this. Lebel et al., (2005) argue that this is a strategic action by those in power to control the frames of reality presented in scales through the discourses and narratives that surround them (Lebel et al., 2005; Lebel et al., 2008). I argue that the case study projects show such processes in action and add to such debates. Taking an example from the CBFM project, the communication systems in place between the VNRC and the village assembly consist of the public declaration of quarterly figures for income and expenditure related to the forest, and the display of these data in written form at the village natural resources office (Interview P34). A similar system is used in the WMA governance structure to report back to the village assembly about the income derived from the WMA within the village budget and its uses, and the activities of the AA (Interview P95). Such meetings are also used as an opportunity for villagers to make suggestions, voice concerns and ask questions relating to the projects (Interviews P34, P94). Discussions with villagers revealed several problems with these governance arrangements. Firstly, villagers argued that reading out information at three-monthly intervals was not frequent enough for them to really understand or keep track of the project activities, and that meetings were often postponed, did not take place at regular intervals and information was sometimes not read out (Focus Group P57, Local Assistant Report P22). There were also concerns about the level of detail included in the information, the opportunities for discrepancies to be hidden within the information read out, and a lack of awareness of the state of the natural resources within the forest and whether they are being used wisely: “We used to know that we had [for example] 50 trees ready to be harvested and this would be advertised...but now we don’t know how many trees we have and even if they are available, the leaders give them to each other” (Local Assistant Report P22). Villagers in the CBFM research village described their understanding and participation in the project as “trifling” as a result, and the contribution of the project to the development of the village as “insignificant” (Local Assistant Report P27). In discussion with village government officials and MBOMIPA representatives, some acknowledged that developing a good understanding of the projects and activities was difficult for ordinary villagers and justified the level of detail divulged to the village assembly as appropriate so as to not take up large amounts of their time: “we take the important and meaningful information to explain to people” (Interview P93). Furthermore, this process was additionally evident within the WMA case study, scaled to the level of the AA, where representatives complained that the 183 leaders of the project acted without the knowledge of the whole Secretariat (Interviews P172, P174), and in a focus group with officials from one participating village they complained that the Chairman in particular “believes he is the only one authorised to speak about the project” (Focus Group P133). In both case study projects information was shared on a ‘need to know’ basis and to the minimum acceptable level. The retention of knowledge and restriction of information by the officials of the project serves to reinforce the power structure in place through maintaining a clear margin between those working within the project and responsible for it, and the rest of the village community. The discursive construction and reinforcement of scalar topographies of power in both case study projects involves distinct patterns of empowerment and disempowerment. This was recognised by villagers in both research villages who commonly voiced their discontent and lack of understanding, blaming the project leaders for not educating them about the project or involving them in it: “The important people know about it, not the rest” (Interview P49); “I don’t have anything or way to participate. If they build a school, my children go, if they build a clinic, I can be treated, but that’s all” (Interview P152). Villagers in both the wildlife and forestry research villages described how this lack of understanding and information led to them not feeling able to make suggestions at the village assembly meetings, and the wishes of the leaders of the project nearly always being accepted as the best course of action, because they were the only people who really understood the project (Local Assistant Report P21, Interview P142). If we take knowledge to be inescapably linked to power (Barnes, 1988), the sharing of information and empowerment of the wider community threatens to destabilise the existing scalar topography of power (Haugaard, 1997), and controlling this information is a mechanism to protect those who hold power, office and rights to benefit from the CBFM and WMA systems. The control of information is a form of institutional violence that is part of a process of power, as described in 6.2.1. It is a clear example of power as a political technology in the social construction of subjects; that villagers felt the leaders of the project were the proper authorities possessing the appropriate knowledge signifies the process of internalisation of the socio-political order taking place within the research villages with regard to management of the forest and wildlife resources (see Lukes, 2005). Both natural resources and their management were being discursively produced and rendered governable through the construction of environmental objects that require ‘proper’ management, and environmental subjects who are equipped to do so. The control of information serves to reinforce the power 184 system in place and has also been internalised by the wider community, making it a key aspect in the reinforcement and construction of power (Feindt and Oels, 2005). These findings demonstrate the third, fourth, and fifth types of power described by Haugaard (1997; see 2.3.1): the internalisation of values that accepts the system described above is supported and maintained by systems of thought (3) that reify power structures, preventing destabilisation through the lack of perception of the work of power within the system (5). Importantly in this example, this takes place through the prevention of the fourth form of power (tacit knowledge) through the control of information that remains in the form of practical rather than discursive consciousness. There is also a clear link to the discussion of corruption and neopatrimonialism in chapter 7: Information has become a resource within the performed governance systems of CBFM and WMAs, and its exchange is subject to the relations of patronage operating within the projects (Moss, 1995). This is supported by Chabal & Daloz (1999) who discussed the control of information as a key component of the political instrumentalisation of disorder and the expansion of patron-client relations (see 7.4 and 7.5). The control of information and restriction of opportunities for the wider community to engage with the governance system surrounding CBFM and the WMA serves to reinforce the existing scalar organisations of power. This is achieved principally through the fifth form of power described by Haugaard (1997) in which these scales are discursively constructed as appropriate and legitimate, for example through paternalistic arguments that justify the simplification and selection of appropriate information to share with the village assembly, as described above. The restriction of training and experience to the leaders of the project and the stagnation of election processes which results from this (see 6.2.1), further serves to discursively create identities for a small group of individuals as trained, experienced and knowledgeable, controlling the frame of reality for the wider community to facilitate their return to office at elections, justifying the perceived pooling of benefits with this group, and explaining and sustaining the lack of wider participation and engagement with the management of the resources as apposite. Thus a system is in place whereby the leaders of the project legitimately benefit from their work to manage the natural resources under their jurisdiction and they are disproportionately empowered through CBFM and the WMA. The control of information-sharing is a political technology employed by the leaders to reinforce the power hegemony, and is legitimised by a discourse of knowledge and experience. 185 6.3.2 Scalar Transformations: The Struggle for Power The performance of policy and the power structures it sets out indicates that alongside the reinforcement of power systems and the protection of power that is described in 6.3.1, there are processes of scalar struggle and transformation taking place continuously within the case study projects. These conflicts are driven by socio-political desire to capture power and control over the projects, the benefits derived from them and the natural resources they manage. In this section, I outline the mechanisms by which power struggles are taking place in both the case study WMA and CBFM, and how individuals and groups both take advantage of existing and create new opportunities to manipulate the projects and scalar topographies of power. These pathways to scalar transformations draw on the ideas set out by Smith (1992: 74) that “the scale of struggle and the struggles over scale are two sides of the same coin”. I include examples of both scale jumping and scale bending (see 2.4.1) and characterise the scalar transformations as being either upward or downward. This direction follows the system of administrative levels operating within Tanzania, and is not intended to suggest that the scales within CBFM and WMAs follow a ladder analogy (see 2.4.1; Bulkeley, 2005). My focus is therefore on the verticality and politics of relations between scales and how they are being transformed by actors (Bulkeley, 2005; Leitner and Miller, 2007; Rangan and Kull, 2009). Scalar transformations were witnessed in the ways in which actors and groups positioned themselves around the existing scalar landscape to both access and alter scales. This can take place through both scale jumping and scale bending (see 2.4.2), but is clearly distinguished in the comparison of the case study projects by the different directions of scaling this involves around the distinct topographies of power in place. A comparison of the politics of scales in Tanzanian CBFM and WMA governance indicates that the politics driving such conflicts are essentially the same, although the prescribed governance structures, and resulting power topographies these create, shape the dominant drivers and opportunities for scalar transformations, producing contrasting scales of struggle and directions of transformations (see 6.3.3). 6.3.2.1 CBFM: Scaling Down Investigation of the governance system within CBFM revealed that the dominant transformations taking place were examples of scaling down to lower levels. Interviews with international conservation practitioners and university researchers revealed that the concentration of power and management responsibilities in the VNRC (village level) has 186 stimulated several attempts by district officials to re-scale their role to within this level, and to take advantage of opportunities to maximize district-level benefits from the projects. The case study of CBFM in Kiwere village revealed the scaling down of election processes with respect to the district’s role, alongside scaling down of accountability mechanisms by the Village Council. Previous studies of CBFM in Tanzania (e.g. Mustalahti, 2007; Ministry of Natural Resources and Tourism, 2009) have identified similar processes at work, particularly with reference to the ways in which the prescribed governance system, and especially revenue systems, are altered in practice (see also 7.1). I adopt a scalar perspective to these issues, and address the struggle for power that lies at the heart of these transformations and deviations from policy. The first of these transformations involves the process of scale bending, whereby district level actors have created a role for themselves in the election processes for the VNRC at the village level. This is closely linked to the discursive construction of scale and protection of power within the hands of a few individuals already on the committee, and the obstacles to new officials being elected (see 6.2.1). Both VNRC officials and local villagers described the encroachment of district officials into electoral processes and decisions around the VNRC; they described how the limited number of trained and experienced individuals within the village led to advice from the district to ensure that these people remained in their positions within the VNRC (Interview P43). The scaling of district-level influence into VNRC electoral processes moved beyond this advice in some instances however, and was described by community members as preventing free elections from taking place because the district officials insisted that at least one quarter of the experienced committee members must remain on the committee, and when previous election results included an entirely new committee, the district refused to allow this (Interview P43, Focus Groups P61, P62). This scalar transformation is discursively legitimized by the district officials and those within the VNRC on the basis of existing committee members, and especially the leaders of the project, possessing the required training, knowledge and expertise to manage the project successfully, which, as discussed in section 6.3.1, partly results from the reinforcement of this situation by these same individuals. Thus a closed circle is created whereby the leaders of the project possess the official training and approval by the district to take up leadership positions within the CBFM project, which is then protected through discursive construction of the VNRC as the appropriate level and institution for successful management of the project, and reinforced through the concentration 187 of power within the level of the VNRC and limitations placed on the empowerment of the wider community. The interference of the district officials into the VNRC election processes benefits both the district, by reducing the need for ongoing training and the existing VNRC members, who remain in post, empowered and able to benefit from the project more easily. Community members also suggested that the interference of the District Officials, the creation of this closed circle and the close relationships between the project leaders and the District Officials potentially reflected the scaling down of corruption networks (see 7.4; Véron et al., 2006). In a focus group discussion, one community member suggested that the VNRC and district officials might be working together for mutual benefit from pre-existing arrangements for private benefits from the exploitation of forest resources (Focus Group P61). A second process of scalar transformation is taking place within the CBFM project with respect to governance arrangements and the roles and responsibilities of the VNRC and the Village Council. In an attempt to foster accountability and transparency, the Village Council in Kiwere has created a supervisory role for itself within the activities of the VNRC. This is a good example of scale jumping by which the Village Council has secured an invitation for one of its members from the social services committee of the Village Council to participate in all formal VNRC meetings and report back to the Village Council (P43). The Village Council has also carried out patrols within the VLFR to verify the findings of the forest scouts and reports and activities of the VNRC (Local Assistant Report P17). 6.3.2.2 WMA: Scaling Up The scalar transformations taking place within the WMA case study were predominantly examples of upward scaling. These were evident in instances of village governments re- scaling investment opportunities through jumping up to regional and national levels by entering into agreements with tourist operators, and scale bending through the creation of a national consortium of AAs to build an increasing voice for these institutions at the national level. There were also examples of conservation NGOs enacting scale bending to increase their role in WMA processes. Within the WMA governance system non-consumptive investment contracts are drawn up between the AA and the private investor, although as discussed in 5.2.4 and 6.1.1, this is a more complex process that involves checks by several other institutions at the district and national levels before final contracts are signed and top-down revenue-sharing arrangements. 188 Whilst the problems with this in terms of the restriction placed upon power devolution has been discussed earlier (see 6.1.1), another result of the prescribed governance arrangements over this issue is the marginalisation of Village Councils from participating villages, who sit in an uneasy position with regards to authority and hierarchy with the AA. There is no official role or space for Village Council officials to participate in decision-making processes regarding investment contracts (see 5.5.3), and the benefits they receive from MBOMIPA are handed down to the village level from the AA (whilst at the same time the Village Officers clearly saw themselves as above the AA, which is made up of elected representatives from each village; Focus Group P129, Interview P178). This shift in power relations had clearly caused tension between the AA and the Village Officers, with one village chairman describing how the village must have the final say, not MBOMIPA, and if they disagree with the decisions taken by the AA they would “first change the MBOMIPA representatives from the village, then break the AA committees, then go after the chairman. We would not wait for the elections, we would chase them away” (Focus Group P129). In an attempt to re-scale power and benefits from tourist investments around the WMA to the village level, rather than that of the AA, several villages have begun to set up investment agreements with tourist operators to provide accommodation facilities and game viewing opportunities outside the WMA, on village land. This is taking place within the villages that are located in areas of potential development for tourist activities nearby the WMA, particularly the villages of Tungamalenga, Mahuninga and Makifu, which directly border the WMA lands and are in close proximity to Ruaha National Park and the infrastructure surrounding it. The village of Tungamalenga has developed sites for tourist investment over a long period, hosting hotels and campsites on village land (numbering 12 at the time of fieldwork) and tourist attractions, including a snake park. Other villages, including the research village have since begun to identify the potential for tourism investments, which in Makifu amounted to negotiations between the Village Council leaders and five investors (all operating at the regional level), although contracts were still being drawn up at the time of fieldwork (Village Meeting P186, Interview P108). Such camps would be located along the boundary of the WMA area within Makifu village, on an area of land that has been designated a conservation area by the village. This area comprises a strip of land beyond the mountains that separate the domestic and agricultural areas of the village from the WMA (Focus group P160). Discussions with Village Council members from several villages indicated that they were strategically re-scaling investments in order to by-pass the prescribed 189 procedures and systems for tourism investment within the WMA, and to exploit the opportunity to increase village revenues directly, rather than receiving a share of new investments through the WMA revenue sharing arrangements (Interviews P176, P180). One MBOMIPA representative discussed this scalar transformation as the cause of conflicts within the AA and between the villages: “People are trying to benefit privately by setting up tourist enterprises, but village investors cause much conflict because people disagree over whether the money should go to the village or to MBOMIPA” (Interview P180). Within the village of Makifu, discussions about these potential investments revealed that they were actually examples of scale jumping in both directions, as members of the Village Council explained that they had both been working to develop these opportunities, copying the model already set out in the neighbouring village of Tungamalenga (scaling up) but in some cases had been initiated by the investors themselves (scaling down), searching for potential sites outside of the WMA to develop their activities at lower costs (Interview P108). Both processes reflect the re-organisation of activities around the existing power structure in attempts to access and re-configure flows of power and revenue around the project and the resources that it manages. The discursive construction of scale is taking place within the wildlife sector through the initiation of a new institution called the National Consortium of Authorised Associations, which began in 2010. The role of this institution, which is formed from representatives of the AAs around the country is to act as a forum for the unification of processes and systems in WMAs across the country, but also to provide the platform for AAs to increase their voice at the national level and their role in governance processes scaled to this level (Interview P80). The consortium was described by the chairman of one WMA as “the equivalent of the National Park Authority for WMAs in Tanzania…which has the power to speak to government and the ministry” (Interview P100). The officially-stated objectives of the consortium include “to liaison (sic) with government, non-governmental organization and/or private sectors in all matter that affect the consortium positively or negatively. These include development policies, legislations, guidelines, community development and distribution of land and natural resources” (Government of the United Republic of Tanzania et al., 2011). This re-organisation of WMA governance scales is potentially also driven by the desire to access and influence the decision-making processes concerning wildlife hunting quotas within hunting blocks inside WMAs and investment contracts for tourist hunting. This is indicated in the expected benefits of the consortium, which include “secured markets for 190 different tourism products in the WMAs…representing AA members in decision making during wildlife quota allocation…having favourable environment (sic) for investment in WMAs” (Government of the United Republic of Tanzania et al., 2011) The consortium has very strong links with external conservation organizations, particularly with funding from USAID and facilitation provided by Worldwide Fund for Nature (WWF; Government of the United Republic of Tanzania et al., 2011). This indicates another process of re-scaling by which WWF, in particular, is enacting scale bending by positioning itself as a key actor within WMA governance. An academic researcher described WWF’s rising status within the WMA governance system through its current role as the external facilitator for at least four of the largest, most successful and high profile WMAs in Tanzania (MBOMIPA, JUKUMU, Wami-Mbiki and Burunge), and stating that he believed WWF were attempting to expand their role as much as possible (Interview P80). 6.3.3 Politics of Scales: Same Politics, Different Scales The examples of scalar reconfiguration ongoing within both case study projects indicate that these manipulations of the prescribed governance structure are centred around power. Drawing on debates in the scale literature about the non-hierarchical nature of scales (Marston et al., 2005; Leitner and Miller, 2007), these examples clearly show that scalar struggle is not simply a case of individuals and groups consistently attempting to scale up or jump or bend scales in an upward direction. Rather, re-scaling takes place around the existing power structures as different individuals and groups attempt to access these scales and construct new scales around them. In the case of the CBFM case study, where power devolution to the lowest level of village government is prescribed, re-scaling is taking place around the VNRC with the district officials attempting to bend CBFM scales to create a larger role for themselves within the governance system, and maintain opportunities to benefit from the management of forest resources, and village government officials attempting to jump down into the VNRC scale. Conversely in the WMA case study, where power is devolved principally to the AA, operating at the inter-village level, but issues of the retention of power at the national level remain (see 5.5), re-scaling is taking place both around the inter-village and the ministerial levels. This is evident in attempts to both allow Village Councils and investors to circumvent the prescribed systems through creating a new scale of investments at the village level, and in attempts to contest the power of the national level through the creation of a National Consortium for AAs. 191 The politics of scales taking place within the Kiwere CBFM and MBOMIPA WMA case studies support the arguments of Purcell & Brown (2005) and Smith’s (1990) definition of politics of scales, which both emphasise the socio-spatial strategies employed by actors for their own political, and in this case environmental agendas. The idea that power is at the heart of these processes is shown in the scalar transformations and constructions taking place around the existing topographies of power within both sectors, producing the same underlying political processes but scaled to different levels within the governance systems of CBFM and WMAs. I argue that a process-based view of scale (Swyngedouw, 1997; Swyngedouw and Heynen, 2003), as utilised in this discussion, is necessary to capture the characteristics of scalar dynamics taking place within the case study CBFM and WMA, which include both transformation and construction. Swyngedouw & Heynen (2003) also argue that conflict and political struggle are at the heart of these scalar processes, which is the focus of the next section. These conflicts also indicate that whilst they are taking place at different scales, the politics driving them is essentially the same: a struggle for power over CBFM or the WMA, the funds they generate and the natural resources they govern. 6.3.4 Scalar Transformations: Conflicts and Power Struggles “Politics is about the formal and informal contests and negotiations of power in, or over, various circumstances and how and what power and decisions might be shared or not” (Brunckhorst, 2010: 17). Alongside the scalar transformations taking place as described in section 6.3.2, there are several conflicts ongoing within the case study projects that reflect struggles for power over the projects and the resources they control. As discussed throughout section 6.3, these struggles take place around the prescribed governance structure and the power systems produced from this. An important feature of these conflicts is the level at which they are focused, therefore; following the concentration of power within the governance system of the WMA in the AA, a major set of conflicts is correspondingly scaled to this level; in the CBFM project, principal conflicts are taking place between different groups within the village, scaled to the same level as the concentration of power in the hands of the VNRC members. These conflicts, which indicate the processes of discursive construction of scale ongoing within the case study projects, are discussed separately because, whilst the processes are comparable, the nature of the conflicts is distinct. 192 6.3.4.1 CBFM Conflicts: Resource User Groups One of the main conflicts taking place in the CBFM research village involved the different resource user groups that utilize the forest to harvest different products. The main uses of the forest are for harvesting timber, firewood for charcoal production and tobacco curing, poles and stones for building uses, alongside numerous non-timber forest product uses of the forest, particularly grazing of livestock, beekeeping and medicinal plant and mushroom collection (Interview P4). Within these uses, revenue is generated from the sale of permits, and collected by the VNRC (Participatory Activity P66). The collection of firewood within the forest for domestic uses67 is permitted under the CBFM management plan, but interviews with villagers, and especially women who usually carry out this task, generally takes place in an area outside of the CBFM project known as the ‘open area68’, which is located closer to village residences (Interview P47). The permits allocated for different uses within the CBFM research village for the year 2010 are shown in Table 6.2 and Fig. 6.3. Month Permit type Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Total Dry wood 63 72 100 98 127 48 145 137 155 101 127 66 1239 Fresh wood 221 17 83 47 43 411 Charcoal 2 3 10 1 17 62 28 3 3 129 Visitors 24 10 14 14 62 Fodder 143 143 Building poles 5 5 Timber 190 190 Total 63 72 124 331 130 89 160 237 407 177 130 259 2179 67 Differentiated from collection of firewood for commercial purposes through the prohibition of cutting tools (machete or axe) and restriction of amounts collected to that which can be carried on foot (P4, P38, P58). 68 ‘Mahitaji wazi’. Table 6.2: Kiwere CBFM Income Derived From Permits Issued in 2010 ($US) 193 The issuing of permits for harvesting of firewood for both tobacco curing and for the production of charcoal purposes puts two resource user groups in direct competition for resources within the village. This conflict has escalated in recent years and, in 2010, resulted in the decision to cease allocation of permits for the purposes of charcoal production within the forest (Interview P43). Interviews with tobacco farmers, charcoal producers, the Village Council and the VNRC all confirmed this decision, and the topic generated lively debate, indicating the unresolved tension surrounding this conflict. The decision to cease approval of permits for charcoal production was explained in several different ways by different groups, indicating the discursive construction of scale around this issue. Charcoal producers explained the decision as one typical of the project which “harm[s] the charcoal producers” by forcing them to stop harvesting in the forest, firstly because they struggled to raise the cash to purchase the necessary permits, and then by removing their rights to use the forest for Fig. 6.3: Graph showing Kiwere CBFM Income Derived From Permits Issued in 2010 ($US) 194 this purpose altogether (Local Assistant Report P13). A focus group discussion with charcoal producers from the research village also showed that they feel unable to participate in the project following the decision, and they described the project as “negligent” (Focus Group P59). The Village Council members explained the exclusion of charcoal producers as a business-driven decision, because their permit requests put them in direct competition with those that could be allocated for firewood collection for the purpose of tobacco curing (Interview P43): “we had already set aside a large area for the tobacco farmers and if we issue permits for the charcoal producers, this would mean the forest would be over-used”. On the contrary, the VNRC members justified the decision to stop charcoal production in the forest and the collection of firewood for this purpose on the basis that the charcoal producers were constantly breaking the rules of the project and they could not be trusted (Interview P43). They invoked a narrative of environmental destruction and the degradation caused by charcoal production in particular, citing it as one of the main reasons why the project was started in the village and then describing the previous degradation and indiscriminate harvesting carried out prior to the project: “the charcoal makers used to empty a whole area of forest” (Focus Group P60, also Interviews P44, P50). The use of the narrative by the VNRC in particular demonstrates the discursive construction of scale with respect to the legitimization of the project as environmentally necessary, the restriction of charcoal production in the forest as a legitimate rule with corresponding sanctions within the project and the VNRC as the appropriate management institution to implement these rules. This discursive scalar struggle is also played out within a particular socio-economic context that had allowed the tobacco farmers to exclude their charcoal-making competitors from the project. There are two elements to this context: the business interests of the members of the VNRC; and the socio-economic hierarchy associated with different livelihood options in rural Tanzania. The first of these elements refers to the fact that the majority of those elected to serve on the VNRC are closely connected and personally involved with tobacco farming in the village, whilst none of them proclaimed to be involved in charcoal production. Of the ten members of the VNRC in Kiwere, six described themselves as tobacco farmers, including the three leaders of the project (Interviews P39, P34, P35, Focus Group P195), which suggests a very strong incentive for the VNRC to prioritise harvesting for tobacco-farming activities, over those of charcoal-making. Tobacco farmers within the community described their fortune at having a tobacco-farmer as the chairman of the project and the benefits they received from the project as including a good and secure source of firewood for curing 195 tobacco, which is organized for them through the chairman of the project and delivered to them as requested (Focus Group P60). Secondly, the discursive construction of charcoal production as environmentally destructive and the producers as untrustworthy and habitual rule-breakers was facilitated by the high socio-economic and social status disparities between the two groups; tobacco farming is generally a livelihood activity of the richer members of a community as it requires large areas of land to cultivate the tobacco plants, special structures for curing the tobacco, and sufficient income to invest in an agricultural system that involves a lengthy growing season (usually 6 months, depending upon rainfall), plus additional processing time (Focus Groups P60, P61, Participatory Activity P70). In contrast charcoal production is an activity associated with less educated and poorer groups within the community: “villagers who have lower incomes occupy themselves with making charcoal to get money to meet their daily needs” (Local Assistant Report P13). It is described as feasible for poorer groups in the community because of the low costs associated with the necessary equipment and the relatively short periods of time required to build and prepare a kiln, cut the firewood and produce the charcoal (minimum of five days, increasing with kiln size; Participatory Activity P69, Local Assistant Report P13). Charcoal producers within the research village described it as a ‘hand to mouth’ livelihood strategy adopted by those struggling to meet the needs of their family, and described themselves as victims of the project and its leaders’ decisions: “the leaders have been refusing to let villagers produce charcoal while at the same time they are using the area to improve their own lives” (Local Assistant Reports P13, P21). The conflict between these two groups involved the discursive construction of identities of charcoal makers as both unethical destructors of the environment and simple, poor people trying to feed their families. The different discursive tactics employed by the two sides of the conflict demonstrate the political nature of this conflict and the tactical use of discourse. The discourse of victimisation, impoverishment and necessity to support their families employed by the charcoal producers indicates an attempt to justify the continuation of charcoal production within the forest, and villagers reported that it is commonly used as a defence by charcoal producers when challenged (Interview P57). The VNRC and tobacco farmers on the other hand employed a discourse of environmental protection and rule-breaking by charcoal producers to legitimise their marginalization within the project, gain support from the wider community for these actions, whilst simultaneously securing resources for their own uses: “the charcoal makers could not be trusted, illegal use was rife” (Interview P34). I argue that 196 the current outcome of the conflict to prohibit charcoal production since 2010 reflects the private interests of the decision-makers, alongside the different socio-economic status of the two groups, which facilitates the prevalence of the negative narratives surrounding charcoal production. The strong hierarchy within Tanzanian society and the emphasis of respect for well-educated and wealthy groups within society, alongside the social norms surrounding the contestation of power (see 7.5) all contribute to the reinforcement of this situation and the dominance of the tobacco-favouring committee over the needs and wishes of charcoal- making groups within the community. 6.3.4.2 WMA Conflicts: Inter-Village Politics and Struggles for Revenue The conflicts taking place within MBOMIPA indicate the same underlying drivers of power and control over the WMA, its funds and its resources. The conflicts are scaled to the level of the AA however, following the concentration of power at this level in the prescribed governance system. The conflicts dominating the project at the time of fieldwork involved two inter-linked themes: the relationship between the two administrative divisions the participating villages are located in; and the distribution of benefits and problems from the MBOMIPA project. Of the 21 villages participating in MBOMIPA, all are located within the district of Iringa Rural, although the district is sub-divided into divisions, which splits the villages involved wherein 12 of the MBOMIPA villages are located in Pawaga division, whilst the remaining nine are located in Idodi division (see Fig. 6.4). This split has some interesting impacts on the relations between the villages and between the MBOMIPA representatives within the AA. Interviews carried out with the MBOMIPA representatives from both villages in Pawaga and Idodi revealed tensions and “problems of not understanding one another” (Interview P107) between the two sets of villages, which were described as each group of MBOMIPA representatives trying to dominate the power and decision-making rights of the AA (Interview P107, Focus Group P127). A focus group with village authorities and MBOMIPA representatives within Idodi revealed a fear of the Pawaga villages dominating the AA and decision-making because “they have more villages involved, which means more representatives, more voice in the project and more power to make decisions” (Focus Group P127). Village authorities and MBOMIPA representatives from some villages in Pawaga corroborated these ideas and argued that Pawaga was more influential within the project (Focus Group P133). Conversely, focus groups in other villages in Pawaga revealed conflict because they perceived that “the villages of Idodi see themselves as above those of Pawaga. 197 This is because they have contributed land, which means they have invested more in the association [WMA] than the Pawaga villages. Those that think themselves above are Kitisi, Mahuninga and Tungamalenga” (Interview P171). A respondent in the focus group also described how power was being concentrated in Idodi because all of the top leadership positions within the AA were all held by representatives from Idodi, whilst all of their assistants came from Pawaga villages (Interview P171; see Table 6.3). Thus a complex set of power relations has developed between the two groups of villages, and is exacerbated by the poor infrastructure between the two divisions, the remoteness of Pawaga compared to Idodi and the perception of Pawaga as poorer and less-developed than Idodi, where the villages benefit greatly from higher rainfall and income from Ruaha National Park (which is accessed through Idodi) and associated tourist activities. 198 Fig. 6.4: Map of MBOMIPA Participating Villages and Land within the Wildlife Management Area 199 The conflict between the administrative divisions is clearly a struggle over power and position within the project, but is also related to a recent conflict that developed within the project concerning the distribution of benefits and problems from MBOMIPA between the participating villages. This conflict centres on the contribution of village land to the area that is managed as a WMA. Table 6.4 compares the area of land contributed by each village (and division) to the MBOMIPA WMA, and highlights that whilst six of the Pawaga villages (notably Mboliboli) have contributed areas of village land to the WMA, six have not. Conversely, within Idodi, eight out of nine of the villages have contributed land, and five out of the six largest areas of land contributed to the WMA come from villages within Idodi. Within the current governance system of the WMA, all villages receive an equal share of the distributed revenue from the AA, as set out in the MBOMIPA constitution (MBOMIPA, 2002). Severe conflict has been created around this issue with a very vocal group of AA Committee Position Officer’s Village Administrative Division Leadership Positions MBOMIPA Chairman MBOMIPA Secretary MBOMIPA Treasurer Assistant Chairman Assistant Secretary Assistant Treasurer Tungamalenga Tungamalenga Idodi Ilolo Mpya Magozi Luganga Idodi Idodi Idodi Pawaga Pawaga Pawaga Protection Committee Chairman Secretary Representative Representative Representative Representative Kimande Nyamahana Kinyika Malinzanga Isele Kimande Pawaga Idodi Pawaga Idodi Pawaga Pawaga Finance Committee Chairman Secretary Representative Representative Representative Nyamahana Kinyika Mahuninga Itunundu Magombwe Idodi Pawaga Idodi Pawaga Pawaga Ethics Committee Chairman Secretary Representative Representative Representative Mkombilenga Makifu Mafuluto Isele Makifu Pawaga Idodi Idodi Pawaga Idodi Table 6.3: MBOMIPA Authorised Association Committee Structure and Representatives’ Villages. The remaining 20 representatives not holding a committee position are ordinary members of the AA. 200 MBOMIPA representatives and associated village authorities arguing that it is unfair that benefits are divided equally when some villages have contributed no land at all to the WMA. This produces arguments that such villages are “eating for free” (Interview P93). These groups, comprising most, but not all, of the village authorities and MBOMIPA representatives from the ‘landed villages’ also invoke a discourse of human-wildlife conflict to justify their arguments for altering the distribution of benefits. They claim that crop- damage impacts from wild animals that move beyond the boundaries of the park are concentrated in the villages that are located closest to the park, and these villages also form the group of landed villages. Idodi Division Pawaga Division Landed Tungamalenga (97.95) Makifu (77.39) Mahuninga (111) Idodi (50.37) Malinzanga (164.81) Mafuluto (81.31) Kitisi (40.31) Mapogoro (34.64) Mboliboli (71.83) Kisanga (13.9) Kinyika (4.16) Isele (17.67) Magombwe (43.67) Luganga (48.48) Total Land Contributed 657.48 199.71 Landless Nyamahana Itunundu Kimande Mkombilenga Magozi Mbuyuni Ilolo Mpya The geographical location of landed villages versus ‘landless villages’ is closely linked with their proximity to the national park, which arose from the design of the WMA, whereby the land within the WMA (gazetted from the previous Lunda-Mkwambi South Game Control Area) was selected to form a buffer around the national park (see Fig. 6.4). This is used by MBOMIPA representatives and village authorities within the landed villages to argue that “here [landed villages] we are hurt by conservation, but they [landless villages] only know the importance of conservation, but we are all paid the same” (Interview P100). A District Official also described how this situation produced a large discrepancy between the participating villages in the benefits received from MBOMIPA (Interview P88). Those who Table 6.4: Axes of Power within MBOMIPA WMA. Amounts shown correspond to the area of land contributed (km2). 201 oppose any alteration to the distribution of benefits within the WMA villages utilised a discourse of biodiversity conservation to challenge these arguments. A Village Officer from a landless village within the WMA described how “[the village’s] participation is valid because animals do not have borders…it needs to be seen from a conservation point of view” (Focus Group P123). Such political use of environmental discourses in the context of rivalries and competition has previously been noted in Tanzania by Brockington (2006). Other landless villages’ MBOMIPA representatives defended their inclusion in the MBOMIPA project on protection grounds, arguing that the villages further away were included so that they could form a cohesive force against hunting within the national park, which justified their inclusion and the equal share of revenue they receive (Interview P178). This group of MBOMIPA representatives and Village Authorities also attempted to discredit the arguments in favour of changing the revenue system, accusing the actors from those villages as “only being interested in money, not conservation” and attempting to “throw out other villages so that their income increases” (Interview P123, Focus Group P178). These two conflicts are inter-related as they feed into one another, escalating tensions and bad-feeling between the representatives. The consequences of these two conflicts became apparent during the period of fieldwork when seven villages wrote officially to the leaders of the AA to complain that the situation was unfair and threatening to remove themselves from the association if the revenue system was not altered in favour of the landed villages (Interview P100). One MBOMIPA representative from Pawaga described how this situation had reached a point where he thought that “MBOMIPA appears to be closer to the villages of Idodi than those of Pawaga” and the effects of this relate not just to the debates about revenue sharing but also “the representatives from Pawaga think that the Chairman is preferential to the Idodi villages and he also gives them priority for many giving [donor- funded] projects” (Interview P171). The complex conflicts taking place within the MBOMIPA villages have produced several axes of power that broadly divide the villages according to administrative division and land contribution, as shown in Table 6.4. It is important to note that not all Village Authorities and MBOMIPA representatives from landed villages within Idodi support the proposal to alter the revenue distribution system, or vice versa. Indeed within the research village there was high differentiation between the members of the Village Council and the MBOMIPA representatives with regards to this issue. One individual complained that villages with little or no land had been included to help prevent poaching and hunting from taking place within 202 the WMA “but now we see that they are eating for free. Their crops are not destroyed, they get no problems from the wild animals, yet they get the same proportion of the funds” (P108). Another individual from the research village was completely against changing the revenue distribution system in favour of the landed villages. The interviewee argued that “those who have little [land] are doing their work just as those that have lots of land, will you withold from them so that they are given little?..We share together because we all have one job to protect our land so that it can be sustainable, and even those who don’t have [land] protect it. It’s something that all of us should receive equally therefore, because the work we do is the same” and demonstrated her point using the following story (Interview P109): “It might be that in your house you have five children, and you are the father. There’s one child who grows up to do eveything that you ask of him. There’s another who just wants to sleep and he doesn’t even want to work. Another from the time they wake in the morning is occupied with cleaning, another is at school and the other works to bring home the things you need at home. Now with the one who doesn’t want to work, now when it comes time to eat do you deprive him of food whilst they are all still your children? He must eat.69” Table 6.4 clearly indicates not only that the majority of landed villages are located within Idodi division, but also that these have contributed the overwhelming majority of the land area inside the WMA. These villages have become more powerful within the AA, assisted by the fact that these villages hold over a third of the committee positions within the AA (8/22), including the top three leadership positions (see Table 6.3). Both the conflicts between the divisions of Idodi and Pawaga with regards to representation on the AA committees, and the debates over revenue distribution between the MBOMIPA villages, are good examples of what Fraser (2010) terms ‘scalecraft’ (2.4.2.1). The example described from MBOMIPA indicates several facets of Fraser’s concept: firstly that the discursive construction of scale around the issue of land contribution is a purposeful attempt to “diminish the influence of actors at other geographical scales” (Fraser, 2010: 336), built 69 “Unaweza ukawa ndani ya nyumba una watoto watano na wewe ndio baba, yupo mtoto anayelima anafanya kazi kila unapomtuma anafanya, yupo mwingine yeye ni kulala tu kabisa yeye hataki hata kazi, yupo ambaye yeye akiamka asubuhi ni kushughulika na usafi, yupo mwingine anaenda shuleni, yupo mwingine kazi yake ni kutafuta vitu vya muhimu vya kuleta pale nyumbani. Sasa katika yule mwingine hataki kazi kabisa sasa je, ukifika muda wa chakula huyo asiyefanya kazi utamnyima wakati ni watoto wako wote? Ni lazima ale.” 203 upon narratives of human-wildlife impact and justifications of ‘fair share of the work’ that strategically empower one set of actors over another and has material consequences for both. Secondly, the MBOMIPA example of inter-village conflict is an excellent example of what Fraser (2010) terms ‘active destruction’, in which the disassembling of scales and reconfiguration around a conflict could be seen as an important form of scalar ‘craft’. This is supported not only by the scalar struggles taking place around this conflict, but by the recognition that specific individuals were largely responsible for starting such a conflict and for deliberately dividing the villages within the WMA around such issues for their own benefit. Respondents from Village Authorities and other MBOMIPA representatives argued that the chairman of the AA was to blame for starting this conflict, as he wanted to gain as much for his village and himself as possible (Focus Group P133). As Fraser (2010) points out, the true aim of many such scalecraft practices may remain hidden beneath neutral justifications, as seen in the reasoning provided to support an altered revenue system. Occasionally, however, as Scott (1990) argues, such hidden agendas, thoughts or ‘transcripts’ may be voiced. For example, a member of the village authorities from one MBOMIPA village explained the tension between the villages of Idodi and Pawaga as “due to business: every person wants to appear to be the big man, so that he may have the power for himself” (Interview P107). 6.4 The Performance of Policy: Power in Reality The scalar transformations taking place within both the case study CBFM and WMA indicate a combination of both reorganisation of existing scales, including instances of both scale jumping and bending, and scalar construction. In line with the discussions in section 6.3, Figs 6.5 and 6.6 indicate these transformations through the relationships and levels that have been created and modified within the power structures for both CBFM and WMAs. Fig. 6.5 shows the re-scaled relationship between the VNRC and the District Council and the construction of a new scale of tobacco investment at the village level. Importantly, the newly constructed scale is not entirely separate but, as discussed in section 6.3.3.1, overlaps with the VNRC and with other investor groups at the village level. This group of investors has emerged as a driving force within the governance of the project. 204 Fig. 6.5: Re-scaled Power Structure as Performed in CBFM. Modified relationships and levels within the structure are highlighted in red. Legend for Figs. 6.5 & 6.6 Village Governance level AA Authorised Association DNRAB District Natural Resource Advisory Board Ministry Ministry of Natural Resources and Tourism PMO-RALG Prime Minister’s Office of Regional and Local Government Regional Sec. Regional Secretariat VC Village Council VGS Village Game Scouts VNRC Village Natural Resources Committee WMA Wildlife Management Area 205 The re-scaled power structure within the MBOMIPA WMA demonstrates several additional relationships, notably the re-scaled relationship between investors and the Village Council discussed in section 6.3.3.2. The National Consortium of Authorised Associations is highlighted as important in two ways in Fig. 6.6: Firstly, the discursive construction of scale is evident in this new organisation; and secondly, this scale has been strengthened by the development of close links with both the AA and with the NGO sector (WWF). Finally, Fig. 6.6 incorporates the scalar conflicts that are taking place within the AA and the axes of power that have been identified in section 6.3.3.2 between administrative division and land contributions to the WMA area. Fig. 6.6: Re-scaled Power Structure as Performed in WMA. Modified relationships and levels within the structure are highlighted in red. 206 6.5 Summary In this chapter I have investigated the performance of policy in Tanzanian CBNRM. I have adopted a scalar perspective to achieve this, arguing that the separate policy pathways adopted in both the forestry and wildlife sectors in Tanzania have produced different governance systems, and I have identified distinct scalar configurations of power that result from this. This is clearly seen in the implementation of a parallel governance system in the wildlife sector, where a new inter-village level sits at the centre of governance processes. It is clear that the governance systems prescribed in policy and performed in practice entail shifts in the relationships between different actors and institutions within the governance system, as seen in the changing power relationship between the Village Council and the newly empowered VNRC within CBFM. This chapter has investigated the characteristics of the scalar configurations of power that were being performed in both case studies, and I argue that one of the clearest features is their shifting nature. There are clear examples from both sectors of the protection and strengthening of the power system prescribed in policy and the benefits that it brings to those elected to serve as leaders on the VNRC and the AA, but also multiple examples of the ways in which different actors were invoking scalar strategies in politics of scales as they attempted to manipulate the scalar topography to their own ends. I argue that discursive practices are central to these processes of transformation and reification and indicate the processes of subjectification which were ongoing within both the WMA and the CBFM case studies. This is most clearly evident in the control of information that was shared with those outside of the principal governance institutions. Struggles over power are at the heart of the politics of scales that are witnessed in the case studies, and this is further indicated in the scales of transformation and conflict, which mirror the scalar configurations of power in each sector. I refer to the politics of scales in both case studies as taking place at different scales but involving the same politics. In the forestry sector scalar re-organisations are focused around the VNRC and conflicts revolve around the dominance of different resource user groups within the village level. In the wildlife sector, re-organisations are taking place to both circumvent the AA and re-scale investments to the village level, but are also evident in the scale bending that is taking place through the National Consortium of AAs and its developing voice at the national level. Conflicts within the WMA governance structure are dominated by those within the AA however, and divide the participating villages along two clear axes of power relating to both geographical locations within the political-administrative system of 207 Tanzania and land contributions to the WMA area. Closely linked to the discussion of the roles of discourse coalitions in policy processes in chapter 5, the utilisation of competing narratives and discourses by actors and groups within scalar practices is seen to be crucial (see Bulkeley, 2000). The conflicts taking place within the MBOMIPA WMA and Kiwere CBFM case studies are indicative of what Hajer & Versteeg (2005) refer to as ‘dislocations’ in which conflicts are used as moments of resistance and challenge to the dominant power structure, producing moments of routine-breaking. The discursive practices used within these conflicts reveal not only the power of discourse within the performance of policy and the politics of scales that it has produced but, I argue, also reveals conflict to be a strategic arena for such practices. This links clearly to the practice of ‘active destruction’ described by Fraser (2010) in which scalar configurations are deliberately dismantled in attempts to re- configure power systems for private gain. The political ecology of scales presented here is closely linked to the discussion in chapter 5 in the ways power devolution is a political project subject to multiple driving forces across a complex network of actors. This chapter adds to this perspective by investigating the non- containerised nature of scales and the ways in which actors are involved in re-shaping the scalar landscape, producing new scales of governance and forging new links between levels, creating what Neumann (2009: 404) refers to as “new relational socioenvironmental spatialities”. The scalar configurations of power prescribed in policy and being performed in reality, including the struggles that are taking place and scalar re-shuffling that is outlined in this chapter, place power as a central theme in the socio-political complexities at work within the case studies. As in chapter 5, where the political and power-driven nature of policy reform and the different policy pathways and distinct governance systems in place in Tanzanian CBFM and WMAs have been discussed, the politics of scales discussed in this chapter indicate that access to, control over and benefit from natural resources are at the core of these issues. This is further highlighted in this chapter as similar processes taking place in both the forestry and wildlife sectors, but as different scalar practices in accordance with the scalar topographies of power prescribed in each sector. In the next chapter I continue the theme of the implementation of the prescribed governance system and its performance within the case studies. Whilst in this chapter I have looked further into the issue of power within the prescribed system and the ways it is being 208 manipulated in reality, in the next chapter I continue this theme by considering the socio- cultural landscape into which CBNRM has been implemented in Tanzania and the ways in which actors are able to manoeuvre themselves and the governance system or avoid it entirely. Chapter 7: The Performance of Policy II Hidden Aspects of Governance and Participatory Spaces 209 7.1 Introduction This chapter continues to explore the enactment of governance within the case studies of Community-Based Natural Resource Management (CBNRM), exploring this performance through non-prescribed and often unseen aspects. I seek to develop the idea of this performed governance system as one that is produced both around and through the system set out in policy (the prescribed governance system). I follow the approach used in institutional theory and discussed by Robbins (2000: 427) that a governance system in performance consists of de jure laws and de facto rules and social norms in combination and is “adapted and curved around the contours of local power” (see 2.1.2). Recognising the political nature of implementing CBNRM in such a context, and the key role that the amalgamation of both prescribed and performed governance systems has upon the outcomes, successes and failures of such endeavours is critical to adding insight to the disenchantment with CBNRM discussed in 1.3, and the socio-political processes shaping such outcomes. Using the terminology of institutional analysis, I consider the ‘rules of the game’ that are in play within the governance system, drawing on cultural factors that have influenced these. The performed governance system is articulated through an informal economy and cultural aspects of participation. These processes and mechanisms are a blend of existing systems and relationships and newly created ones around the prescribed governance system. I also consider the ‘tricks of the game’ within this chapter, and I use this term to include the processes uncovered through fieldwork which indicate the importance of informal economies operating around these CBNRM systems. These tricks of the game constitute the mechanisms and processes by which actors have both adapted and circumvented the prescribed governance system, producing a hidden, but equally important aspect within the performed system. It is widely observed that the structures of the state are altered in performance, used by people and re-shaped by them to their own advantage (Brockington, 2008). In chapter 6 I looked at how this was taking place in politics of scales, scalar transformations, conflicts and power struggles within the case studies. I now extend this and look at how the rules of the game, in both the prescribed and performed governance systems, facilitate procedures of rule-breaking that go unremarked, unchallenged or unsanctioned. I also begin to explore issues of neopatrimonialism, bribery and clientelism as mechanisms of corruption in this chapter, investigating the context of power that such systems draw on, and 210 how they contribute to the inequalities in benefits described in 6.2 in both legitimate and illegal ways. In this chapter my focus is principally on the informal realm and the connections between formal and informal types of power, rules of the game and the relationship between the prescribed and performed governance systems. I consider power in two principal ways: firstly through the power structures into which CBNRM has been amalgamated, discussing types of power already present within the research villages and how these have merged with and transformed the governance system put in place; secondly through the power of different discourses enacted by groups within the governance system and how these have an effect upon decision-making and governance arrangements. The initiation of Wildlife Management Areas (WMAs) and Community-Based Forest Management (CBFM) policy within the research villages marked important shifts in the power structures, the politicisation of natural resources, and the processes by which different groups benefitted from them (see 5.2, 6.1 & 6.2). A former MBOMIPA project staff member described how significant informal economies were present around the project both before and during the REWMP phase and through to the MBOMIPA phase (Interview P73). The emergence of national policy implementing WMAs in Tanzania shifted the systems in place and “ushered in the transferral of control away from the informal economy” (Interview P73), which had complex results for the different groups involved. The politics of devolution described in section 5.4 must also acknowledge these hidden systems of power operating around natural resources, therefore. Some villagers had been part of hunting networks controlled by resident hunters and involving district and regional officials, tied together in processes of private benefit through bribery, clientelism70 and neo-patrimonialism. The initiation of WMAs heralded the transfer of control over managing wildlife resources to a new set of actors, removing hunting rights and shifting sources of funds away from such groups. These shifts in the power system, an academic researcher argued, had significant impacts in the forestry sector also, where actors and groups began searching for ways to benefit from the new systems and sources of revenue (Interview P2). The performed governance system is a continually changing arrangement of previous relationships and systems, the implementation of the prescribed governance system and the integration of new 70 Whilst neopatrimonialism and clientelism both consider ties of dependence and processes of patron- client distribution, clientelism, in modern day Africa, is closely associated with practices of favouritism in decision-making Olivier de Sardan (2008). 211 processes as actors attempt to position themselves and the scalar configurations of power for private benefits. The outcomes of such attempts is a mixture of ways in which people have utilised the prescribed governance system, shifting and manipulating it to their interests (see 6.3), and ways in which people have found ways to circumvent it, diverge from policy and evade sanctions. In the next section I outline examples of the rules not being followed within both the CBFM and WMA governance systems, and I attempt to highlight some of the differences between the prescribed and performed governance systems within the case studies. As with villagers’ perceptions of those who benefit the most from CBNRM, the leaders of both the VNRC and the AA were frequently highlighted as persistent rule-breakers who employed several tricks of the game (but see also 7.4.1). Further instances of rule- breaking were identified as impracticalities with the prescribed governance system, rather than tactical efforts to avoid the rules (see 7.2.1). 7.2 Deviations from Policy: Rule Breaking in the Performance of Policy The devolution of power and authority over revenue collection and natural resource management to the local level in both CBFM and WMAs has been a contentious and not universally-supported shift in natural resource management policy in Tanzania (see 5.4.2). Deviations from policy have been recorded, therefore, in cases of revenue policy simply not being followed, as described by an academic researcher concerning CBFM revenue systems, where some districts have ignored the 5% district cess collection rule, choosing instead to collect 100% cess (Interview P2). Other cases of non-acceptance of the changes brought in by the initiation of WMAs and CBFM have also been noted, especially within the WMA case study where claims were commonly made that resident hunting groups simply ignored the changes in rules governing their rights to hunt (Interview P86). A lack of adherence to the rules put in place within the WMA and CBFM management plans runs alongside problems with enforcement of these rules within both the WMA and the CBFM case studies. One owner of a tourist lodge described a key aspect of this problem when he depicted illegal hunting as “common at night and uncontrollable because of lack of funding for patrolling, and technology such as mobile phones, which mean hunting groups can be tipped off about raids” (Interview P79). Whilst the assertion that there are significant issues with enforcement of rules concerning hunting within WMA was strongly supported by the research carried out, it was difficult to judge the extent to which hunting was taking place within the WMA and harvesting was occurring within the CBFM forest outside of the official 212 systems as this was obviously a sensitive issue to discuss with respondents. Respondents’ estimations of how much of an issue these presented ranged from insignificant to the dominant problem that threatened the sustainability of the entire system within the forest or the WMA. Quantification of the problem has not been attempted. However, it was clear from the research carried out that illegal harvesting from the forest and illegal hunting within the WMA were both facilitated by enforcement issues, notably the effectiveness of patrols, and the systems in place to sanction offenders. In both the research villages, the inability of scouts to patrol the areas effectively was discussed by the scouts and the wider community as a product of the size of the area they need to patrol and the lack of resources available to enable scouts to achieve this (including shelter, shoes, clothing and monetary allowances; Focus Group P65, Local Assistant Report P25). In the CBFM village the result is that a small group of scouts were responsible for protecting a large area of forest and often did not patrol some areas at all (Interview P4), patrols were not always carried out as frequently as set out in the management plan (four times per month for CBFM), and the frequency of patrols varied according to the weather and season and the responsibilities faced by the scouts with regards to their farming activities at different times of the year (Interview P4, P94, P108, Local Assistant Report P25). In both the research villages, respondents described how the long rainy season particularly between March and May is often associated with less frequent patrols (Local Assistant Report P25; Focus Groups P65, P164 and P162, Interview P99). In the CBFM village, respondents described this as well known by the villagers, who could cause a lot of damage in the forest during this period, with little likelihood of being apprehended (Local Assistant Report P25). These respondents raised questions about the feasibility of carrying out patrols to the extent set out in policy under the current system due to the size of the forest, the long-distances involved in patrols and the opportunity costs associated with patrolling rather than working on their own farming activities (Focus Groups P164, P162; Interview P192). A further issue was the low allowances paid to the scouts, which is discussed further in 7.2.1. Within this context of problematic enforcement, instances of rule-breaking by different groups within both the CBFM forest and the WMA were widely perceived. Rule-breaking was also a sensitive issue and it was inherently difficult to gauge the level of rule-breaking taking place within either case study. Instead, I focused on identifying the mechanisms and ways in which rules could be broken in both case studies. A lack of fear or perceived likelihood of being caught and punished for rule-breaking was a key factor revealed through 213 fieldwork, and closely associated with the size of the forest and the WMA and the patrolling issues described above. The likelihood of being caught by the forest scouts in the CBFM research village was generally described as low, due to the size of the forest, alongside knowledge of when the scouts were going to carry out their patrols and the areas they usually covered (Interview P4). Likewise villagers within the WMA research village described how certain areas were popular with illegal users as they were infrequently patrolled by the village scouts (Interview P192). Types of rule-breaking include exceeding quotas for harvesting activities (Interview P86), for example permits for charcoal production are given according to the number of bags that will be produced, but this amount is often exceeded and extra bags may be paid for retrospectively, may be excused through a bribe (see 7.4 & 7.5), or may be hidden from authorities and transported out of the forest for sale in Iringa town (Interview P5, Local Assistant Reports P13, P20). Further tricks of the game were identified by villagers and local assistants in the transportation of illegal goods from the forest. Permit holders should both be escorted to the forest by a forest scout and should then pass through the village or the boundary gate at the end of their activities, where vehicles are inspected by members of the Village Natural Resource Committee (VNRC) or forest scouts to verify the goods harvested match those allocated in the permit (Local Assistant Reports P13, P22, Focus Group P59, Participatory Activity P67). Village respondents reported that “they have several tricks” (P59) to avoid these checks. In conjunction with issues of bribery opportunities (see 7.4.2), it was reported that concealment of goods was used to transport illegally harvested products. This was identified as taking place particularly for illegally harvested timber, which was concealed at the bottom of the truck and covered by the firewood for which a permit had been purchased (Local Assistant Report P25). Secondly, illegally harvested goods were thought to be transported at night to avoid being stopped by the forest scouts or the VNRC (who were also implicated in these activities), and this trick was used by many villagers as hard evidence that illegal harvesting was taking place because they reasoned that “if they have purchased a permit, why is it that they transport at night and not in the afternoon?” (Local Assistant Report P22). 7.2.1 Disincentives within the Prescribed Governance System The difficulties met by scouts with regards to the demanding nature of their patrols, the poor equipment available to them and the threats they faced from armed groups within the forest 214 and WMA areas all acted as disincentives to carrying out the patrol system as prescribed in policy. This is compounded by the low wages paid to forest and wildlife scouts within the villages which has created a system that encourages illegal and informal economy processes, resulting in a perception by research academics that a significant proportion of transactions continually take place outside of the prescribed governance structure: “accountability in accounting is not so much a case of where the money goes to from the accounts, but what doesn’t make it into the official receipt book” (P5). Whilst greed is often blamed for the processes taking place in the informal economy and instances of bribery and corruption (e.g. Leader-Williams et al., 2009; see also 5.4.1), the research findings point to the facilitation of such processes by the prescribed governance system itself. Using the CBFM case study as an example, the role set out for forest scouts positions them as the protectors and enforcers of the rules of the management plan for the forest (see 5.2.1). They are responsible for carrying out patrols, apprehending offenders and taking them back to the natural resources office within the village to face the appropriate sanction, which is the responsibility of the members of the VNRC. The distribution of benefits and problems surrounding the low allowances paid to forest scouts has created a relationship whereby these sanctioning processes were de-incentivised: “The scouts are often angry with the authorities because they feel they work hard and receive little benefit” (Interview P7); “Their allowance for patrols is very small, their work is hard, especially when it rains, and the fines [given to offenders] can be quite big, so they should get more for an allowance” (Focus Group P65). The performance of policy in this respect involves the financial incentivisation of bribery, as described by a local assistant, who said “A villager who is caught making charcoal will offer a bribe to the person who has caught him in order to finish the problem. For the villagers who don’t have the money, they are able to promise to give them an amount from the sale of the goods” (Local Assistant Report P20). The WMA system has attempted to move away from such processes through the decision taken to employ MBOMIPA scouts, rather than utilising the village scouts. There are 37 scouts now employed by MBOMIPA, which means a smaller number of scouts are carrying out patrols over a vast area. However, MBOMIPA is able to pay them an allowance that is very substantial compared to that of the forest scouts ($558 annually compared to between 215 $50.4 and $8471 annually). The differential politicisation of resources and the high value of wildlife resources within the hunting sector (see 5.4.1) has driven this move on the basis that adequate allowances for the MBOMIPA scouts will enhance the perceived value of employment in such roles, encourage abidance with the rules of the WMA, promote effective patrolling and encourage reporting of report rule-breaking, rather than acceptance of bribes. Whether this rational-economic shift in policy will deliver a reduced incentive to benefit from bribery could be called in to question however by the informal rules of the game discussed in 7.3. 7.2.2 Discourses of Rule-Breaking In both research villages, local respondents argued that beyond the ‘insider’ rule-breaking associated with the leaders of the Village Natural Resources Committee (VNRC) and Authorised Association (AA), cases of rule- breaking were dominated by outsiders (non village residents). This probably reflected an unwillingness to admit to rule-breaking or to incriminate others within the village. Where rule-breaking by the villagers was admitted, it was usually explained as inevitable given the harsh conditions in which people live, levels of poverty and necessity of providing for one’s family (Local Assistant Report P25, Interview P92): “the conditions we live in sometimes cause people to break the rules of the forest...many people cut firewood in the forest without being scared of being caught and they are of one heart that if you are caught, it is just bad luck” (Local Assistant Report P25). Villagers invoked a discourse of necessity and poverty therefore to justify rule-breaking as necessary to live. Contrastingly, rule-breaking by outsiders was described as unacceptable, and surrounded by a discourse of others not respecting their rules and not understanding the importance of managing resources properly. As mentioned earlier, the range in perceived levels of rule-breaking and the impacts it had upon the WMA and CBFM forest was very large. Whilst many village respondents portrayed illegal activities as largely a small-scale nuisance within a generally very successful example of CBNRM in Kiwere, local assistants’ reports included testimonies of villagers who perceived the condition of the forest to be deteriorating and management of the forest to be very poor, with high levels of illegal harvesting, particularly of firewood and charcoal making, taking place by villagers, outsiders and the committee themselves (Local Assistant 71 This is calculated from the number of patrols supposed to be carried out each month and the average number of scouts that carry out each patrol (three to five). 216 Reports P13, P14, P15). These respondents argued that persistent rule-breaking was encouraged by the rules themselves, which were not strong and fines that were not high enough to deter people from using the forest which, alongside people’s perception of the low likelihood of being caught, encouraged people to carry on harvesting as they wished (Local Assistant Report P13). A similar discourse of forceful protection was invoked by a group of tourist lodge owners operating in the WMA area and Ruaha National Park, who described the ability of the AA to successfully handle hunting pressures as inadequate and blamed both resident hunters and local communities for persistent illegal hunting taking place within the WMA (Interviews P79, P81, P85, P103). 7.3 Enforcement, Moral Economy & Reciprocity The issues described in 7.2 with respect to the enforcement of rules and deviations from policy regarding patrolling and sanctioning of offenders are closely linked to de facto rules and social norms within the research villages. Not only is enforcement tempered by the difficulty in apprehending offenders and the disincentives described above that facilitate extra-legal harvesting and bribery to accommodate this, but research revealed several aspects of a moral economy, or what Neumann (1998) terms ‘local justice’ resulting from the tensions between state policy and local subsistence requirements and customary use (see 2.1.3). Local management of natural resources, as seen in both CBFM and WMAs in Tanzania involve local enforcement of the rules governing such resources. This creates an intrinsic tension between enforcers (in this case forest and wildlife scouts) and villagers using the resources. In both research villages scouts within the villages described how their job was made more difficult because of the close personal connections they often had with people they were supposed to apprehend and send to the office to be sanctioned for breaking the rules of the forest or WMA. An academic researcher summarised these issues as “It is difficult to exercise the law when relations are involved...When you catch a culprit who is the son of your aunt, and you decide to prosecute, you should expect problems from family members. These are big problems for scouts and committee members” (Interview P7). Leaders of the CBFM project also described how they faced severe pressure from villagers for imposing rules and fines for breaking those rules, especially in the early stages of CBFM (Interview P35). Within MBOMIPA this issue was reduced by the patrol system of the scouts, who did not carry out patrols within their own or neighbouring villages, but would 217 travel to the other administrative division (Idodi or Pawaga) to carry out their work (Interview P89)72. The social costs of being a forest scout in Kiwere were revealed by VNRC members’ descriptions that “after they catch somebody, he [the offender] begins to blame the forest scout and to build hatred towards them, even though it was he [the offender] who committed the error” (Interview P44). The scouts also reported that enforcing the rules of the forest caused them to damage their good relations with the village community (see also 7.5), leading to threats against themselves, their families and their livelihoods. Scouts argued that when such threats were made, they reported it to the VNRC, so that one of the leaders would speak to the offender and warn them against doing anything to harm or cause problems for the scouts. However, local assistants reported that threats of curses and superstitions still often prevented the scouts and the VNRC from carrying out the jobs as they were supposed to because they feared for their lives (Local Assistant Report P14). Research also indicated that there is a significant reciprocity element to enforcement and the level of sanctions given to offenders within the CBFM project. Local assistants reported from villagers that the level of fine faced by someone caught breaking the rules was related to the socio-economic status of the offender “for example a rich person who destroys the forest for timber, they [VNRC] often fail to follow the steps properly because that person has lots of money” (Local Assistant Report P22); “punishments that are given out by the leaders show favouritism for those who are rich over those that are poor...poor people are held to the rules” (Local Assistant Report P25). Local assistants also reported that lower fines or no punishment may be given, or fines may not be followed up when the offender is seen to be potentially useful to the scout or the offender and they could ask for the favour to be returned at a later date (Local Assistant Report P25). Villagers also reported through a local assistant report that leaders feared heavy fines would encourage some offenders to reveal the leaders’ own illegal activities in the forests, leading to significant bias in the sanctions imposed (Local Assistant Report P20). The performed governance system in both WMAs and CBFM is one where the moral economy of local communities partly governs the processes of enforcement and sanctioning. It is important to note that these informal institutions not only govern the actions of scouts, the VNRC and the offenders, but are also a vehicle for the development of patron-client relations and the emergence of multiple forms of corrupt practice (Chabal and Daloz, 1999). 72 It is unclear how this system has changed since the new MBOMIPA scout and patrol system was brought in. 218 In the next section I discuss three processes of corrupt practice ongoing within Tanzanian CBNRM: neopatrimonlialism, bribery and clientelism. These practices exist in tandem with the prescribed governance system in a balance of political and moral-economic factors making up the processes of ‘real governance’ (see Olivier de Sardan, 2008). They are therefore, further core features of the performed governance system. 7.4 The Corruption Complex In this section, I begin to consider the hidden aspects of governance that constitute different types of corruption. There are many different classifications within this area, and I follow Olivier De Sardan’s (1999: 26-7) terminology of the ‘corruption complex’ as an umbrella term for “a number of illicit practices...[which] offer the possibility of illegal enrichment, and the use and abuse to this end of positions of authority”. Such a definition could encompass many techniques including classic corruption, nepotism, clientelism, embezzlement, abuse of power, misappropriation, influence pedalling, prevarication, insider trading and abuse of public funds (Olivier De Sardan, 1999). It is important to note that the processes of corruption identified within the Kiwere Village Land Forest Reserve (VLFR) and MBOMIPA WMA are closely linked to the power struggles described in 6.3: Neopatrimonialism, bribery and clientelism all take place within a formal context of established rules and governance processes (see Kelsall, 2011). Uncertainty and a lack of transparency are vital to the continuation of these informal processes, upon which political legitimacy rests (Chabal and Daloz, 1999). The protection of power and discursive construction of scale described through the control of information and empowerment in 6.3.1 are vital to the maintenance of these corrupt practices (Moss, 1995; Chabal and Daloz, 1999), and constitute important tricks of the game therefore. 7.4.1 The Corruption Complex: Neopatrimonialism Neopatrimonialism involves the distribution of public goods for private benefit (see 2.1.3), and within African societies is widely regarded as a ‘perk of the job’ to which those in positions of responsibility are entitled (Bayart, 1993; Chabal and Daloz, 1999). The perks of the job available to different actors within CBFM and the WMA are clearly closely related to the issue of elite capture of benefits discussed in section 6.2. The capture of benefits within CBFM and WMAs, particularly by those working as leaders on the VNRC and AA takes place partly through legitimate means within the prescribed governance system (see 6.2.1), 219 but is joined by myriad accusations of neopatrimonlialism through money being ‘creamed off’ from public funds, creative accounting that covers private use of funds (Interview P9), alongside the issues of bribery discussed in this chapter. Within the CBFM research village such claims were extremely widespread, whereas within the WMA research village, most of the interviewed village residents claimed not to know enough about the WMA to understand the sources of revenue, how these were distributed and how they were used for good or ill (see 6.3.1). Amongst the AA Representatives however, accusations if improper use of funds became common again (Interviews P168, P175). Discussions with NGO staff and research assistants described a widespread culture within Tanzania of perks of the job. They described this as especially related to NGO and development funding where supplementing pay with benefits such as expensive cars, travel expenses, invitation to training sessions and conferences and the ubiquitous per diem were standard practice expected by those working on such projects (Interviews P1, P4). Olivier De Sardan (2011: 26) describes this as a characteristic of ‘project modes’ of local governance, which create an “oasis of prosperity, which is the envy of all public servants, whose ambition is to be hired by a project, or failing that, to be financed by it”. Whilst this distribution of funds and perks may operate within the governance system, and be justified by a discourse of project maintenance and administrative duties, it exacerbates the capture of benefits by those already legitimately benefitting from their involvement in the VNRC and the AA. The MBOMIPA case study provides an excellent example of such processes and techniques at work. The AA recently purchased a new project vehicle, which was partly funded by the USAID totalling $22 500, alongside additional funding sources to cover the remaining costs of purchasing and paying the necessary duties on the vehicle (Interview P101). This large expense was justified as enabling effective patrols and facilitating management of the project, but is likely to bring large benefits to the leaders of the AA, who will gain access to the vehicle both to travel to meetings and for personal use when they feel able to justify or conceal such use. The AA also exemplifies neopatrimonialism through its expenses system. Large amounts of money can be accrued through expenses claims: MBOMIPA representatives were eligible to claim up to TSh 45 00073 ($31) for each meeting they attended (Interview P109). Recent conflict because some representatives claim more expenses than others resulted in each 73 Consisting of TSh 20,000 for attendance and TSh 25,000 for accommodation, food and other costs (P109). 220 representative being given a flat rate of TSh 51 000 ($35) plus the cost of their travel expenses (Interview P109). Respondents involved with the Board of Trustees for MBOMIPA reported that some representatives were very keen to hold as many meetings as possible in order to be eligible to claim such expenses as frequently as possible (Interview P103). They described this as a type of corruption within the AA, whereby the committees “swallow up huge amounts of the budget, leaving just a small amount for the villages” (Interview P103). The major categories of expenditures in the MBOMIPA accounts are shown in Table 7.1. This corroborates the argument that management costs are a legitimate but high proportion of the expenditure of the AA. Whilst this category is the smallest proportion of the total, the accounts detail that it was spent exclusively on allowances, travel and accommodation expenses related to meetings, training and visits to other WMAs and mobile phone credit for representatives (totalling 44 individuals). These claims are not against the rules, but do represent clear examples of the abuse of office and the maximisation of benefits and perks of the job by those elected to the MBOMIPA AA. Expenditure (US$) 2007-8 Proportion of total (%) 2008-9 Proportion of total (%) Management costs 7,785 18 19,348 21 Protection costs 12,933 30 25,764 28 Development costs 22,910 52 45,553 50 There were also a number is discrepancies within the accounts, however, which complicated assessment of how funds were spent. For example, a driver’s allowance is recorded twice for the year 2008-9 with different amounts for each, and adjustments to the constitution of the AA are recorded under ‘development activities’. Whilst the development activities budget is the largest category in the expenditure, these funds are split between 21 participating villages (and a total population of over 30,000 people). Each village received just $828 for the year 2007-8 and $1,448 for the year 2008-9. Likewise the expenditure for protection is large, but conceals the fact that the expenditure that ended up in the pockets of scouts and drivers for each year was a total of $7,634 (2007-8) and $14,883 (2008-9), making up just 59% and 58% Table 7.1: MBOMIPA Authorised Association Expenditure by Category 2007-9. These data were provided in the official accounts of the AA. 221 of the total protection costs respectively. It is unclear from the accounts what the remaining, fairly large protection expenditures were. It becomes very difficult to assess the ways in which funds, as recorded in the official accounts, are actually being used. Many of the transactions taking place are hidden (deliberately or due to poor accounting skills) within the official figures, and a large number of transactions reportedly took place outside of the formal economy of the VNRC and AA. For example, at a village meeting requests for made for the committee members not to “hide yourselves in the information” (Interview P148), relating to the perception that figures provided to the wider community concealed many private uses and benefits under official categories. Such processes and techniques take place within the “realm of rumours and suspicion” (Olivier De Sardan, 1999: 30) and, considering the sensitivity of such topics within interviews and discussion groups, it was very difficult to verify such accusations. Although extensive triangulation was attempted to gain as much depth of insight into the validity of such claims, caution must be advised in accepting such accusations as truth. For example, within the CBFM research village, accusations that the leaders of the project “use the opportunities of being leaders to harvest timber and to make charcoal” (Local Assistant Report P13) were common, but it was not clear how much of this use is legitimate and how much is illegal. It is clear from the research I carried out in Kiwere, and in other studies carried out (e.g. Lund and Treue, 2008) that leaders of the VNRC are able to benefit from CBFM through the purchase of permits, partly because they usually come from richer socio- economic groups and have the funds available to participate in the project, and also partly because, as leaders of the VNRC, their applications for permits are unlikely to be refused. The latter of these processes indicates a ‘perks of the job’ component of such neopatrimonlialism, which was commonly thought to be accompanied by more sinister components of deliberate embezzlement and illegal harvesting: “the leaders make charcoal and steal timber without permits for their own activities” (Local Assistant Report P13); “the question to ask is whether they follow the procedure of obtaining a permit, and if they do, how is it that they can produce large amounts of charcoal whilst the activity for the rest of the community is forbidden?” (Local Assistant Report P22). The boundaries between unfortunate (but legitimate) capture of participation and benefits, perks of the job and the abuse of power and funds within neopatrimonialism remain very unclear. 222 The accusations and widely-held beliefs that those working on the VNRC and the AA were closely involved with illegal activities could support the argument made by Brockington (2008: 118) that in Tanzania he witnessed a lack of faith in official figures and those holding positions of public office, claiming that, generally, “the expectation was that money will be misspent and positions abused for personal profit”. Chabal & Daloz (1999) similarly outlined this situation as one of resigned fatalism and the expectation of the abuse of power. Such expectations and a level of acceptance of the benefits enjoyed by those holding positions of power was expressed in a report by a local assistant in the CBFM village: “every leader does his work considering his own interests first, before he considers those of the community. Even if others are elected, they can have the same customs as the first leaders who are involved in bad uses of the money and leadership positions” (Local Assistant Report P20). Such discourses of acceptance and inevitability could be seen to represent the internalisation of such values and the subjectification of the corruption complex (Gaventa, 1982; Lukes, 2005). This is indicative of the first form of power outlined by Haugaard (2003). He termed this type of power construction ‘causal predictability’ created through structuration and confirm structuration whereby social meaning is reproduced and recognised by actors (see Table 2.1). This example indicates that the social power entitling certain actors to perks of the job and to employ tricks of the game have been internalised. In this situation the members of the community may recognise the practice of neopatrimonialism and perks of the job but do not attempt to overthrow the system which supports these processes, but seek to position themselves to benefit in these ways. The politics of scales and conflicts described in 6.3 were described as a struggle for power over natural resources and the benefits from them. Neopatrimonial forms of exchange and the perks of the job form an important element of these (Chabal and Daloz, 1999; Robbins, 2000). Olivier De Sardan (1999) argues that African forms of corruption are simultaneously despised, generalised and banalised, usually going uncontested and supporting Brockington’s (2008) thesis that people have become both accustomed to and expectant of such practices. Such a discourse of corruption expressing both expectation and inevitability of perks for those in power was exemplified by one villager in the CBFM village who said that he “had no suggestions about improving the management of the forest because even if they changed the leaders, the new people would do exactly the same as the last ones” (Interview P53). Stories of the embezzlement of funds by specific members of the VNRC and accusations that the MBOMIPA leaders had utilised the funding available for educating two orphans from 223 each village for their own children, or those of their friends and relatives, were good examples of this (Interviews P4, P109, Local Assistant Reports P13, P17). Chabal & Daloz (1999) argued that the informality of African politics and the disorder on which it rests provides opportunities for those that know how to play the game and get the most from it. This is similar to an argument put forward by Olivier De Sardan (1999), who claimed that a ‘logic of predatory authority’ in African political systems creates a belief by those in power, and often more widely, that such a position entitles one to certain benefits, creating a discourse of entitlement and serving oneself. He states that “illegal enrichment and nepotism are definitely supported by positive social values, namely the necessity to seize all opportunities” (Olivier De Sardan, 1999: 43). In Tanzania these arguments can be closely linked to a discourse of self-help, which has played an important role during the socialist and multi-party democracy periods in Tanzania (see 3.2). Such arguments are important to the issues surrounding patrols discussed in 7.2.1, within which it is commonly argued that increasing scouts’ wages for carrying out patrols will provide the incentive required to reduce instances of bribery and deviations from policy. According to Olivier De Sardan’s view (1999), taking such measures will not necessarily lead to any such changes on behalf of the scouts, who will be driven to utilise opportunities for bribery regardless of the wages they receive. 7.4.2 The Corruption Complex: Bribery As discussed above, there were extensive rumours and beliefs of the widespread acceptance of bribes by the scouts, the VNRC and leaders of the AA. Bribes were believed to be offered and accepted for allowing illegal goods to leave the forest, for allowing hunting to take place in the WMA in excess of the quota or without proper permits to hunt (Local Assistant Report P25, Focus Group P59, Interviews P103, P168), which one villager from Kiwere discussed as a process of “they [users, scouts and the VNRC]join hands and the goods pass through” (Focus Group P59), whilst a local assistant recorded in his report that people believed that “the fines that are given out by the leaders, for the most part are for their own advantage but to help the community is just a small percentage” (Local Assistant Report P25). Similarly a tourist operator discussed how in MBOMIPA “the project has got worse as its got bigger because corruption has grown along with it” (Interview P103). My study confirms the arguments made by Brockington (2008) and Olivier De Sardan (1999) that such corrupt practices invoke criticism but little retribution or social action. In the previous section I 224 discussed the internalisation of values that emphasise the maximisation of private benefits to explain this, yet here I offer an additional explanation that works alongside this. Investigation into the suspicions, rumours and accusations of bribery and other techniques suggested that these could also be influenced by the lack of understanding and engagement by the wider community with CBFM and the WMA. For example, suspicions of bribery within the CBFM research village were very prevalent, and closely related to people’s perceived lack of understanding of how the system works and inability to hold the VNRC and forest scouts to account. One villager described widespread bribery and pocketing of bribes and confiscated goods but, when asked about this in further detail, the respondent explained that he/she presumed this was going on because, if it wasn’t being pocketed by the scouts and the VNRC, he did not know where these goods or the money was going (Local Assistant Report P14). A local assistant corroborated such processes by reporting that the lack of information shared with villagers led many to believe the something was being hidden (Local Assistant Report P15). The accusations made against the leaders of the committees in both of the case studies suggest that, whilst the power structure is reified by the control of information and prevention of the construction of discursive consciousness within the wider community (see Table 2.1), this does not go entirely unchallenged. The internalisation of values that facilitate corruption does not prevent it from being widely decried in Tanzania (see Fig. 7.1), and featuring prominently in political scandals. It was very difficult to gauge to what extent the accusations of bribery and other forms of corruption were accurate within both the WMA and CBFM case studies. I argue that claims of corruption may be a political weapon employed by the wider population as a form of resistance to the power system in place. Actors may attempt to position themselves to benefit from the mechanisms of corruption and tricks of the game discussed in this chapter by facilitating the removal of the leaders of the AA or VNRC through accusations of corrupt practice. Scott (1990) describes how this kind of resistance, presented as a hidden transcript, does not necessarily reflect the truth of the situation, but it is an important mechanism by which the dominant power system can be destabilised. Here, I argue, there is a link to the discussion of politics of scales and the role of conflicts within such power struggles, as described in 6.3. The accusations of bribery and other forms of corruption may be a political strategy employed as part of politics of scales and scalecraft. Actors use such tactics to destabilise the scalar topography in the process of attempting to 225 negotiate the system of power and position themselves to benefit from it (Fraser, 2010; see also 6.3.4 and 2.4.2.1). 226 Fig. 7.1: Government Anti-Corruption Poster (Institute for the Cessation and Contestation of Corruption, Government of Tanzania). The poster, displayed in the Village Council Offices in Kiwere, shows a political campaign in which the speaker is making promises of food and money to the local community, whilst his assistant is anticipating the bribes they will receive from these people. A member of the crowd is protesting that the people are hungry, whilst one man is walking away saying that this type of corruption only brings bad leaders and he refuses to be a part of it. The bottom of the poster reads: “If you involve yourself with electoral corruption, you inhibit the growth of democracy and good governance. Stop corruption.” 227 7.4.3 The Corruption Complex: Clientelism There are several ways in which the boundaries between ‘good neighbourliness’, bribery and clientelism are blurred within the case studies, especially in MBOMIPA, where such issues were a dominant complaint of the MBOMIPA representatives. Whilst the processes that are described in this section could be classified in several different ways according to the type and characteristics of tricks of the game they represent, I discuss them together as they all display impacts upon the decision-making processes taking place within the case studies, and relationships of patronage can be identified. Some elements of the processes discussed here are open, whilst others are hidden, although all operate outside of the prescribed governance system. Within MBOMIPA a clear set of close relationships has developed between the Board of Trustees, the leaders of the AA, current WMA investors, and potential future investors. In one sense these relationships are cultivated by the investors themselves, who declare the importance of building good working relationships with the leaders of the AA, following the concept of ‘good neighbourliness’ that is very prevalent in Tanzania since its socialist era (see 3.2.2). The importance of gift-giving and the moral duty this involves within African cultures (see Olivier De Sardan, 1999) contributes to such processes and has resulted in investors providing gifts of food, drinks, trips to visit the investors and cash (Interview P102). The importance of such gifts to smooth the process of applications was demonstrated within the research village where a potential investor in the village (not the WMA) had, prior to the signing of investment contracts, provided TSh 400 000 ($276) to the Village Council, which was explained as a gift to the village to help with development activities (Village Meeting P186). Whilst members of the Board of Trustees claimed that for the WMA they always discouraged such activities (Interview P102), researchers working in the area and MBOMIPA representatives claimed that the trustees were complicit in such activities, had close relationships with investors and that their advice to the AA was often influenced by these clientelist relationships (Interviews P86, P175): “it has appeared over the past few years that the advice they give is not in the best interests of the project or villages, but involves personal gain for them, or support for their friends’ bids” (Interview P86). The Board of Trustees is made up of prominent local businessmen, the majority of whom run tourist-related enterprises, including safari companies and hotels. The nature of their work was described as a benefit by district officials who claimed that it gave them a real interest and often valuable experience in this sector, which MBOMIPA must work within (Interview P88). One 228 MBOMIPA representative described how the nature of the business interests of those serving on the Board of Trustees made them too closely connected to MBOMIPA. One of the members of the board had been the owner of an existing tourist lodge within the MBOMIPA area and is now one of the official investors in the WMA, whilst others were setting up investment contracts on village land around the WMA (Interview P175). The respondent argued that they could not provide good advice when they were so closely related and financially involved with the WMA and its activities. The manipulation of decision-making for private interests is closely linked to the discussion of manipulation as part of the re-scaling of both forest harvesting within CBFM and MBOMIPA revenue systems in section 6.3.3. MBOMIPA representatives and village authorities from one village who participated in a group discussion described how they would prefer the Chairman and Secretary of MBOMIPA to be employed from outside the participating villages because “these are very influential positions and important decisions are made by them and, at the moment, these people still represent their own villages’ interests, not solely the interests of the WMA” (Interview P137). The leaders of the AA were also accused of biased opinions regarding the conflict within the WMA (concerning the distribution of revenues between the participating villages; see 6.3.4.2), and these respondents argued that as the leaders came from villages in Idodi that have all contributed land, their handling of the conflict would be severely compromised by the additional benefits these villages would receive if the revenue system were altered (Interview P137). Clearly, mechanisms of clientelism hold potentially significant financial benefits for the agents within the relationship, who respondents described as benefitting from large sums of money from investors wishing to hold favour with the leaders of the AA (Interview P86). The impacts of such relationships extend beyond these financial transactions however, into the decision-making processes within the AA, where reports of those thought to be benefitting from such arrangements are described as pushing the other representatives to make decisions according to the wishes of the investors they are closely related to, and even to attempt to circumvent the prescribed governance system and the constitution of the WMA in favour of these investors. MBOMIPA Representatives claimed that “at the last budget meeting, the leaders tried to use the WMA regulations of 2005 instead of those issued in 2010. This brought an argument and caused many of the representatives to believe that they wanted to benefit some of the investors with whom they have close relationships” (Interview P172). Interviews with MBOMIPA representatives revealed that a decision had been made to 229 allow one of the investors to hunt within a zone of the WMA that, according to the management plan, is reserved for non-consumptive tourist activities only. The respondents described how the leaders of the AA had persuaded the others to allow this for the year 2009 on the basis that the investor could not collect any revenue from the planned lodge until construction is complete. The leaders of the AA suggested that they should support his proposal of hunting as a way to allow the investor to collect revenue in the meantime “but the shocking thing is that nobody knows how many animals he hunted” (Interview P172). The representatives decided that this was unacceptable and refused to allow the leaders to let this continue, causing much conflict within the AA and between the AA and the investor, who “after he was refused his request for the year 2010 refused to pay the money owed to the AA as part of his investment contract” (Interview P175). This refusal to pay his investment contract fee to the AA is one of the primary reasons why the development budget for the year 2009-10 decreased dramatically, and each village received just $552. The impacts of such relationships can be very tangible for the communities participating in MBOMIPA, therefore. The close relationships between the leaders of the AA and investors were explained by others as due to demands from the leaders for rewards, gifts and money, amounting to extortion rather than clientelist offerings from the investors: “the chairman called the investor on the phone and demanded 15 million shillings [$10 344] so that he could rent a tractor, and when the investor refused, he [chairman] phoned the wife of the investor and told her that her husband had denied him the money and the chairman would see if the investor would continue hunting in the WMA” (Interview P175). The same MBOMIPA representative went on to explain that the investor had been pressured into putting payments into private accounts, rather than the MBOMIPA account and had often been asked to provide extra money for MBOMIPA resources, such as the purchase of a spare wheel for the vehicle (Interview P175). Cash transactions were also reported to take place between the leaders of the AA and local villagers applying for work when available within the WMA “so that they get preferential treatment...the work available is down to the leaders who look after their friends and family and those that can provide an incentive” (Interview P191). It is not clear whether these techniques of clientelism and nepotism are initiated by those applying for work and wishing to enhance their application, or by the leaders themselves. Once again, the boundaries between bribery, clientelism and extortion within neopatrimonial systems are unclear. 230 Such relationships of clientelism were also apparent in the CBFM research village, although I sensed to a smaller extent. The domination of forest harvesting by village tobacco farmers using firewood for curing of their crops, as discussed in 6.3.4.1, also represents a relationship of patronage. The chairman of the VNRC in the research village is a tobacco farmer, and is also heavily involved with a tobacco cooperative between the villages of Kiwere and neighbouring Mgera. The chairman organises the permits for collection of the firewood and transport from the forest, which as mentioned before puts the tobacco farmers within the cooperative at an advantage over the other uses of the forest, particularly the charcoal producers who have been excluded from the project (see 6.3.4.1), creating a clientelist relationship. This relationship, however, extends beyond the tobacco farmers of Kiwere; a recent controversy had erupted at the time of fieldwork because the chairman of the VNRC had purchased a single permit for the farmers from Kiwere and those members of the cooperative who come from Mgera. As external users of the forest they should pay a higher price for access to these resources according to the management plan, but they had been charged the same rate as the Kiwere members of the cooperative (Interview P36). 7.4.4 Scales of Corruption In this thesis, I have focused on gaining depth of insight into the everyday, petty forms of corruption that contribute to the disappointing and unexpected outcomes from both WMAs and CBFM. Whilst the financial value of these mechanisms of corruption is far less than the rent-seeking behaviour and neopatrimonial systems apparent at the national and international levels between ministerial staff, wildlife hunting companies and timber companies (see also 5.4.1), in this section I discuss how local forms of corruption may be linked to further scales of corruption and important as a component within a broader system. In the wildlife sector, several studies have pointed out the opacity of the hunting block allocation system and the opportunities for rent-seeking this provides (Nshala 1999; Sachedina, 2008; Benjaminsen & Bryceson, 2012). Leader-Williams et al. (2009) describe how the liberalisation and neoliberal reform that began in the 1990s in Tanzania, alongside low public servant salaries, has produced a system of corruption around wildlife hunting that is endemic and systemic. They argue that rent-seeking surrounding hunting block allocation, the issuing of presidential licences for hunting to personal friends of high level officials and the allocation of hunting blocks based on the private financial interests of high level officials (many of whom own smaller hunting companies) pose serious constraints on the transparency 231 and revenue generation of the wildlife hunting industry in Tanzania. The roles of neopatrimonial forms of exchange and clientelist relations is clearly equally important at the national level as at the local, as discussed in this chapter. Similarly, in the forest sector, the hidden aspects of timber markets at the national level have been discussed by Milledge et al. (2007), who argue that there are huge degrees of revenue loss (up to 96%) related to undervaluation and illegal forms of exchange that line the pockets of individuals involved in networks of forestry-related corruption spanning the international to district levels. Milledge et al. (2007) argue that the expansion and globalisation of timber markets from Tanzania, alongside the opening up of new areas to timber trade through infrastructure developments (particularly through the Mkapa bridge in Southwest Tanzania) has been accompanied by the expansion of markets for collusion and concealed transactions. As in the wildlife sector, they cite neopatrimonial networks and personal and institutional relationships between timber exporters and senior officials74, mechanisms of bribery and rent-seeking behaviour as critical to national level forestry-related corruption. They also argue that fraudulent documentation and certification (particularly with respect to the authentication and certification of timber by ‘hammer’ marking of timer products by District Authorities) is an important component within these networks. There may be important links between these national scales of corruption and the politics of scales described in section 6.3. In the wildlife sector, the emergence of the AA consortium at the national level, and its specific interest in gaining political voice regarding hunting and block allocation decision-making (see 6.3.2.2) could indicate either an important challenge by members of the AAs across Tanzania to challenge the corruption complex surrounding hunting at the national level, or could represent the integration of these scales of corruption and the assimilation of local level actors into networks of corruption. The discussion in 7.4.3 of clientelist relationships and their impacts on decision-making regarding the MBOMIPA WMA indicate that this integration may be well underway, with increasing links between investors, the Board of Trustees and the leaders of the AA shaping investment agreements and neopatrimonial relationships, with real and material consequences on the WMA, the villages, and villagers that participate in it. In the forestry sector, the indication of widespread corruption relating to forest products, and particularly the involvement of District Officials in these networks, indicates a potential link with the politics of scales discussed in 74 Over half of the 28 timber companies they investigated had institutional linkages with senior Tanzanian officials (Milledge et al., 2007). 232 section 6.3.2.1. The interference of the District Authorities in local level electoral processes and their vested interest in keeping the same individuals in office within the VNRC was described by one respondent as due to the facilitation of corruption (see 6.3.2.1), indicating the incorporation of CBFM, and specifically project leaders at the local level into these networks. This is supported by Sachedina’s discussion of corrupt practice, in which he cites a district officer offering to teach a member of the village government the “tricks of fraud” (2008: 170). In both sectors, the scaling down of national level corruption networks and mechanisms, and the incorporation of project leaders into these networks provides further incentive for the control of information and reinforcement of power through discursive scalar construction described in 6.3.1, which serves not only to hide from view the local processes of the corruption complex, but also the links to wider networks and scales of corruption ongoing within the natural resources sector. In the next section, I discuss how the integration and maintenance of these networks of corruption need to be considered in the context of the socially-embedded mechanisms that support them. 7.5 Cultural Aspects of Power and Corruption “In a face-to-face society, the price of open conflicts is too high” Olivier De Sardan (1999: 30) This section discusses the performance of policy by considering the cultural context into which CBNRM has been implemented in the research villages. In exploring corruption as a part of a moral economy of production (see 2.1.3; Chabal and Daloz, 1999), the social mechanisms of corruption and “the value systems and cultural codes, which permit a justification of corruption by those who practice it (and do not necessarily consider it to be such – quite the contrary), and to anchor corruption in ordinary everyday practice” (Olivier De Sardan, 1999: 26-27) must be a research focus. In this section I address such social norms, considering them as a component of the tricks of the game that, to borrow the word from Olivier De Sardan, ‘communicate’ with corruption. Such factors are also important for adding depth of insight to the power struggles discussed in chapter 6 and to the aspects of power that could be involved with the issues explored in this chapter. Power systems within Tanzanian communities are clearly not limited to the politics surrounding the management of natural resources under CBNRM. Such politics are instead mapped onto existing systems, hierarchies and inequalities within the village context and 233 wider institutions (Hyden, 2005; 2008). Pre-colonial systems are mixed with colonial regimes and post-colonial developments to produce complex governance systems that make up the performed governance system that has been the focus of this chapter (Hyden, 2008). In this section, I attempt to highlight the key features of Tanzanian culture that I witnessed influencing and shaping the management of natural resources by communities. Participant observation in village meetings and time spent with villagers in both research villages supported a common observation of the prevalence of hierarchy and respect, both for elders (see Pepinsky, 1992) and for those with an official title, and the significance placed upon bureaucratic procedures and following protocol. This was also particularly apparent when dealing with administrative tasks related to conducting fieldwork in Tanzania. Discussions with other researchers and international NGO staff working in the study area also highlighted the important role of ‘accepting one’s place’ within the hierarchy and following the social norms surrounding dealing with official matters (Interviews P6, P7). Such procedures were very apparent at village meetings, where a strict protocol was followed regarding the agenda and management of meetings, and where demonstrating appropriate respect for Village Council officials was of paramount importance. Such demonstrations usually included addressing the person by their title, and using a discourse of deference with respect to those in power. Raised eyebrows from those holding office and uncomfortable shuffling and floor gazing from the wider community were noted when an individual failed to follow these rules. In the rare cases where I witnessed outright defiance from an individual towards those in power, this was quickly addressed by a high-ranking official within the village who would berate the outspoken individual for their poor manners and exert their authority over the meeting, insisting that all contributions were invited by the Village Officers. Whilst an important feature of the participatory processes within CBNRM, hierarchy and respect for those in power was not dictatorial in the research villages, and whilst speaking out of turn was severely frowned upon, the strict rules governing how and when a person may address a meeting also created space to ensure that all participants were given an opportunity to speak. The importance of hierarchy and respect for those who sit at the top of it was discussed by an academic researcher as contributing to problems within CBNRM because “It is difficult to challenge up the hierarchy, and this is exacerbated by the current set up of the institutions... Who has the power to remove the Chairman? Someone has to just come up. This requires courage” (Interview P7). The same respondent went on to describe how this 234 created problems when an individual working on the committees was deemed to be doing a poor job, because it can be very difficult to remove such a person (Interview P7). Disagreement with the leaders of the VNRC or WMA was usually voiced using specific responses that thanked the official for their contribution and then requested permission to contribute to their ideas, in an attempt to make suggestions rather than appear confrontational (Interview P7). Brockington (2006) previously identified a similar process taking place in Tanzania whereby the ‘weak speak to the strong’ through official discourses, for example in requests for increased accountability rather than direct accusations of corruption, and thereby employ tactics of manipulation and subversion, rather than conflict and open resistance. Avoiding conflict was seen as an important social skill, underlying which is a consideration of long-term relationships over immediate issues combined with expected reverence for those in power. Social norms can also play an important role when family members work together on a committee, making it inappropriate to disagree with a relation in such a setting (Interview P7). Customary rules regarding the participation of women in such meetings and their subordination to men also played a role in the meetings that I observed, where even when invited to speak, women were uneasy doing so. When discussed with groups of women in the village, one explained to me that she and her friends didn’t like to speak at the meetings in case they said something stupid. As we discussed this topic whilst cooking outside her house following a meeting earlier that day she said “for women it’s different. Here we can speak, but not there” (Interview P138). The gender aspects of participation are well-known (e.g. Guijt and Shah, 1998; Cornwall, 2003; Cornwall, 2004), and the policies for CBNRM in Tanzania include regulations to ensure a proportion of women are elected to serve on the VNRC and as WMA Representatives. The comments of women within the research villages highlight the fact that ensuring presence does not necessarily lead to meaningful participation, and that participation is still subject to multiple types of power relations within and around committees Aspects of power were clear in relation to the norms of participation and reverence for those in office. The protocols surrounding speaking publicly and in challenging the leaders of the VNRC and the AA could be seen as exemplifying the structuration and confirming- structuration of power, as described by Haugaard (2003), whereby the symbolic act of attaching meaning through following these rules confirms and strengthens the power structures they represent. I also observed that many villagers were unwilling to appear to be 235 complaining about individuals that hold office or the management of the forest and the WMA. One group of respondents explained that they could not discuss the problems they saw with the project “because it would be like saying we don’t want MBOMIPA” (Village Preliminary Survey P132). Taken together, these embodiments of power in the way people speak and act reflect the subjectification of villagers and the internalisation of values and norms that legitimise the power systems in place and the appropriate behaviour with respect to those in power. They form an example of governmentality (see Foucault, 1980). The research findings also support the argument that there is a strong spatial element to both participation and power, with the women I spoke to happier to discuss CBNRM, the committees and the problems they perceived with the management of natural resources away from men, but also away from spaces they perceived as the domain of men (see Kesby, 2005; Kesby, 2007). In this respect they demonstrated the internalisation of values regarding the subordination of women to men, reinforced these power systems through their behaviour at meetings. In the next section I bring together the discussions of rule-breaking and local systems of power, including neopatrimonialism and respect for officials to consider the emergence of big men within the performed governance systems of Tanzanian CBNRM. 7.5.1 Cultural Aspects of Corruption: The Tanzanian Big Man In combination with the clustering of power in the hands of a small number of individuals working on the VNRC and AA, both the CBFM and WMA case studies indicated a phenomenon similar to that of ‘big man’ politics (see 2.1.3; Sahlins, 1963). The concept of the big man has been applied outside of Melanesia, with parallels drawn in Africa, for a person who has achieved leadership status through wealth and personality and has created a large group of supporters (Brown, 1990). Here I argue that the personal power of such individuals, elected to positions of power within the VNRC and the AA, are combined with the authority of public office in Tanzania, and the culture of respect and reverence discussed above. A key aspect of the ‘big man’ concept is loyalty, which has also been noted in Tanzania in the dangers of speaking out against village authorities and the tensions between Village Officers, whereby “the council was not a balance to the chairman’s powers” (Brockington, 2008: 118). The demands placed upon myself and my research assistants when dealing with the leaders of MBOMIPA supported the concept of them acting as big men, and I argue that this is more pronounced in the WMA than in the CBFM case study because of the scalar topographies of power associated with each. In the WMA, the scaling of governance processes to the level of the AA (and the uncomfortable relationship with the 236 Village Government this produces), alongside the large sums of money that are managed by the AA, and the direct links formed between the leaders of the AA, national level investors and influential businessmen that make up the Board of Trustees have contributed to this. The concept of big man politics is present not only in the respect that must be shown to the leaders of the projects, particularly in the WMA, but also in the decision-making processes ongoing within the VNRC and the AA. As discussed in section 6.2.1, there is a clear concentration of power, knowledge, training, and to a certain extent benefits with those who are elected on to the VNRC and to serve as MBOMIPA Representatives. Furthermore, within the committees such benefits are gathered at the level of the leaders in particular, and the pattern of decision-making often follows this. Powerful individuals holding the most prestigious and powerful roles within the committee often take on the responsibility of making management decisions, without necessarily following participatory mechanisms with the wider committee (Interview P7). The status of such individuals makes it difficult for such a situation to be challenged, regardless of the written management agreement for the forest or WMA. Further to the instances of rule-breaking and deviations from policy described in 7.2, the performed governance systems indicated that participatory processes of decision-making in both the AA and the VNRC (see 5.2.1 and 5.2.3) were often not followed, and decision- making power was commonly described as concentrated in the hands of the leaders of the VNRC and the AA, resulting in decisions being made by these same few individuals without consultation or with retrospective consultation with other members of the AA, VNRC or the wider communities (e.g. Interview P7). One MBOMIPA representative voiced his frustration that “an investor can send a letter of application and then you hear that the leaders have sat to discuss and have decided that the investor is suitable without consulting with other representatives” (Interview P171). Other MBOMIPA representatives complained that “there are times when the leaders of MBOMIPA do not implement decisions that have been agreed by the Chairmen of the villages” (Interview P170), and explained that “it’s possible that [decisions are taken by the leaders alone] because of the geography of the area and the difficulty of meeting with the other representatives if such a situation is shown to be an emergency. Other times you discover that one of the leaders makes a decision for himself, not according to the rules of the association” (Interview P174). Narratives and explanations for individuals acting on their own commonly drew on ideas of necessity and the large distances between the WMA villages making it infeasible to bring MBOMIPA 237 Representatives together for all decisions. These politically-neutral explanations also served to reify the power structures within the AA, positioning the leaders of the AA as legitimately able to make decisions on behalf of the WMA in its entirety. Here there is clearly a link to the protection of power, the discursive construction of scale and the control of information (see 6.3.1) which, together with the social norms protecting the social standing of individuals in office, served to promote and maintain the big man status of the leaders of the VNRC and the AA. 7.6 Summary In this chapter I have discussed the hidden aspects of the performed governance systems in Tanzanian CBNRM. These hidden aspects are made up of ‘rules of the game’, which do not necessarily follow the governance systems prescribed in policy, but are seen to incorporate ideas of moral economy, reciprocity and social norms, alongside ‘tricks of the game’ which are processes and mechanisms by which actors can get around the rules of CBFM and the WMA and the sanctions that they might face. It is clear that deviations from the prescribed governance system are plentiful, especially in instances of rule-breaking by numerous groups both within and outside of the research villages. There are also significant issues with enforcement of the rules which both facilitates rule-breaking and is caused by deficiencies within the prescribed governance system. The patrol systems in place within the CBFM case study highlight these issues and currently represents a system in which patrolling, as outlined in the management plan for the forest, is unrealistic for the scouts, thereby promoting rule- breaking within the forest, as people do not expect to be apprehended, but also facilitates systems of bribery when they are. Power is critical to understanding the hidden aspects of the governance systems being performed in the case studies. The common belief that leaders of both the VNRC and the AA benefit from illegal use of the forest and wildlife resources they are elected to manage, and the funds generated from their management, generally goes unchallenged within the case studies, and in many cases has been accepted as inevitable. This highlights the ways in which such individuals hold protected positions of power within the research villages, shielded from investigation by the control of information and social norms that demand respect from the wider community, although still subject to accusations. I have argued that the leaders of the VNRC and the MBOMIPA representatives generally hold important social positions within their own villages, characterised by wealthy individuals who employ many other villagers 238 and are large personalities within the community, or so called ‘big men’. Such status may be a contributing factor to their election to the VNRC and the AA, which consolidates their power and status within the governance systems of CBFM and the WMA. The case studies also highlight the ways in which such social norms communicate with corruption, normalising the seizing of opportunities and expectation of ‘perks of the job’ for those elected to serve on the VNRC and the AA. The examples of clientelism discussed in this chapter also highlight how such relationships are facilitated by the communal, rather than individual culture in Tanzania, and the importance of good neighbourliness, fostering good inter-personal relations within working relationships and the practice of gift-giving. These are crucial aspects of the socio-political processes within the performance of policy, which are hidden and unrecognised by the prescribed governance system, but which have important impacts upon the governance systems in place and the distribution of benefits resulting from the WMA and CBFM. I argue that the status of the leaders of the VNRC and the AA and their ability to benefit from these positions is both a product of their protected social positions, but also a process of governmentality in which villagers have internalised and automatically confirm-structure the legitimacy of their authority and the normality of the processes through which they benefit from these positions (see also 6.3.1). This does not mean that these processes are blindly supported, however: instances of the corruption complex at work within both case studies are a common source of complaints within the village, and serve as a focal point for resistance to the power regime, but, like the form of power they involve, these forms of resistance are subtle. Villagers often speak privately of the injustices they see within the governance system, but in public they call upon official discourses of accountability, rather than conflict- inducing accusations. This suggests support for the ‘public and hidden transcripts’ described by Scott (1990). In the case studies the public transcript was used not only to appear acquiescent with the regime, but also as a vehicle to address the issues within the hidden transcript, where this was too politically sensitive to be voiced directly. Officials and the wider community are caught in a delicate discursive game therefore, in which the politics of natural resource management go undisclosed, and non-confrontational discourses are invoked by both the leaders of the project, who hide aspects of the performed governance system within them, and also by those seeking to challenge the power system in place, who utilise the official discourse to confront (without ever appearing to do so) the prescribed governance system, the power system it involves and the ways in which it is being performed. 239 The nature of power in the case study projects of CBNRM involves both the prescribed and performed governance systems, and the intricacies these entail. As described in chapter 6, these power systems are not static, but involve constant manoeuvres by actors as part of the politics of natural resource management. The hidden aspects of governance and especially the tricks of the game discussed in this chapter highlight some of the mechanisms and processes by which such manoeuvrings take place and the ways in which the prescribed governance system is mapped onto and melded around pre-existing power systems and social practices. Power is at work in the subjectification of people in both the case studies in several ways. As described in 6.3.1, the protection of power and strengthening of the power exercised particularly by the leaders of the VNRC and the AA is evident in the control of information shared with the wider community within each of the case studies. I argue that this situation, whether strategic or not, serves to legitimise the power of those elected to serve and reinforces the hierarchical nature of power and respect for official status described in this chapter. Both processes drive the subjectification of the wider community, instilling values of acceptance, legitimacy and necessity of those individuals (who possess the appropriate knowledge, training and experience), to govern the Village Land Forest Reserve and the WMA on their behalf. Concurrently the process of subjectification whereby the corruption complex described in this chapter has become socially normalised, especially with reference to the ‘perks of the job’ expected for leadership positions, legitimises and removes obstacles to the capture of benefits and misappropriation of funds apparent in both case studies. Members of the community are automatically confirm-structuring these systems of power, having internalised such systems as in their best interest and the inevitability of some ‘leakage’ within the system. Ordinary villages are not blind to these processes however, which raises interesting questions about the nature of power and resistance within these performed governance systems. It is clear from the findings in both this chapter and in chapter 6 that power and resistance are complex systems within the case studies. Resistance could be characterised as the ways in which actors attempt to reconfigure power for their own benefit, particularly in the struggles and conflicts described in 6.3.4, and in the destabilisation of power systems through political accusations of bribery as described in this chapter. I find that resistance is not always present or, at least, not always presented by everyone. Instead, the performed governance systems in Tanzanian CBNRM indicate a delicate balance of the protection of power, subjectification to 240 that power, alongside processes of resistance to it. The processes of resistance described within the politics of scales and the accusations made against elected officials within the project do not necessarily represent attempts to dismantle the power system completely, but instead are often attempts to re-shape the configuration of power in order to access and benefit from it. Chapter 8: Conclusions 241 Previous studies of CBNRM in Tanzania have highlighted many issues, both with policy and with the outcomes witnessed. The large discrepancies in the outcomes from CBNRM across the country (Lund, 2004), the presence of elite capture (Lund and Treue, 2008; Ministry of Natural Resources and Tourism, 2009; Nelson et al., 2009), negative impacts of CBNRM (Meshack et al., 2006; Igoe and Croucher, 2007), concern over revenue systems in both Wildlife Management Areas (WMAs) and Community-Based Forest Management (CBFM; Walsh, 2000; Meshack et al., 2006; Mustalahti, 2007) and poor implementation that has diverged from policy (Ministry of Natural Resources and Tourism, 2009; Mustalahti and Lund, 2009) have all been previously noted, although little explored. In the wildlife sector, concerns over the ways in which power has been devolved to local communities for managing natural resources, and especially hunting systems, within Wildlife Management Areas (WMAs) has been a popular topic of debate and concern (Leader-Williams, 2000; Nelson, 2007; Nelson et al., 2007; Leader-Williams et al., 2009; Nelson and Blomley, 2010). Many of these issues are acknowledged in the recent ministerial reviews of both WMAs and CBFM around the country (Institute of Resource Assessment, 2007; Ministry of Natural Resources and Tourism, 2008; Ministry of Natural Resources and Tourism, 2009; Health and Development International Consultants (HDIC), 2010). This thesis has been designed to meet the challenge previously set by such studies to consider these issues in more depth and to investigate the socio-political processes underneath these phenomena (see Igoe and Croucher, 2007; Blomley et al., 2008b; Blomley et al., 2008a). In this thesis I set out to take a specifically socio-political lens to the study of Community- Based Natural Resource Management (CBNRM) in Tanzania. In the preceding empirical chapters I have done this through an investigation of three over-lapping areas: the processes of policy reform and the prescription of policy; the performance of policy as seen through power structures and the struggles that take place around them; and the performance of policy as including hidden aspects of power and governance. These three areas bring together the context in which CBNRM has developed and is being implemented in Tanzania, and theorises CBNRM as an integration of socio-political processes of politics, power and governance that take place across levels ranging from international and national level policy formulation through to the micro-politics of natural resource management at the local level. I argue that an understanding of the outcomes from CBNRM and the issues described by other researchers as discussed above cannot ignore the role of such micro-politics, but also cannot consider them in isolation from the political-economic-social context in which CBNRM 242 developed as a national policy, and the socio-political forces that shaped it. For example, as discussed in chapter 7, the problems with enforcement and the processes of rule-breaking within both the case studies were the result of the influences of both the prescribed governance system, which created disincentives for patrolling and enforcement, and were also partly facilitated by the moral economy of local communities in Tanzania, where cultural aspects of power and corruption also shaped enforcment and mechanisms of rule-breaking within the performed governance system. In investigating how the reality of CBNRM in Tanzania is different from that prescribed in policy, my first conclusion is that the two exist in tandem, sometimes in conflict and always under processes of change and struggle. The performed governance system is an amalgamation, and occasionally conflict, between the power systems prescribed in policy and local contexts of power. The socio-political processes that take place around these contexts of power are partly outside of the prescribed governance system and operate within an informal realm, often hidden from view and not discussed openly, but they are present in reality and cannot be explained simply as poor implementation. Closely linked to this, my second conclusion is that the ‘how’ of governance processes is inescapably socio-political and represents a tangled mix of power-driven processes. Power is a central driver in the process of policy reform and the prescribed governance system that results from the politics of natural resource management, power devolution and the authority to manage natural resources and revenues that can be generated from them. Power is also at the heart of the performed governance system and the struggles taking place as actors try to benefit from the governance and power systems, and shape them according to their private interests. In this concluding chapter, I draw on the insights from the three empirical chapters to address some of the over-arching questions set out in chapter 1 through an examination of three socio-political processes cutting across these issues and my empirical chapters: politics, power and governance. 243 8.1 Politics, Power and Governance: Empirical and Theoretical Research Contributions 8.1.1 A Comparative Perspective of CBNRM in the Wildlife and Forestry Sectors In this thesis, I have attempted to broaden the discussion of CBNRM in Tanzania by bringing together the concerns over power devolution, elite capture and the unequal and unexpected outcomes from CBNRM in a comparative study of WMA and CBFM policy prescription and performance. There are important similarities between the developments of these policies, particularly relating to their objectives, the timescale of their development and the role of external organisations in agenda setting and policy formulation. These twin policies of CBNRM in Tanzania have followed distinct but parallel trajectories in both agenda setting, policy development and the assessment of predicted ‘win-win’ outcomes. In this section, I bring together the findings concerned with the distinctions between the two sectors and the underlying processes that have led to these distinctions and the differences in outcomes that are witnessed. It is clear that this process of political reform is a complex context of competing agendas and discourses and differential politicisation of natural resources in the forestry and wildlife sectors. In chapter 5, I examined the process of policy reform that introduced CBNRM as a national policy in Tanzania in both the forestry and wildlife sectors. One of the three original foci of my research was to explain the distinctions between the prescribed governance systems in place in CBFM and WMAs. Three of these differences are key: Firstly, the ways in which power is devolved to the local level within WMAs and CBFM, and the different configurations of power that result; secondly the application procedure to gazette a WMA and a VLFR; and thirdly in the revenue arrangements and mechanisms of distributing funds between the national and district levels, and amongst the individuals participating in CBNRM in their villages. The divergences between the policy pathways in the wildlife and forestry sectors manifest in the prescribed governance structures, which devolve power in distinct ways. The wildlife sector and the introduction of WMAs represents a much more cautious devolution of power and involves a much larger role for the district and national levels within the governance system for a WMA. Furthermore, WMAs demonstrate the recentralisation of power through restrictions and bureaucratic obstacles within power devolution within the original policies that introduced WMAs and subsequent amendments to policy that set out further constraints 244 on the autonomy of the local level with regards to the generation and collection of revenues from WMAs. However, the example from the forestry sector in Tanzania indicates that state devolution of power to local communities is not inevitably restricted wherever possible. The introduction of CBFM aligned with the interests of the Forestry and Beekeeping Division and presented an opportunity to increase the area of land under a category of Protected Area without threatening the material interests of the policy community. There is clearly a territorial component to the devolution of power over natural resources. In the forestry sector, where this does not threaten the interests of the state and its officials, and offers an opportunity to expand the area under a category of PA (over which the state has some advisory capacity at least), it is more broadly accepted and implemented. Where such devolution effectively removes territory from under the control of a state organisation, as with WMAs in Tanzania, and threatens officials’ networks and systems of personal gain, devolution is much less forthcoming, highly restricted and politically contested. The case studies indicate that the politics of power devolution do not always represent attempts by the state to restrict devolution wherever possible, but are a complex interaction between different forces shaping natural resource use and their management across different levels. The power to control areas of land, make decisions about and benefit from natural resources drove the politics at work in both sectors throughout the processes of policy reform and creation of the prescribed governance system, and continues to drive the politics of their performance. The differential politicisation of natural resources in the wildlife and forestry sectors is key to understanding the different policy pathways adopted, the distinctions in power devolution, alongside the attempts at recentralisation that have emerged, and the patterns of winners and losers witnessed in both WMAs and CBFM, therefore. The economic value of wildlife resources, particularly through the hunting system in Tanzania and the opportunities for rent-seeking by national level officials and ministerial authorities within an opaque block allocation system was a driving force behind the processes of controlled devolution and recentralisation that have been witnessed in WMAs. In contrast, the comparatively low value of forest resources in CBFM projects has facilitated the more rapid and full devolution of power in the prescribed VLFR governance system. Through the high value of wildlife resources, they have become increasingly politicised and a focus of power struggles as the national level continues to seek opportunities to re-capture control and revenue associated with both hunting activities and other forms of revenue generation within WMAs. The key role of natural resource politicisation is also highlighted by the emergence 245 of conflict and resistance to power devolution in CBFM when high value forest resources are involved (as discussed in 5.4.2), and also at the local level once the economic potential of forest resources becomes clear, sparking conflicts and power struggles within the community (see 6.3.4.1). The importance of the differential politicisation relating to wildlife and forestry resources been crucial in shaping the governance structures prescribed in policy and generating the distinctions between WMAs and CBFM described above. These determining factors also have ongoing effects in shaping the patterns of winners and losers within Tanzanian CBNRM and the ways in which the governance structures are being performed and contested. 8.1.2 Winners and Losers in Tanzanian CBNRM The politics of power devolution surrounding wildlife resources have created a governance system in reality where the authority of the local level and financial benefits of CBNRM are restricted. This has clear implications on the patterns of winners and losers from WMAs, whereby a significant proportion of revenue is being retained at national and district levels, and local level empowerment is strictly controlled. My research has highlighted that the patterns of winners and losers in both WMAs and CBFM are a product of these prescribed governance systems and the scalar configuration of power that they produce, alongside the political struggles over natural resource management taking place within and between the different levels of administration. As discussed in chapter 6, the scalar configuration of power set out in policy for both WMAs and CBFM concentrates power devolution at the local level into the Village Natural Resources Committee (VNRC) for CBFM and the Authorised Association (AA) for WMAs. The concentration of benefits is correspondingly scaled to the level of these institutions, taking place within the village level for CBFM and at the inter-village level for WMAs. In both case studies the concentration of power within these organisations was associated with a lack of empowerment and low levels of knowledge concerning the WMA or the VLFR, but this was especially pronounced in the WMA, where village residents were excluded through the scaling of power and governance processes to the inter-village level. Furthermore, within these organisations, the capture of benefits by a small number of individuals elected to serve in leadership positions was consistently identified. These individuals benefit from the allowances that they are entitled to from carrying out their duties, 246 but also benefit from the training provided to carry out these duties. The elite capture of benefits was further maintained through the restriction of information sharing and the lack of ongoing training, by which these individuals were successfully protecting their positions within the governance systems, and the benefits they enjoy from this. The fundamental relationship between power and knowledge and the concept of power as a political technology is very clear in both the case studies, therefore. The use of this political technology is intrinsically tied to the social construction of subjects and the internalisation of values that legitimises and protects the system of power in place, and the elite capture that results from it. Elite capture and its acceptance by the wider community as both inevitable and acceptable (to an extent) is also closely tied to local contexts of power within the case studies. The role of neopatrimonial systems of power which surround the leaders of the VNRC and the AA, and the internalisation of one’s rights to ‘perks of the job’ when in positions of power are vital to the maintenance of these processes of elite capture. Actors continually try to re-position themselves to benefit from the management of natural resources, employing power as a political technology, tricks of the game and politics of scales to do so. Such tactics are generally accepted and internalised as both legitimate and inevitable. 8.1.3 The Reality of Community-Based Natural Resource Management Power is not only exercised and constructed in the prescription of policy and the formal governance structure set out, but is adapted around existing systems of power (within the local community and between these communities and other levels operating within the WMA/VLFR), and is re-configured by the actors involved across these levels. These processes of integration with previous arrangements and the re-configuration of power systems that is constantly taking place are largely responsible for the differences seen between the prescribed and performed governance systems. This is not a case of poor implementation, but one of the unrecognised or unanticipated socio-politics surrounding implementation. The devolution of power to manage natural resources to the local level within CBNRM necessarily involves changing relationships of power between the state and the local level, although subject to the politics of this change as described above. CBNRM also introduced changing power relationships at the local level however; the introduction of the Authorised Association (AA) to manage a WMA has created new scalar configurations of power with 247 respect to natural resources and altered the roles of the Village Council dramatically; similarly in the forestry sector the introduction of CBFM has shifted relationships of power between the Village Council and the Village Natural Resources Committee (VNRC). CBNRM is introduced at the local level within an existing context of power, and the performed governance system emerges as an integration and struggle between these two systems of power and the shifts described above. The evidence from both case studies clearly shows that the performed governance system and the power systems within them are under constant struggle and re-negotiation as actors try to position themselves and manipulate the projects for their own interests. These struggles are a key feature of the politics of scales taking place within Tanzanian CBNRM and an important shaping force underlying the patterns of winners and losers within both the WMA and CBFM case studies. These processes of scalar transformation shaped people’s participation within the projects, the election processes within CBFM and the revenue opportunities available to the village. I argue that a scalar perspective makes it evident that power is at the centre of these conflicts. Whilst the nature of conflicts is specific to the case studies and the natural resources sector in which they operate, the identified struggles all converge around the locus of power at the local levels and the distinct features of the topographies of power in each case. I have referred to this as indicating the same politics at different scales within the politics of scales and processes of contestation of power. I have employed a definition of governance in my research that acknowledges the roles of both formal and informal means to creating a management system (see 1.4). It is clear that the implementation and reality of CBNRM in Tanzania as seen in the performed governance systems, and their divergence from that prescribed in policy, is constantly being shaped by informal processes and de facto rules of the game. CBNRM is initiated within a context of existing local contours of power and socially accepted norms and rules. Neopatrimonial systems of power remain hugely influential and are re-shaping the governance systems, their performance and the outcomes of CBNRM. The unequal distribution of benefits and presence of elite capture are widely lamented within Tanzanian CBNRM, and both the case studies demonstrated these processes in action with little challenge or resistance from the wider community. Those that benefit most from CBNRM at the local level are consistently identified as those that hold positions on the committees charged with decision-making and natural resource management on behalf of the community. These individuals hold positions of status within the communities and enjoy many ‘perks of the job’. I have likened them to 248 the ‘big man’ in politics (Sahlins, 1963), whereby individuals construct power through the creation of a network of supporters and the ability to redistribute resources to these supporters. I have argued that the deviations from policy and the unequal distribution of benefits from CBNRM witnessed in the case studies were largely unchallenged due partly to these neopatrimonial networks in place around the leaders of the VNRC and AA (see 7.4). The privileged and protected position of these big men is also due partly to the social norms and culture of respect surrounding such positions. Furthermore cultural factors are important to understanding the differences between the prescribed and performed governance systems, especially in terms of the acceptance of unequal distribution of benefits from CBNRM. Tanzanian culture respects and rewards the individuals who play the game successfully. These values are internalised by members of the community, who accept the right of those that are able to benefit in whatever way that they can, and simultaneously their own right to benefit in such ways if they can position themselves to access such benefits. The power systems and social norms shaping the performed governance system are continually undergoing processes of structuration and confirm structuration, serving to legitimise the status of the leaders of the VNRC and the AA, and justifying the processes and maintenance of elite capture. In my research I have focused on the devolution of power within CBNRM and the struggles this invokes. My research also highlights features of resistance however. The research findings with respect to resistance are complex; whilst the politics of scales discussed clearly indicate that power is clearly being contested and re-configured in both WMAs and CBFM, the discussion above indicates that power, and the inequalities it supports, often go unchallenged. Power and resistance are neither uniform nor universal, but are woven throughout the socio-political fabric of everyday life. It is not a case of power either being totalitarian and imperceptible, or always subject to resistance (however subtle it may be). Instead, power is at times invisible and internalised and in other contexts it is resisted, and this resistance is itself bound by the system of power. For example, I have argued that the case studies indicate that power is rarely questioned or resisted in an outright fashion. The conflicts taking place within both case studies represent a form of defiance and resistance, but they do so within the confines of the performed governance system, and in ways that reduce the confrontational manner of these conflicts, as dictated by socio-cultural aspects of power within Tanzania. In place of outright challenges, the politics of scales perspective used in 249 this thesis shows that processes of active destruction were used to destabilise scalar configurations of power (Fraser, 2010). As discussed in 2.4.2, these processes of scalecraft indicate the inter-relatedness of structure and agency as actors organise themselves and their agency around the structures in place. 8.1.4 The Discursive Process of Policy Reform A final theoretical and empirical contribution made in this thesis concerns the discursive process of policy reform, and particularly the role of the local level in these negotiations. The prescribed governance systems that set up the differences between WMAs and CBFM described in this research were formed from influences across levels, including Transnational Conservation Organisations (TNCOs), the state, key individuals within the policy community and actors at the local level. The different policy pathways adopted in the forestry and wildlife sectors, and that underpin WMAs and CBFM in Tanzania result from the preferences, values and beliefs of these actors and groups, alongside multiple political- economic factors that drove policy processes. My research supports and contributes to the body of work that has begun to highlight the political economic process of policy reform in Tanzania, therefore (e.g. Nelson and Blomley, 2010), and has expanded this discussion to consider the discursive processes that took place within agenda setting and policy formulation. Furthermore, the research presented in this thesis has highlighted the role of the local level in these policy processes, particularly the influence of the REWMP/MBOMIPA projects in the wildlife sector. Such roles of the local level and the influence of donor-funded pilot projects are often left un-discussed, but here I have emphasised their potential roles in the co-construction of policy with the national level, and especially the policy community. It is also apparent that the role of the local level in policy processes is an invited one, and their position within an institutionalised discourse coalition does not guarantee influence in policy formulation, and that this is subject to power relations between the policy network, the discourse coalition and the policy community. Policy space is both constructed by individuals and restricted by them according to these power relations and the discursive processes that surround them, therefore. The discursive nature of policy reform is also clear: Policies are, after all, created by people, all of whom approach the process of policy reform with their own understandings, priorities and motivations. The resulting policy pathway and prescribed governance structures is an embodiment of those actors’ success at discursively negotiating policy processes within the context of driving forces that surround natural resource management (or any policy issue). Pilot projects are not only capable of influencing 250 policy reform processes however; they can also be shaped by them. In 5.5 I argued that strategic positions within a policy network make pilot projects useful to the policy community, and the discursive processes of policy formulation they are engaged in, by providing an opportunity to witness policy options in practice. The REWMP/MBOMIPA example highlights that pilot projects can play a role in shaping policy processes, but also the substantial impact they may have on pilot projects. 8.2 Practical Research Contributions, Implications and Future Directions In this thesis, I have focused specifically on the complexities and struggles taking place within case study CBFM and WMA projects. I have argued that it is vital to understand the ways in which such policies and projects are performing differently to what is expected, and I have attempted to delve into the messiness, highlight the loopholes, inequalities and undesirable mechanisms and processes that are shaping the case studies. What emerges is a picture of CBNRM in Tanzania that is fraught with struggle and tactical manipulation, and where actors across levels are engaged in a ceaseless game to benefit from natural resources, often at the expense of others. I believe this to be true, and I argue that such processes are inherent to CBNRM, but I also argue that they should be taken in context. The two case studies discussed in this thesis are both well-regarded examples of WMAs and CBFM within Tanzania. They are successful in many ways: they both have a long history of generating revenue at the local level; they have both been in operation for over a decade and have successfully transitioned from donor-funded projects to implementing national policy; they are both fully authorised and functioning examples of WMAs and CBFM that have operated under community authority for at least five years and do not appear to be in danger of collapsing in the near future. The longevity of these projects and their continuing presence should be recognised as an achievement. My research has been at pains to point out their imperfections and inadequacies, but they function, in unexpected and unequal ways, but function nonetheless. I have not focused on the biophysical elements of these CBNRM projects, and I have not attempted to investigate the conservation outcomes witnessed in these projects. This is an area for rewarding future research and collaboration with other research projects in the area75. However, the findings of this research with regards to rule-breaking and corruption, and 75 See the Ruaha Carnivore Project (www.ruahacarnivoreproject.com) and The Wildlife Connection (www.thewildlifeconnection.org), accessed 07/04/13 251 particularly the potential integration of local level actors involved in CBNRM into regional and national networks of illegal resource exploitation and personal gain, do raise important questions for conservation outcomes. Local villagers argue that wildlife and the forest are benefitting from CBNRM, hunting is reduced, illegal activities are curbed and the impacts of this are seen. However, my research has highlighted the unequal distribution of costs and benefits being experienced by local residents, which may lead to disillusionment and destructive environmental practices as people abandon the project and its rules, as previously discussed by Gillingham (1998) and Songorwa (1999). The discourse of CBNRM argues that conservation outcomes from both WMAs and CBFM are twinned with the development outcomes that are achieved. I have highlighted the inequalities and often disappointing results for development in both case studies, which suggests that conservation objectives are jeopardised by the complexities discussed in my research. I am hesitant to suggest that the socio-political complexities of CBNRM discussed here should be taken to show that such policies and projects are failing. I have maintained that these projects indicate that CBNRM in Tanzania is working in unique and evolving ways, and I argue that the patterns of both development and conservation outcomes can be expected to follow similarly intricate and dynamic paths. Future research that could link the socio-political complexities discussed in this thesis with analysis of the conservation status of the MBOMIPA WMA and Kiwere VLFR would be highly beneficial. The intrinsically political and power-driven nature of natural resource management, policy reform and CBNRM are not new themes within political ecology (see Forsyth, 2003; Robbins, 2004; Adams and Hutton, 2007; Blaikie and Springate-Baginski, 2007). The objective of my research in this thesis was to carry out an in-depth investigation into the socio-political processes at work within CBNRM in Tanzania, and to provide insight into the outcomes and inequalities that are witnessed in WMAs and CBFM. In doing this, I have brought together several different bodies of literature, connected through the theme of power, to provide a holistic perspective of the prescribed and performed governance systems in place within a WMA and a VLFR. I have used political ecology as an approach, placing power at the core of these socio-political issues and drawing on these further areas of study to gain depth of insight into the processes at work. I argue that a political ecology approach, and particularly the freedom this provides to integrate further areas of research, offers the opportunity and highlights the necessity of serious consideration of the complexities of such socio-political processes. 252 There is ongoing debate within and around political ecology as to how it can, or whether it should be useful (see Vayda and Walters, 1999; Walker, 2005; Walker, 2006; Forsyth, 2008; Blaikie, 2011). Below I argue that the political ecology approach implemented within this thesis has highlighted several ways in which recommendations can be made to improve the problems associated with both WMAs and CBFM in Tanzania. However, I also argue that the biggest contribution made by this kind of research is not the adjustments to policy that may come out of such research, but the recognition and attempts at mainstreaming the importance of investigating the political nature of such issues. As I stated in chapter 1, understanding the outcomes, successes and failures of CBNRM is simply incomplete without attempting to understand the ‘how’ of governance processes (Büscher and Dressler, 2007). Whilst it is perhaps easier to create new forms of devolved environmental management, new markets for environmental commodities (fictitious or not) and new revenue generation schemes, such technocratic fixes will always be subject to similar power struggles, political processes and shaping forces that have been witnessed with CBNRM. We must learn to accept the messiness, not as poor implementation, but as characteristic and a permanent feature. There are always adjustments that could improve the outcomes of CBNRM in Tanzania. My research has highlighted the concentration of power, opportunities for training and benefits in the hands of a small group of individuals elected to serve as leaders of the AA and the VNRC. Creating a rule that the same individual cannot serve in these positions indefinitely, and that training must be provided so that newly elected members can carry out their duties effectively might alleviate this issue and spread the benefits of such roles to a wider group of people. It is important to note, however, that such ‘tweaks’ to policy and the rules of the game would only spread these benefits so far within the participating communities, and are not a silver bullet to the problem. Secondly, the prescription of a new rule does not guarantee its performance as envisaged. As is clearly indicated by the MBOMIPA example, stipulation of the proportion of revenue that should be distributed to participating villages does not prevent divergence from this figure, and constitutional decree of the equal sharing of revenue between all participating villages does not prevent attempts to change this (see 6.3.4.2). Similarly, prescription of the rules for sanctioning offenders within CBFM does not mean that different rules which adhere to a moral economy or which facilitate bribery will cease to have influence (see 7.2 and 7.3). Attempts to increase the remuneration available to forest scouts is often advocated as a solution to bribery and enforcement problems within CBFM, 253 but as discussed in 7.4.1, these problems rest not simply on the poor wages and harsh conditions experienced by forest scouts, but also upon complex cultural factors that uphold practices of accepting a bribe for turning a blind eye and encourage the seizure of opportunities to improve one’s own situation, and that of those who depend upon you, wherever possible. I argue that an ongoing process of ‘tweaking’ is advisable to continually monitor the devolved governance of natural resources. The performed governance system is not static, but is constantly shifting as actors adapt to the de jure and de facto rules. There is no optimal configuration of CBNRM governance systems, or end point to aim for. CBNRM requires constant attention to the power struggles and manoeuvrings within the performed governance system to understand and address the imbalances that emerge. Similarly I would argue that the issues and inequalities within Tanzanian CBNRM cannot be resolved by expurgating all forms of local level corruption from within the governance systems being performed in WMAs and CBFM. This research has highlighted the complex cultural foundations of practices that others may classify as morally corrupt, but which can be alternatively viewed as adherence to a different set of rules, accepted and upheld by actors. My research is closely aligned with efforts to investigate the ways in which development efforts can work alongside and through local contexts of power and political systems, such as neopatrimonialism (Kelsall et al., 2010). The further depth of understanding demonstrated in this thesis as to how these systems of power relate to the management of natural resources within CBNRM in a Tanzanian context opens up a potentially fruitful area for further research that draws together my own research area with questions about how to work with, and get the most for development objectives, from such systems of power. A final core finding of my research is the constantly shifting nature of political struggles within devolved natural resource management. In 6.3 I discussed the politics of scales ongoing within the case studies of CBFM and a WMA. These struggles take place as actors try to position themselves to benefit from CBNRM in Tanzania, and it is important to note that they have emerged in response to, and around the configurations of power within CBFM and WMAs (see also Smith, 1992). What these scalar transformations and conflicts highlight is that governance systems and the power systems within them are not static, and that actors are engaged in constant struggle. Whatever changes are made to the prescribed and performed governance systems, these processes of struggle will remain, and new forms of conflict and scalar re-configurations will emerge. As discussed in 2.2.3, policy reform does not begin with agenda setting, move through processes of policy formulation and end with 254 implementation, but is an ongoing process of negotiation and development (Keeley and Scoones, 2003). Further research into the case studies discussed in this thesis might trace the shifting politics that continue to feature in both Kiwere and MBOMIPA as the AA and VNRC go through subsequent elections, new AA investment contracts are signed and the policy environment begins to concentrate seriously on the opportunities presented by REDD in Tanzania (see Burgess et al., 2010). The adoption of REDD in Tanzania may also present an interesting opportunity to investigate further comparisons between the different types of market-based mechanisms for local development within Tanzania. Finally, future research might expand the discussion in 5.5 regarding the role of different non-state actors in policy processes and the politics of scales discussed in 6.3. The role of the Hunting Association of Tanzania in policy processes, for example, is yet to be explored with respect to the development of the 1998 Wildlife Policy for Tanzania. Future policy changes and amendments to the prescribed governance structures within WMAs may also emerge through the ongoing roles of TNCOs (and particularly WWF; see 6.3.2.2), and their roles in the initiation and shaping of the National Consortium for AAs is a potentially interesting field of research. The politics of scales and conflicts that might emerge in Tanzania through such an organisation at the national level may have profound impacts on the devolution of power within WMAs, particularly with respect to the hunting revenues distributed by the Wildlife Division, which has been an ongoing debate for several years (as discussed in 5.2.4; see also Shauri, 1999; Mniwasa and Shauri, 2001; Nelson et al., 2007). All of these topics might produce interesting future research that continues the contribution of this thesis to dedicating research efforts to deepen understanding of the complexities of CBNRM, both in a Tanzanian empirical context, and in a theoretical context by outlining further aspects of the role of power in shaping policy processes and its continual performance. Both of these contexts will benefit from continued efforts to put politics and power first. References 255 Adams, W. M. (2004), Against extinction: The story of conservation, Earthscan, London, 311pp Adams, W. M. (2009), Green development: environment and sustainability in a developing world, Routledge, 445pp Adams, W. M. & Hulme, D. (2001a), If Community conservation is the answer, what is the question?, Oryx, 35, (3), 193-200 Adams, W. M. & Hulme, D. (2001b), Conservation & community: Changing narratives, policies and practices in African Conservation, IN Hulme, D. & Murphree, M. W. (Eds.) African Wildlife and Livelihoods: The Promise and Performance of Community conservation. James Currey Ltd., p.9-23 Adams, W. M. & Hutton, J. (2007), People, parks and poverty: Political ecology and biodiversity conservation, Conservation and society, 5, (2), 147-183 Adams, W. M., Aveling, R., Brockington, D., Dickson, B., Elliott, J., Hutton, J., Roe, D., Vira, B. & Wolmer, W. (2004), Biodiversity conservation and the eradication of poverty, Science, 306, 1146-1148 Adger, W. N., Benjaminsen, T. A., Brown, K., & Svarstad, H. (2001), Advancing a political ecology of global environmental discourses. Development and Change, 32, (4), 681- 715 Agar, M. & MacDonald, J. (2003), Transboundary conservation: The politics of ecological integrity in the Great Limpopo transfrontier park, Journal of Southern African Studies, 29, (1), 261-278 Agnew, J. (1997), The dramaturgy of horizons: geographical scale in the 'reconstruction of Italy' by the new Italian political parties, 1992-1995, Political Geography, 16, (2), 99- 121 Agrawal, A. (2001), Common property institutions and sustainable governance of resources, World Development, 29, (10), 1649-1672 Agrawal, A. (2005), Environmentality: technologies of government and the making of subjects, Duke University Press Durham, NC Agrawal, A. & Gibson, C. C. (1999), Enchantment and disenchantment: the role of community in natural resource conservation, World Development, 27, (4), 629-649 Agrawal, A. & Ostrom, E. (2001), Collective action, property rights and decentralisation in resource use in India and Nepal, 485-514 Allen, J. (1999), Spatial assemblages of power: from domination to empowerment, Human geography today, 194-218 256 Anderson, D. & Grove, R. (Eds.) (1987) Conservation in Africa: People, policies and practice, Cambridge University Press, Cambridge, 359pp Andersson, K. P. & Ostrom, E. (2004), Decentralization reforms: Help or hindrance to forest conservation, Paper presented at the conference of the International Association for the Study of Common Property (IASCP) in Oaxaca, Mexico, August 9-13, 2004 Andersson, K. P. & Ostrom, E. (2008), Analyzing decentralized resource regimes from a polycentric perspective, Policy sciences, 41, (1), 71-93 Arts, B. & Tatenhove, J. V. (2004), Policy and power: A conceptual framework between the 'old' and 'new' policy idioms, Policy sciences, 37, (3), 339-356 Attfield, R., Hatinghi, J. & Matshabaphala, M. (2004), Sustainable development, sustainable livelihoods and land reform in South Africa: a conceptual and ethical enquiry, Third World Quarterly, 25, (2), 405-421 Bachrach, P. & Baratz, M. S. (1962), Two faces of power, The American Political Science Review, 56, (4), 947-952 Baldus, R. D. (2001a) Introduction: Conservation by the people, Tanzania Wildlife Discussion Paper 29, Deutsche Gesellschaft für Technishce Zusammenarbeit, Dar es Salaam, Tanzania Baldus, R. D. (2001b), Wildlife conservation in Tanganyika under German colonial rule, Internationales Afrikaforum, 37, (1), 73-78 Baldus, R. D. & Cauldwell, A. E. (2004), Tourist hunting and its role in development of wildlife management areas in Tanzania, Proceedings of the 6th International Game Ranching Symposium, Paris Barnes, B. (1988), The nature of power, Polity Press Cambridge, UK, 205pp Barrett, C. B. & Arcese, P. (1995), Are Integrated Conservation-Development Projects (ICDPs) sustainable? On the conservation of large mammals in sub-Saharan Africa, World Development, 23, 1073-84 Barrow, E. & Murphree, M. (2001), Community conservation: from concept to practice, IN Hulme, D. & Murphree, M. (Eds.) African Wildlife and Livelihoods: The promise and practice of community conservation. James Curry Ltd., Oxford, p.24-37 Barrow, E., Gichohi, H. & Infield, M. (2001), The evolution of community conservation policy & practice in East Africa, IN Hulme, D. & Murphree, M. (Eds.) African wildlife and livelihoods: The promise and performance of community conservation. p.59-73 257 Barton, G. (2001), Empire forestry and the origins of environmentalism, Journal of Historical Geography, 27, (4), 529-552 Bassett, T. J. (1988), The Political Ecology of Peasant-Herder Conflicts in the Northern Ivory Coast, Annals of the Association of American Geographers, 78, (3), 453-472 Bawa, K. S., Joseph, G. & Setty, S. (2007), Poverty, biodiversity and institutions in forest- agriculture ecotones in the Western Ghats and Eastern Himalaya ranges of India, Agriculture, ecosystems & environment, 121, (3), 287-295 Bayart, J. F. (1993), L'Etat en Afrique: la Politique du Ventre (The state in Africa: the politics of the belly), Longman, London Behera, B. & Engel, S. (2006), Institutional analysis of evolution of joint forest management in India: A new institutional economics approach, Forest Policy and Economics, 8, (4), 350-362 Bell, R. H. V. (1987), Conservation with a human face: conflict and reconciliation in African land use planning, IN Anderson, B. & Grove, R. (Eds.) Conservation in Africa: People, policies and practice. p.79-101 Benjaminsen, T. A., & Svarstad, H. (2010), The death of an elephant: Conservation discourses versus practices in Africa, Forum for Development Studies, 37, (3), 385- 408 Benjaminsen, T. A. & Bryceson, I. (2012), Conservation, green/blue grabbing and accumulation by dispossession in Tanzania, Journal of Peasant Studies, 39, (2), 335- 355 Benjaminsen, T.A., Goldman, M.J., Minwary, M.Y. & Maganga, F.P., (in press), Wildlife Management in Tanzania: Recentralization, Rent-Seeking and Resistance, Geoforum Berg, B. L. (2009), Qualitative Research Methods for the Social Sciences, Allyn & Bacon, Boston, 418pp Bergin, P. & Dembe, E. (1996), Parks people in Tanzania: An overview of the Tanzania national parks Community Conservation Service (CCS), Participatory Wildlife Management, Rural Extension Bulletin, (10), 14-18 Bergin, P., (2001), Accommodating new narratives in a conservation bureaucracy: TANAPA & community conservation, IN Hulme, D. & Murphree, M. (Eds), African wildlife and livelihoods: the promise and performance of community conservation, James Currey Ltd, p.88-105 Berkes, F. (2004), Rethinking community-based conservation, Conservation biology, 18, (3), 621-630 258 Birkenholtz, T. (2011), Network political ecology: Method and theory in climate change vulnerability and adaptation research, Progress in Human Geography, 36, (3), 295- 315 Blaikie, P. (1985), The political economy of soil erosion in developing countries, Longman, London, 188pp Blaikie, P. (2011), Should some political ecology be useful? The Inaugural Lecture for the Cultural and Political Ecology Specialty Group, Annual Meeting of the Association of American Geographers, April 2010 (in press), Geoforum Blaikie, P. & Springate-Baginski, O. (2006), Is small really beautiful? Community-based natural resource management in Malawi and Botswana. 1942-1957 Blaikie, P. M. & Brookfield, H. C. (1987), Land degradation and society, Routledge Kegan & Paul, London, 296pp Blaikie, P. M. & Springate-Baginski, O. (2007), Forests, people and power: the political ecology of reform in South Asia, Earthscan, London, 394pp Blom, B., Sunderland, T. & Murdiyarso, D. (2010), Getting REDD to work locally: lessons learned from integrated conservation and development projects, Environmental science & policy, 13, (2), 164-172 Blomley, T. (2006) Mainstreaming participatory forestry within the local government reform process in Tanzania, Gatekeeper Series, International institute for environment and development (IIED) Blomley, T., Ramadhani, H., Mkwizu, Y. & Böhninger, A. (2008a) Hidden harvest: Unlocking the potential of community based forest management in Tanzania, Workshop on forest governance and decentralisation in Africa, Durban, South Africa, 8-11 April, 2008, 8-11 April, 2008 Blomley, T., Pfliegner, K., Isango, J., Zahabu, E., Ahrends, A. & Burgess, N. (2008b), Seeing the wood for the trees: An assessment of the impact of participatory forest management on forest condition in Tanzania, Oryx, 42, (3), 380-391 Bloor, M., Frankland, J. & Thomas, M. (2002), Farmers' participation in soil and water conservation activities in the Chemoga watershed, Blue Nile basin, Ethiopia, Land Degradation and Development, 13, (3), 189-200 Boiesen, J. H. & Lund, J. F. (2003), Participatory forest management in Tanzania: A socio- economic study of the implementation of PFM in Iringa district, Masters thesis, Department of economics and natural resources, Royal veterinary and agricultural university, Copenhagen 259 Bond, I. (2001), CAMPFIRE & the incentives for institutional change, IN Hulme, D. & Murphree, M. (Eds.) African wildlife and livelihoods: The promise and performance of community conservation. p. 227-243 Booth, D. (2011), Introduction: Working with the grain? The Africa power and politics programme, IDS bulletin, 42, (2), 1-10 Booth, V., Nangale, G. & Majamba, H. (2002) Procedures for communities to enter into joint ventures in WMAs, , Tanzania Wildlife Discussion Paper No. 31, Prepared for the Ministry of Natural Resources and Tourism, Wildlife Division Booth, W. J. (1994), On the idea of the moral economy, American Political Science Review, 88, (3), 653-667 Born, B. & Purcell, M. (2006), Avoiding the local trap: Scale and food systems in planning research, Journal of Planning Education and Research, 26, (2), 195 Borzel, T. A. (1998), Organizing Babylon: On the Different Conceptions of Policy Networks, Public Administration, 76, (2), 253-273 Borzel, T. A. (2011), Networks: Reified Metaphor or Governance Panacea?, Public Administration, 89, (1), 49-63 Bourdieu, P. (1989), Social space and symbolic power, Sociological theory, 7, (1), 14-25 Bourdieu, P. (1992), The logic of practice, Stanford Univ Press, Stanford, California, 337pp Brandon, K., Redford, K. & Sanderson, S. (1998), Parks in peril: people, politics, and protected areas, Island Press, Washington D.C., 521pp Brandon, K. E. & Wells, M. (1992), Planning for people and parks: design dilemmas, World Development, 20, (4), 557-570 Brenner, N. (2001), The limits to scale? Methodological reflections on scalar structuration, Progress in Human Geography, 25, (4), 591-614 Briggs, C. L. (1986), Learning how to ask: A Sociolinguistic Appraisal of the Role of the Interview in Social Science Research, Cambridge University Press, Cambridge, 155pp Brock, K., Cornwall, A. & Gaventa, J. (2001), Power, knowledge and political spaces in the framing of poverty policy, Institute of Development Studies, Sussex, 41pp Brockington, D. (2002), Fortress conservation: the preservation of the Mkomazi game reserve, Tanzania, The International African Institute, London, 196pp Brockington, D. (2006), The politics and ethnography of environmentalisms in Tanzania, African Affairs, 105, (418), 97-116 Brockington, D. (2008), Corruption, taxation and natural resource management in Tanzania, Journal of Development Studies, 44, (1), 103-126 260 Brockington, D. & Igoe, J. (2006), Eviction for conservation: A global overview, Conservation and society, 4, (3), 424-470 Brockington, D., Duffy, R. & Igoe, J. (2008), Nature Unbound: Conservation, Capitalism and the Future of Protected Areas, Earthscan, London Brooks, J. S., Franzen, M. A., Holmes, C. M., Grote, M. N. & Mulder, M. B. (2006), Testing hypotheses for the success of different conservation strategies, Conservation biology, 20, (5), 1528-1538 Brown, C. J. & Purcell, M. (2005), There's nothing inherent about scale: political ecology, the local trap, and the politics of development in the Brazilian Amazon, Geoforum, 36, (5), 607-624 Brown, P. (1990), Big man, past and present: Model, person, hero, legend, Ethnology, 29, (2), 97-115 Brunckhorst, D. J. (2010), Using context in novel community-based natural resource management: landscapes of property, policy and place, Environmental Conservation, 37, (01), 16-22 Bryant, R. L. (1998), Power, knowledge and political ecology in the third world: a review, Progress in physical geography, 22, (1), 79-84 Bryant, R. L. & Bailey, S. (1997), Third world political ecology, Routledge, London, 243pp Bulkeley, H. (2000), Discourse Coalitions and the Australian Climate Change Policy Network, Environment and Planning C: Government and Policy, 18, 727-748 Bulkeley, H. (2005), Reconfiguring environmental governance: towards a politics of scales and networks, Political Geography, 24, (8), 875-902 Burgess, N., Hales, J., Underwood, E., Dinerstein, E., Olson, D., Itoua, I., Schipper, J., Ricketts, T., Newman, K. & Hales, J. (2004), Terrestrial ecoregions of Africa and Madagascar: a conservation assessment, Island Press, Washington, 501pp Burgess, N. D., Bahane, B., Clairs, T., Danielsen, F., Dalsgaard, S., Funder, M., Hagelberg, N., Harrison, P., Haule, C. & Kabalimu, K. (2010), Getting ready for REDD+ in Tanzania: a case study of progress and challenges, Oryx, 44, (03), 339-351 Büscher, B. (2010), Anti-politics as political strategy: Neoliberalism and transfrontier conservation in southern Africa, Development and Change, 41, (1), 29-51 Büscher, B. & Dressler, W. (2007), Linking neoprotectionism and environmental governance: On the rapidly increasing tensions between actors in the environment- development nexus, Conservation and society, 5, (4), 586-611 261 Cameron, J. (2005), Focussing on the focus group, IN Hay, I. (Ed.) Qualitative research methods in human geography. Oxford University Press, p. 83-102 Campbell, J. R. (2001), Participatory rural appraisal as qualitative research: Distinguishing methodological issues from participatory claims, Human Organization, 60, (4), 380- 389 Campbell, L. M. (2007), Local conservation practice and global discourse: A political ecology of sea turtle conservation, Annals of the Association of American Geographers, 97, (2), 313-334 Campbell, L. M. & Vainio-Matila, A. (2003), Participatory development and community- based conservation: Opportunities missed for lessons learned?, Human Ecology, 31, (3), 417-436 Carr, E. S. (2003), Rethinking empowerment theory using a feminist lens: The importance of process, Affilia, 18, (1), 8-20 Chabal, P. & Daloz, J. P. (1999), Africa Works: Disorder as Political Instrument (African Issues), Indiana University Press, Indiana, 169pp Chambers, R. (1983), Rural development: Putting the last first, Longman London, 198pp Chambers, R. (1997), Whose Reality Counts?: Putting the First Last, ITDG Publishing, 297pp Charlton, R. & May, R. (1995), NGOs, politics, projects and probity: a policy implementation perspective, Third World Quarterly, 16, (2), 237-256 Charnley, S. & Poe, M. R. (2007), Community Forestry in Theory and Practice: Where Are We Now?, Annual Review of Anthropology, 36, 301-336 Cleaver, F. (1999), Paradoxes of participation: questioning participatory approaches to development, Journal of International Development, 11, (4), 597-612 Cleaver, F. (2001), Institutions, agency and the limitations of participatory approaches to development, IN Cooke, B. & Kothari, U. (Eds.) Participation: The new tyranny. Zedbooks, New York, p.36-55 Clegg, S. (1989), Frameworks of power, Sage Publications Ltd, London, Clegg, S. & Haugaard, M. (Eds.) (2009) The SAGE handbook of power, Sage Publications Ltd, London, 484pp Cloke, P. J., Philo, C. & Sadler, D. (1991), Approaching Human Geography : An Introduction to Contemporary Theoretical Debates, Chapman, London, 240pp Cooke, B. & Kothari, U. (2001), Participation: The New Tyranny? Zed Books, New York, 207pp 262 Cornwall, A. (2003), Whose voices? Whose choices? Reflections on gender and participatory development, World Development, 31, (8), 1325-1342 Cornwall, A. (2004), Spaces for transformation? Reflections on issues of power and difference in participation in development, IN Hickey, S. & Mohan, G. (Eds.) Participation: From Tyranny to Transformation. Zed Books, London, p.75-92 Coulson, A. (1982), Tanzania: a political economy, Oxford University Press, Oxford, 394pp Crang, M. (1997), Analyzing Qualitative Materials, IN Flowerdew, R. & Martin, D. (Eds.) Methods in Human Geography: A Guide for Students doing a Research Project. Pearson Educations Ltd, Harlow, p.218-233 Crook, R. C. (2003), Decentralisation and poverty reduction in Africa: The politics of local- central relations, Public Administration and Development, 23, (1), 77-88 Dahl, R. A. (1961), Who Governs: Democracy and power in an american city, Yale University Press, New Haven, 357pp De Haan, L. & Zoomers, A. (2005), Exploring the frontier of livelihoods research, Development and Change, 36, (1), 27-47 Dressler, W., Büscher, B., Schoon, M., Brockington, D., Hayes, T., Kull, C. A., McCarthy, J. & Shrestha, K. (2010), From hope to crisis and back again? A critical history of the global CBNRM narrative, Environmental Conservation, 37, (01), 5-15 Dreyfus, H. & Rabinow, P. (1982), Michel Foucault: Beyond Hermeneutics and Structuralism, Harvester, Brighton Duffy, R. (2000), Killing for conservation: wildlife policy in Zimbabwe, James Currey Ltd, 209pp Dwyer, C. & Limb, M. (2001), Introduction: Doing Qualitative Research in Geography, IN Limb, M. D., Claire. (Ed.) Qualitative Methodologies for Geographers: Issues and Debates. Arnold, London, p.1-23 Edwards, B., Goodwin, M., Pemberton, S. & Woods, M. (2001), Partnerships, power, and scale in rural governance, Environment and Planning C, 19, (2), 289-310 Elliott, J. (2001) Wildlife and Poverty Study: Phase 1 Report, Department for International Development England, K. V. L. (1994), Getting Personal: Reflexivity, Positionality, and Feminist Research, The Professional Geographer, 46, (1), 80-89 EPIQ (2000) Community-Based Conservation (CBC) Experience in Tanzania: an Assessment of Lessons Learned, USAID (Environmental POlicy and Institutional Strengthening), Dar es Salaam 263 Escobar, A. (1995), Encountering development: The making and unmaking of the Third World, Princeton Univ Prress, 290pp Etzioni, A. (1967), Mixed-scanning: A "third" approach to decision-making, Public administration review, 25, (5), 385-392 Etzioni, A. (1986), Mixed scanning revisited, Public administration review, 46, (1), 8-14 Evans, S. L., Robson, E. & Willis, K. (1994), The Logistics of Undertaking Fieldwork in Developing Areas, IN Developing Areas Research Group (Ed.) Postgraduate Fieldwork in Developing Areas: A Rough Guide. Developing Areas Research Group of the Institute of British Geographers, London, p.7-18 Fabricius, C. (2004), The fundamentals of community-based natural resource management, IN Fabricius, C., Koch, E., Magome, H. & Turner, S. (Eds.) Rights, resources and rural development: Community-based natural resource management in Southern Africa. p.3-43 Fabricius, C., Koch, E., Magome, H. & Turner, S. (Eds.) (2004) Rights, resources and rural development: community-based natural resource management in Southern Africa, Earthscan, London, 288pp Fairhead, J. & Leach, M. (1996), Misreading the African landscape: society and ecology in a forest-savanna mosaic, Cambridge University Press, 357pp Feindt, P. H. & Oels, A. (2005), Does discourse matter? Discourse analysis in environmental policy making, Journal of Environmental Policy & Planning, 7, (3), 161-173 Ferguson, J. (1990), The anti-politics machine:" development," depoliticization, and bureaucratic power in Lesotho, Cambridge University Press, Cambridge, 323pp Ferraro, P. J. (2002), The local costs of establishing protected areas in low-income nations: Ranomafana National Park, Madagascar, Ecological Economics, 43, (2-3), 261-275 Fischer, F. & Forester, J. (Eds.) (1993) The argumentative turn in policy analysis and planning, Duke University Press, Durham, 329pp Flick, U. (2002), An introduction to qualitative research, SAGE, London, 310pp Flowerdew, R. & Martin, D. (1997), Methods in Human Geography: A Guide for Students doing a Research Project, Pearson Education Ltd, Harlow, 296pp Foddy, W. (1993), Constructing questions for interviews: Theory and Practice in Social Research, Cambridge University Press, Cambridge, 288pp Forsyth, T. (2003), Critical political ecology: The politics of environmental science, Routledge, London, 322pp 264 Forsyth, T. (2008), Political ecology and the epistemology of social justice, Geoforum, 39, (2), 756-764 Foucault, M. (1975), Discipline and Punish: the birth of the prison, Allen Lane, London, Foucault, M. (1980), Power/knowledge: Selected interviews and other writings, 1972-1977, IN Gordon, C. (Ed.). Pantheon New York, 270pp Foucault, M. (1982), Afterword: The subject and power IN Dreyfus, H. L. & Rabinow, P. (Eds.) Michel Foucault: Beyond structuralism and hermeneutics. p.208-26 Francis, P. & James, R. (2003), Balancing rural poverty reduction and citizen participation: the contradictions of Uganda's decentralization program, World Development, 31, (2), 325-337 Fraser, A. (2010), The craft of scalar practices, Environment and Planning A, 42, (2), 332- 346 Frey, J. H. & Fontana, A. (1991), The group interview in social research, IN Morgan, D. L. (Ed.) Successful Focus groups: Advancing the State of the Art. SAGE, California, p.175-87 Freire, P. (1970), Cultural action and conscientization, Harvard Educational Review, 40, (3), 452-477 Frost, P. G. H. & Bond, I. (2008), The CAMPFIRE programme in Zimbabwe: Payments for wildlife services, Ecological Economics, 65, (4), 776-787 Gaventa, J. (1982), Power and powerlessness: Quiescence and Rebellion in an Appalachian Valley, University of Illinois Press, Gibson, C. C. & Marks, S. A. (1995), Transforming rural hunters into conservationists: an assessment of community-based wildlife management programs in Africa, World Development, 23, (6), 941-957 Gibson, C. C., Ostrom, E. & Ahn, T. K. (2000), The concept of scale and the human dimensions of global change: a survey, Ecological Economics, 32, (2), 217-239 Giddens, A. (1984), The constitution of society: Outline of the theory of structuration, University of California Press, 407pp Gillingham, S. (1998), Giving wildlife value: a case study of community wildlife management around the Selous Game Reserve, Tanzania, PhD, Department of Biological Anthropology, University of Cambridge, Cambridge Gillingham, S. & Lee, P. C. (1999), The impact of wildlif-related benefits on the conservation attitudes of local people around the Selous Game Reserve, Tanzania, Environmental Conservation, 26, (3), 218-228 265 Glaser, B. G. & Strauss, A. L. (1967), The Discovery of Grounded Theory :Strategies for Qualitative Research Aldine de Gruyter, Hawthorne, N.Y, 274pp Göhler, G. (2009), 'Power to' and 'power over', IN Clegg, S. & Haugaard, M. (Eds.) The SAGE handbook of power. SAGE, London, p.27-40 Goldman, M. (2003), Partitioned nature, privileged knowledge: community-based conservation in Tanzania, Development and Change, 34, (5), 833-862 Goldman, M. (2011), Strangers on their own land: Maasai and wildlife conservation in northern Tanzania, Conservation and society, 9, (1), 65-79 Goldman, M. J., Nadasdy, P. & Turner, M. D. (2011), Knowing nature: conversations at the intersection of political ecology and science studies, University of Chicago Press, London, 367pp Government of the United Republic of Tanzania, WWF, USAID & Muungano wa Jumuiya Zilizoidhinishwa za Hifadhi ya Maeneo ya Wanyamapori Tanzania (2011), Muungano wa Jumuiya Zilizoidhinishwa za Hifadhi ya Maeneo ya Wanyamapori Tanzania(Authorised Associations Consortium). Dar es Salaam, Government of the United Republic of Tanzania (1998a), Wildlife Policy of Tanzania, Ministry of Natural Resources and Tourism (Wildlife Division) Government of the United Republic of Tanzania (1998b), National Forest Policy of Tanzania, Tourism, Ministry of Natural Resources and Tourism (Froestry and Beekeeping Division) Government of the United Republic of Tanzania (2002a), The wildlife conservation (Wildlife Management Areas) Regulations, Ministry of Natural Resources and Tourism (Wildlife Division) Government of the United Republic of Tanzania (2002b), Forest Act, Ministry of Natural Resources and Tourism, Ministry of Natural Resources and Tourism (Froestry and Beekeeping Division) Government of the United Republic of Tanzania (2005), Wildlife conservation (Wildlife Management Areas) Regulations, Ministry of Natural Resources and Tourism (Wildlife Division) Government of the United Republic of Tanzania (2007), The Wildlife Policy of Tanzania, Ministry of Natural Resources and Tourism (Wildlife Division) Government of the United Republic of Tanzania, (2008), Wildlife Conservation (Non- Consumptive Wildlife Utilization) Regulations 2007 (Amended), Ministry of Natural Resources and Tourism (Wildlife Division) 266 Government of the United Republic of Tanzania (2009), Wildlife Conservation Act, Ministry of Natural Resources and Tourism (Wildlife Division) Grindle, M. S. (2004), Good enough governance: poverty reduction and reform in developing countries, Governance, 17, (4), 525-548 Grindle, M. S. (2007), Good enough governance revisited, Development Policy Review, 25, (5), 533-574 Grindle, M. S. & Thomas, J. W. (1991), Public choices and policy change: The political economy of reform in developing countries, Johns Hopkins University Press Baltimore, 222pp Grove, R. (1987), Early themes in African conservation: the Cape in the nineteenth century, IN Anderson, B. & Grove, R. (Eds.) Conservation in Africa: People, policies and practice. Cambridge University Press, Cambridge, p.21-39 Guijt, I. & Shah, M. K. (1998), The myth of community: Gender issues in participatory development, Intermediate technology publications, London Haas, P. M. (1992), Introduction: epistemic communities and international policy coordination, International organization, 46, (1), 1-35 Hackel, J. D. (1999), Community conservation and the future of Africa's wildlife, Conservation biology, 13, (4), 726-734 Hadiz, V. R. (2004), Decentralization and Democracy in Indonesia: A Critique of Neo- Institutionalist Perspectives, Development and Change, 35, (4), 697-718 Hahn, R. & Kaggi, D. (2001) Selous Game Reserve Development of CBC in the Buffer Zone: Facts and Figures, Experiences With Community Based Wildlife Conservation in Tanzania, Deutsche Gesellschaft für Technische Zusammenarbeit, Dar es Salaam, Tanzania Hajer, M. (1995), The Politics of Environmental Discourse: Ecological Modernisation and the Policy Process, Clarendon Press, Oxford, 332pp Hajer, M. & Versteeg, W. (2005), A decade of discourse analysis of environmental politics: achievements, challenges, perspectives, Journal of Environmental Policy Planning, 7, (3), 175-184 Hammersley, M. (1995), The politics of social research, Sage Publications Ltd, London, 186pp Hargreaves, J. R., Morison, L. A., Gear, J. S. S., Makhubele, M. B., Porter, J. D. H., Busza, J., Watts, C., Kim, J. C. & Pronyk, P. M. (2007), 'Hearing the voices of the poor': 267 Assigning poverty lines on the basis of local perceptions of poverty. A quantitative analysis of qualitative data from participatory wealth ranking in rural South Africa, World Development, 35, (2), 212-229 Harriss-White, B. & White, G. (1996), Corruption, liberalization and democracy: editorial introduction, IDS bulletin, 27, (2), 1-5 Hartley, D. (1996), Initiating community-based conservation within a weak policy framework and from a complex inistitutional base: The case of local implementation in the Great Ruaha ecosystem of Tanzania, Draft paper Hartley, D. (1997) Community Wildlife Management: A Review of the ODA's Experience in Tanzania, Report to the Overseas Development Administration Harvey, D. (1989), The urban experience, Johns Hopkins University Press Baltimore, 293pp Harvey, C. A., Dickson, B. & Kormos, C. (2010), Opportunities for achieving biodiversity conservation through REDD, Conservation Letters, 3, (1), 53-61 Haugaard, M. (1997), The Constitution of Power: A theoretical analysis of power, knowledge and structure, Manchester University Press, Manchester, 234pp Haugaard, M. (2002), Power: A reader, Manchester University Press, 347pp Haugaard, M. (2003), Reflections on seven ways of creating power, European Journal of Social Theory, 6, (1), 87 Hayes, T. M. (2006), Parks, people and forest protection: An institutional assessment of the effectiveness of protected areas, World Development, 34, (12), 2064-2075 Hayward, C. & Lukes, S. (2008), Nobody to shoot? Power, structure, and agency: A dialogue, Journal of Power, 1, (1), 5-20 Health and Development International Consultants (HDIC) (2010) Socio-economic baseline studies in selected wildlife management areas under the financial crisis initiative/cash-for-work program: MBOMIPA WMA report, Report prepared for WWF, USAID funded. Henry, A. D. (2011), Ideology, power and the structure of policy networks, The Policy Studies Journal, 39, (3), 361-383 Hickey, S. & Mohan, G. (2004a), Participation--from tyranny to transformation?: Exploring new approaches to participation in development, Zed books, London, 292pp Hickey, S. & Mohan, G. (2004b), Towards participation as transformation: critical themes and challenges, IN Hickey, S. & Mohan, G. (Eds.) Participation: From tyranny to transformation. Zed Books, London, 3-24 268 Hildyard, N., Hegde, P., Wolvekamp, P. & Reddy, S. (2001), Pluralism, Participation and Power: Joint Forest Management, IN Cooke, B. & Kothari, U. (Eds.) Participation: The New Tyranny. Zed, London, p.56-72 HIMA (2000) HIMA Marketing Study, Final Report, Danish Ministry of Foreign Affairs, Hirschman, A. (1963), Problem solving and policy-making: A Latin American style, Twentieth Century Fund, Journeys toward Progress: Studies in Economic Policy- Making in Latin America, New York Hofferbert, R. I. (1974), The study of public policy, Bobbs-Merrill Indianapolis, Minnesota, 275pp Hoggart, K., Lees, L. & Davies, A. (2002), Researching human geography, A Hodder Arnold Publication, London, 275pp Homewood, K., Kristjanson, P. & Trench, P. C. (Eds.) (2009) Staying Maasai?: Livelihoods, conservation, and development in East African rangelands, Studies in human ecology and adaptation, Springer Verlag, New York, 417pp Howard, S. (1994), Methodological issues in overseas fieldwork: experiences from Nicaragua’s Northern Atlantic Coast, IN Robson, E. & Willis, K. (Eds.) Postgraduate Fieldwork in Developing Areas: a Rough Guide. Developing Areas Research Group of the Institute of British Geographers, p.19-35 Hulme, D. & Murphree, M. (1999), Communities, Wildlife and the ‘New Conservation’in Africa, Journal of International Development, 11, (2), 277-285 Hulme, D. & Murphree, M. (2001a), Community conservation in Africa: An introduction, IN Hulme, D. & Murphree, M. (Eds.) African wildlife and livelihoods: The promise and performance of community conservation. James Currey Ltd., Oxford, p.1-8 Hulme, D. & Murphree, M. (2001b), African Wildlife and Livelihoods: The Promise and Performance of Community Conservation, James Currey Ltd, Oxford, 320pp Hurst, A. W. (2004), Not Yet Out of the Woods: A Political Ecology of State Forest Policy and Practice in Mainland Tanzania, 1961-1998, PhD thesis, University of Oxford, Oxford Hutton, J., Adams, W. M. & Murombedzi, J. C. (2005), Back to the barriers? Changing narratives in biodiversity conservation, Forum for Development Studies, 32, (2), 341- 370 Hyden, G. (1980), Beyond Ujamaa in Tanzania: underdevelopment and an uncaptured peasantry, University of California Press, Berkeley, 270pp 269 Hyden, G. (2005) Why do things happen the way they do? A power analysis of Tanzania, Report to the Swedish International Development Agency Embassy of Sweden in Dar Es Salaam Hyden, G. (2008) Institutions, power and policy outcomes in Africa, Power and Politics in Africa Discussion Paper, No.2, published on behalf of the African Power and Politics Programme with support from the Overseas Development Institute, London Hyden, G. & Karlstrom, B. (1993), Structural adjustment as a policy process: The case of Tanzania, World Development, 21, (9), 1395-1404 Igoe, J. & Brockington, D. (2007), Neoliberal conservation: A brief introduction, Conservation and society, 5, (4), 432 Igoe, J. & Croucher, B. (2007), Conservation, commerce and communities: The story of community-based wildlife management areas in Tanzania's northern tourist circuit, Conservation & Society, 5, (4), 534-561 Iliffe, J. (1979), A modern history of Tanganyika, Cambridge University Press, Cambridge Ingle, C. R. (1972), From Village to State in Tanzania: The politics of rural development, Cornell University Press, Ithaca, New York, 279pp Institute of Resource Assessment (2007) Assessment and Evaluation of the Wildlife Management Areas in Tanzania, report to the Ministry of Natural resources and Tourism (Wildlife Division) United Republic of Tanzania Jenkins-Smith, H. C. & Sabatier, P. A. (1994), Evaluating the advocacy coalition framework, Journal of Public Policy, 14, 175-175 John, P. (2003), Is there life after policy streams, advocacy coalitions, and punctuations: using evolutionary theory to explain policy change?, Policy Studies Journal, 31, (4), 481-498 Jones, J. P. G., Andriamarovololona, M. M. & Hockley, N. (2008), The importance of taboos and social norms to conservation in Madagascar, Conservation biology, 22, (4), 976- 986 Jones, S. (2006), A political ecology of wildlife conservation in Africa, Review of African Political Economy, 33, (109), 483-495 Kapoor, I. (2002), The devil's in the theory: a critical assessment of Robert Chambers' work on participatory development, Third World Quarterly, 23, (1), 101-117 Kapoor, I. (2005), Participatory development, complicity and desire, Third World Quarterly, 26, (8), 1203-1220 270 Keeley, J. & Scoones, I. (2003), Understanding Environmental Policy Processes: Cases from Africa, Earthscan, London, 224pp Keil, R. & Debbané, A. M. (2005), Scaling discourse analysis: Experiences from Hermanus, South Africa and Walvis Bay, Namibia, Journal of Environmental Policy & Planning, 7, (3), 257-276 Kelsall, T. (2011), Rethinking the Relationship between Neo-patrimonialism and Economic Development in Africa, IDS bulletin, 42, (2), 76-87 Kelsall, T. & Mercer, C. (2003), Empowering people? World vision &'transformatory development'in Tanzania, Review of African Political Economy, 30, (96), 293-304 Kelsall, T., Booth, D., Cammack, D. & Golooba-Mutebi, F. (2010), Developmental patrimonialism? Questioning the orthodoxy on political governance and economic progress in Africa, Africa Power and Politics Working Paper No, 9 Kepe, T. & Cousins, B. (2001), Resource tenure and power relations in community wildlife: The case of Mkambati area, South Africa, Society and Natural Resources, 14, 911- 925 Kesby, M. (2005), Retheorizing empowerment through participation as a performance in space: beyond tyranny to transformation, Signs, 30, (4), 2037-2063 Kesby, M. (2007), Spatialising participatory approaches: The contribution of geography to a mature debate, Environment & Planning A, 39, (12), 2813-2831 Kideghesho, J. R. (2006), Wildlife Conservation and Local Land use Conflicts in Western Serengeti Corridor, Tanzania, PhD thesis, Department of Biology, Norwegian University of Science and Technology, Trondheim Kingdon, J. W. (1984), Agendas, alternatives, and public policies, Longman, New York, 253pp Kitchin, R. & Tate, N. J. (2000), Conducting Research into Human Geography, Prentice Hall, Harlow, 330pp Kitzinger, J. (1994), Focus groups: method or madness, IN Boulton, M. (Ed.) Challenge and Innovation: Methodological Advances in Social Research on HIV/AIDS. Taylor & Francis, New York, p.159-75 Knodel, J. (1993), The design and analysis of focus group studies: A practical approach, IN Morgan, D. L. (Ed.) Successful focus groups: Advancing the state of the art. SAGE, California, p.35-50 Koppers, B., Mariki, S., Shauri, V. & Vignon, C. (2004) Development of guidelines and regulations regarding sharing of costs and forest revenues/benefits in participatory 271 forest management in Tanzania - Review of participatory forest management related legislation in Tanzania, Report to HTSPE Development, Dar es Salaam Krueger, J. (Ed.) (1998) Developing Questions for Focus Groups, Focus Group Kit 3, Sage, London, 107pp Kumar, S. (2002), Does 'Participation' in Common Pool Resource Management Help the Poor? A Social Cost-Benefit Analysis of Forest Management in Jharkand, India, World Development, 30, (5), 763-782 Lahat, L. (2011), How can leaders' perceptions guide policy analysis in an era of governance?, Policy sciences, 44, 135-155 Larson, A. M. & Soto, F. (2008), Decentralization of natural resource governance regimes, Annual Review of Environment and Resources, 33, 213-239 Leach, M., Mearns, R. & Scoones, I. (1997a), Community-based sustainable development: consensus or conflict?, IDS bulletin, 28, (4), 1-3 Leach, M., Mearns, R. & Scoones, I. (1997b), Challenges to Community Based Sustainable Development: Dynamics, Entitlements, Institutions, IDS bulletin, 28, (4), 4-14 Leach, M., Mearns, R. & Scoones, I. (1999), Environmental entitlements: dynamics and institutions in community-based natural resource management, World Development, 27, (2), 225-247 Leader-Williams, N. (2000), The effects of a century of policy and legal change on wildlife conservation and utilisation in Tanzania, IN Prins, H. H. T., Grootenhuis, J. G. & Dolan, T. T. (Eds.) Wildlife conservation by sustainable use. Kluwer Academic Publishers, Boston, p.219-245 Leader-Williams, N., Kayera, J. A. & Overton, G. L. (1996), Community-based Conservation in Tanzania, Occasional paper of the IUCN Species Survival Commission No. 15, Leader-Williams, N., Baldus, R. D. & Smith, R. (2009), The influence of corruption on the conduct of recreational hunting, IN Dickson, B., Hutton, J. & Adams, W. M. (Eds.) Recreational Hunting, Conservation and Rural Livelihoods: Science and Practice. Wiley-Blackwell, Chichester, p.296-316 Lebel, L., Garden, P. & Imamura, M. (2005), The politics of scale, position, and place in the governance of water resources in the Mekong region, Ecology and Society, 10, (2), 18-38 Lebel, L., Daniel, R., Badenoch, N., Garden, P. & Imamura, M. (2008), A multi-level perspective on conserving with communities: Experiences from upper tributary 272 watersheds in montane mainland Southeast Asia, International Journal of the Commons, 2, (1), 127-154 Lefebvre, H. (1974), The production of space (translated by Nicholson-Smith, D.), Blackwell, Oxford, 454pp Leitner, H. & Miller, B. (2007), Scale and the limitations of ontological debate: a commentary on Marston, Jones and Woodward, Transactions of the Institute of British Geographers, 32, (1), 116-125 Leys, C. (1996), The rise and fall of development theory, James Currey, London, 205pp Lindblom, C. E. (1959), The science of" muddling through", Public administration review, 19, (2), 79-88 Lindblom, C. E. (1979), Still muddling, not yet through, Public administration review, 39, (6), 517-526 Lovett, J. C. (2003), Tanzanian Forest Law, IN Chaytor, B. & Gray, K. (Eds.) Environmental Law and Policy in Africa. Kluwer Law International, Dordrecht, p.151-180 Lovett, J. C. & Pócs, T. (1993) Assessment of the condition of the Catchment Forest Reserves, a botanical appraisal: Iringa region Catchment Forestry Report 93.3 Lukes, S. (1974), Power: A radical view London, Macmillan, London Lukes, S. (2005), Power: a radical view, 2nd edition, Palgrave Macmillan, Basingstoke, 192pp Lund, J. F. (2004), Participatory Forest Management and Poverty: Distributional Effects of Participatory Forest Management in Tanzanian Miombo Woodlands, draft paper, Centre for Forest, Landscape and Planning, Royal Veterinary and Agricultural University, Unit of Forestry, Copenhagen Lund, J. F. (2007), Is small beautiful? Village level taxation of natural resources in Tanzania, Public Administration and Development, 27, (4), 307-318 Lund, J. F. & Treue, T. (2008), Are we getting there? Evidence of decentralized forest management from the Tanzanian Miombo woodlands, World Development, 36, (12), 2780-2800 Mackenzie, C. (1995) Village-based wildlife management areound Ruaha National Park, Tanzania: Planning for implementation, Report to the Overseas Development Administration MacKenzie, J. (1987), Chivalry, social Darwinism and ritualised killing: The hunting ethos in Central Africa up to 1914, IN Anderson, D. & Grove, R. (Eds.) Conservation in Africa: People, policies and practice. Cambridge University Press, Cambridge, 41-78 273 Madge, C. (1994), The Ethics of Research in the ‘Third World’, IN Developing Areas Research Group (Ed.) Postgraduate Fielwork in Developing Areas; A Rough Guide. Developing Areas Research Group of the Institute of British Geographers, p.91-102 Magome, H. & Fabricius, C. (2004), Reconciling biodiversity conservation with rural development: the Holy Grail of CBNRM, IN Fabricius, C., Koch, E., Magome, H. & Turner, S. (Eds.) Rights, Resources and Rural Development: Communty-Based Natural Resource Management in Southern Africa. Earthscan, London, p.93-111 Majamba, H. (2001), Regulating the Hunting Industry in Tanzania: Reflections on the Legislative, Institutional, and Policy-making Frameworks, Lawyers' Environmental Action Team, Marsh, D. & Rhodes, R. A. W. (Eds.) (1992) Policy networks in British politics, Clarendon, Oxford, 312pp Marsh, D. & Smith, M. (2000), Understanding policy networks: towards a dialectical approach, Political studies, 48, (1), 4-21 Marsh, G. P. (1865), Man and nature: or, physical geography as modified by human action, Scribners (reprinted Harvard University Press, 1965), New York Marston, S. A. (2000), The social construction of scale, Progress in Human Geography, 24, (2), 219-242 Marston, S. A., Jones III, J. P. & Woodward, K. (2005), Human geography without scale, Transactions of the Institute of British Geographers, 30, (4), 416-432 Massao, J. (2003) Participatory Forest Management Activities Supported by MEMA: Progressive Report on MEMA Projects in Iringa 1999-2003, District Natural Resources Office, Iringa Massey, D. (1978), Regionalism: some current issues, Capital & Class, 2, (3), 106-125 Matta, J., Alavalapati, J., Kerr, J. & Mercer, E. (2005), Agency perspectives on transition to participatory forest management: a case study from Tamil Nadu, India, Society and Natural Resources, 18, (10), 859-870 Matti, S. & Sandström, A. (2011), The Rationale Determining Advocacy Coalitions: Examining Coordination Networks and Corresponding Beliefs, Policy Studies Journal, 39, (3), 385-410 MBOMIPA (2002), Constitution, MBOMIPA Authorised Association McClintock, C. C., Brannon, D. & Maynard-Moody, S. (1979), Applying the logic of sample surveys to qualitative case studies: The case cluster method, Administrative Science Quarterly, 24, (4), 612-629 274 McShane, T. O., Hirsch, P. D., Trung, T. C., Songorwa, A. N., Kinzig, A., Monteferri, B., Mutekanga, D., Thang, H. V., Dammert, J. L. & Pulgar-Vidal, M. (2011), Hard choices: Making trade-offs between biodiversity conservation and human well-being, Biological Conservation, 144, 966-972 MEMA (2001) Review of MEMA Phase I and forumulation/pre-appraisal of a second phase, Report to the Danish International Development Agency, Ministry of Foreign Affairs, Copenhagen MEMA (2003) MEMA: Initial learning experiences, Report to the Ministry of Natural Resources and Tourism (Forestry and Beekeeping Division) & Iringa District Council Mercer, C., Mohan, G. & Power, M. (2003), Towards a critical political geography of African development, Geoforum, 34, (4), 419-436 Meshack, C. K., Ahdikari, B., Doggart, N. & Lovett, J. C. (2006), Transaction costs of community-based forest management: Empirical evidence from Tanzania, African Journal of Ecology, 44, 468-477 Midgley, J., Hall, A., Hardiman, M. & Narine, D. (1986), Community participation, social development and the state, Methuen, London, 187pp Miles, M. B. (1979), Qualitative data as an attractive nuisance - the problem of analysis, Administrative Science Quarterly, 24, (4), 590-601 Ministry of Natural Resources and Tourism (2000a), Wildlife Management Area Regulations, Wildlife Division Ministry of Natural Resources and Tourism (2000b), Wildlife Conservation (tourist hunting) Regulations, Wildlife Division Ministry of Natural Resources and Tourism (2000c), Draft Wildlife Management Area Guidelines Wildlife Division Ministry of Natural Resources and Tourism (2000d), Second Draft Wildlife Management Area Guidelines, Wildlife Division Ministry of Natural Resources and Tourism (2001a), Community-Based Forest Management Guidelines, Forestry and Beekeeping Division Ministry of Natural Resources and Tourism (2001b), Draft Wildlife Conservation (Wildlife Management Areas) Regulations, Wildlife Division Ministry of Natural Resources and Tourism (2002), Wildlife Conservation (tourist hunting) Regulations, Wildlife Division 275 Ministry of Natural Resources and Tourism (2003) Baseline Survey of the Eastern Pilot Wildlife Management Areas, Volume 2: Idodi-Pawaga, Report for the Wildlife Division by Mung'ong'o, C.G., Mwakaje, A., Kijazi, A. , Dar es Salaam Ministry of Natural Resources and Tourism (2008) Action Research into Poverty Impacts of Participatory Forest Management (ARPIP): Case studies from the Eastern Arc mountians area of Tanzania, Forestry and Beekeeping Division, Dar es Salaam Ministry of Natural Resources and Tourism (2009) Participatory forest management in Tanzania: An overview of status, progress and challenges ahead Report for the Forestry and Beekeeping Division by Blomley T. & Iddi T., Dar es Salaam Mishler, E. G. (1986), Research Interviewing: Context and narrative, Harvard University Press, Cambridge, MA, 89pp Mkumbukwa, A. R. (2008), The evolution of wildlife conservation policies in Tanzania during the colonial and post-independence periods, Development Southern Africa, 25, (5), 589-600 Mniwasa, E. & Shauri, V. (2001) Review of the Decentralization Process and it's Impact on Environmental and Natural Resources Management in Tanzania, Report to the Lawyers Environmental Action Team, Dar es Salaam Mohan, G. (2001), Beyond participation: Strategies for deeper empowerment, IN Cooke, B. & Kothari, U. (Eds.) Participation: The New Tyranny?, Zed books, London, p.153- 168 Mohan, G. & Stokke, K. (2000), Participatory development and empowerment: The dangers of localism, Third World Quarterly, 21, 247-268 Moore, D. S. (1993), Contesting terrain in Zimbabwe's eastern highlands: political ecology, ethnography, and peasant resource struggles, Economic Geography, 69, (4), 380-401 Morgan, D. L. (Ed.) (1998) Planning Focus groups, Focus Group Kit 1, Sage, London, 139pp Morgan, D. L. & Spanish, M. T. (1984), Focus Groups: A New Tool for Qualitative Research, Qualitative Sociology, 7, (3), 253-270 Morgan, D. L. & Krueger, R. A. (1993), When to use focus groups & why, IN Morgan, D. L. (Ed.) Successful focus groups: Advancing the state of the art. Sage, London, p.3-20 Moss, D. (1995), Patronage revisited: The dynamics of information and reputation, Journal of Modern Italian Studies, 1, (1), 58-93 Mosse, D. (2001), People's knowledge, participation and patronage: Operations and representations in rural development, IN Cooke, B. & Kothari, U. (Eds.) Participation: The new tyranny?, Zed Books, London, p.16-36 276 Murombedzi, J. (2001), A case study of ecotourism development from Mahenye, Zimbabwe, IN Hulme, D. & Murphree, M. (Eds.) African wildlife & livelihoods: the promise and performance of community conservation. James Currey Press, Oxford, p.208-226 Murombedzi, J. (2003), Devolving the expropriation of nature: The 'devolution' of wildlife management in Southern Africa, IN Adams, W. M. & Mulligan, M. (Eds.) Decolonizing Nature: Strategies for Conservation in a post-colonial era. Earthscan, London, p.136-151 Murombedzi, J. C. (1999), Devolution and stewardship in Zimbabwe's CAMPFIRE programme, Journal of International Development, 11, (2), 287-293 Murphree, M. (1996), Community Based conservation in Tanzania. IUCN Species Survival Commission, Occasional Paper no. 15 Murphree, M. W. (2000) Review of the Government of Tanzania/DfID MBOMIPA Project, Report for the Department for International Development Murphree, M. W. (2005), Congruent Objectives, Competing Interests, and Strategic Compromise: Concept and Process in the Evolution of Zimbabwe’s CAMPFIRE, 1984-1996, IN Brosius, J. P., Tsing, A. & Zerner, C. (Eds.) Communities and conservation: histories and politics of community-based natural resource management. p.105-147 Mushi, S. S. (2001), Development and democratisation in Tanzania: a study of rural grassroots politics, Fountain Publishers Ltd, Kampala, 90pp Mustalahti, I. (2007), Msitu wa Angai: Haraka, haraka, haina baraka! Why does handing over Angai forest to local villages proceed so slowly?, IN Gould, J. & Siitonen, L. (Eds.) Anomalies of aid. A festschrift for Juhani Koponen Interkont Books, Institute of Development Studies, University of Helsinki, Helsinki, p.168-86 Mustalahti, I. & Lund, J. F. (2009), Where and How Can Participatory Forest Management Succeed? Learning From Tanzania, Mozambique, and Laos, Society & Natural Resources, 23, (1), 31-44 Nagothu, U. S. (2001), Fuelwood and fodder extraction and deforestation: mainstream views in India discussed on the basis of data from the semi-arid region of Rajasthan, Geoforum, 32, (3), 319-332 National Bureau of Statistics (2007), Iringa Region Socio-Economic Profile, Published by the Government of Tanzania in collaboration with the Iringa Regional Commissioner's Office 277 Nelson, F. (2007) Emergent or illusory? Community wildlife management in Tanzania, International Institute for Environment and Development, issue paper no. 146 Nelson, F. (Ed.) (2010) Community Rights, Conservation and Contested land: the politics of Natural Resource Governance in Africa, Earthscan, London 342pp Nelson, F. & Blomley, T. (2007), Eating from the same plate: Integrating community-based wildlife and forestry management. The Arc Journal of the Tanzanina Forest Conservation Group. Dar es Salaam Nelson, F. & Agrawal, A. (2008), Politics of co-optation: Community forest management versus Joint Forest Management in Orissa, India, Environmental Management, 41, 707-718 Nelson, F. & Blomley, T. (2010), Peasants' forests and the King's game? Institutional divergence and convergence in Tanzania's forestry and wildlife sectors, IN Nelson, F. (Ed.) Community Rights, Conservation and Contested Land. The Politics of Natural Resource Governance in Africa. Earthscan, London, p.79-106 Nelson, F., Nshala, R. & Rodgers, W. M. (2007), The evolution and reform of Tanzanian wildlife management, Conservation & Society, 5, (2), 232-261 Nelson, F. (2009). Conservation and aid: designing more effective investments in natural resource governance reform. Conservation Biology, 23, (5), 1102-1108 Nelson, F., Gardner, B., Igoe, J. & Williams, A. (2009), Community based conservation and Maasai livelihoods in Tanzania, IN Homewood, K., Kristjanson, P. & Trench, P. (Eds.) Staying Maasai?, Springer, New York, p.299-333 Nelson, N. & Wright, S. (1995), Power and Participatory Development: Theory and Practice. IT Publications, London, 225pp Neumann, R. (2004a), Moral and discursive geographies in the war for biodiversity in Africa, Political Geography, 23, (7), 813-837 Neumann, R. P. (1998), Imposing Wilderness: Struggles over Livelihood and Nature Preservation in Africa. University of California Press, 256pp Neumann, R. P. (2001), Africa's' last wilderness': Reordering space for political and economic control in colonial Tanzania, Africa, 71, (4), 641-665 Neumann, R. P. (2004b), Nature-state-territory: Toward a critical theorization of conservation enclosures, IN Peet, D. & Watts, C. (Eds.) Liberation ecologies: environment, development, social movements, 2nd edition. Routledge, London, p.179-200 Neumann, R. P. (2005), Making Political Ecology, Hodder Arnold, Human Geography in the Making, London, 213pp 278 Neumann, R. P. (2009), Political ecology: theorizing scale, Progress in Human Geography, 33, (3), 398 Ngware, S. S. A. & Chale, G. (2002) Diwani na maendeleo (Local Government Handbook), Report to Friedrich Ebert Stiftung, Dar es Salaam Ngwilizi, H. (2002) The local government reform programme in Tanzania country experience, A paper delivered by Minister of State, President's Office, Regional Administration and Local Government, Dodoma North, D. C. (1990), Institutions, institutional change, and economic performance, Cambridge University Press, Cambridge, 153pp Nshala, R. (1999), Granting hunting blocks in Tanzania: The need for reform, Lawyers' Environmental Action Team Nshala, R. (2002) Village rights relating to land management, tourism and tourist hunting, Lawyers' Environmental Action Team Nunan, F. (2006), Empowerment and institutions: managing fisheries in Uganda, World Development, 34, (7), 1316-1332 Nygren, A. (2005), Community-based forest management within the context of institutional decentralisation in Honduras, World Development, 33, 639-655 Olivier de Sardan, J. (2008) Researching the practical norms of real governance in Africa, The Africa Power and Politics Programme (APPP) Discussion Paper (no.5), London Olivier De Sardan, J. P. (1999), A moral economy of corruption in Africa?, The Journal of Modern African Studies, 37, (1), 25-52 Olivier de Sardan, J. P. (2011), The eight modes of local governance in West Africa, IDS bulletin, 42, (2), 22-31 Olowu, D. (1999), Local organizations and development: The African experience, IN McGinnis, M. D. (Ed.) Polycentric governance and development: Readings from the workshop in political theory and policy analysis. University of Michigan Press, p.209- 240 Ostrom, E. (1987), Institutional arrangements for resolving the commons dilemma: some contending appraoches, IN McCay, B. J. & Acheson, J. M. (Eds.) The question of the commons: the culture and ecology of communal resources. University of Arizona Press, Tucson, p.250-265 Ostrom, E. (1990), Governing the commons, Cambridge University Press, 280pp 279 Ostrom, E. (1992), Making the commons work: Theory, practice, and policy, IN Bromley, D. W. & Feeny, D. (Eds.) Making the Commons Work: Theory, Practice, Policy. ICS Press, San Francisco Ostrom, E. (2002), The drama of the commons, National Academy Press, Washington DC, 525pp Ostrom, E., Burger, J., Field, C. B., Norgaard, R. B. & Policansky, D. (1999), Revisiting the commons: local lessons, global challenges, Science, 284, (5412), 278-282 Parfitt, T. (2004), The ambiguity of participation: A qualified defence of participatory development, Third World Quarterly, 25, (3), 537-556 Parsons, T. (1963), On the concept of political power, Proceedings of the American Philosophical Society, 107, (3), 232-262 Peck, J. & Tickell, A. (2002), Neoliberalizing space, Antipode, 32, (3), 380-404 Peet, R. & Watts, M. (Eds.) (2004) Liberation ecologies :environment, development, social movements, Routledge, London 444pp Peet, R., Robbins, P. & Watts, M. (Eds.) (2011) Global political ecology, Taylor & Francis US, New York, 444pp Peluso, N. L. (1992), The political ecology of extraction and extractive reserves in East Kalimantan, Indonesia, Development and Change, 23, (4), 49-74 Peluso, N. L. (1993), Coercing conservation?: The politics of state resource control, Global environmental change, 3, (2), 199-217 Pepinsky, H. E. (1992), Corruption, bribery and patriarchy in Tanzania, Crime, Law and Social Change, 17, (1), 25-51 Pieterse, J. N. (1998), My paradigm or yours? Alternative development, post-development, reflexive development, Development and Change, 29, (2), 343-373 Pretty, J. (2002), People, livelihoods and collective action in biodiversity management, IN O’Riordon, T. & Stoll-Kleeman, S. (Eds.) Biodiversity, sustainability and human communities: Protecting beyond the protected. Cambridge University Press, p.61-86 Pulhin, J. M. & Dressler, W. H. (2009), People, power and timber: The politics of community-based forest management, Journal of environmental management, 91, (1), 206-214 Punch, K. F. (1998), Introduction to Social Research :Quantitative and Qualitative Approaches Sage, London, 319pp Purcell, M. & Brown, J. C. (2005), Against the local trap: scale and the study of environment and development, Progress in Development Studies, 5, (4), 279-297 280 Puri, S. & Sahay, S. (2003), Participation through communicative action: A case study of GIS for addressing land/water development in India, Information Technology for Development, 10, (3), 179-199 Putnam, R. (1993), Making Democracy Work: civic traditions in modern Italy, Princeton University Press, Princeton NJ, 263pp Rangan, H. & Kull, C. A. (2009), What makes ecology 'political'?: Rethinking scale in political ecology, Progress in Human Geography, 33, (1), 28-45 Redford, K. H. & Sanderson, S. E. (2000), Extracting humans from nature, Conservation biology, 14, (5), 1362-1364 Ribot, J. C. (1999), Decentralisation, participation and accountability in Sahelian forestry: Legal instruments of political-administrative control, Africa, 69, (1), 23-65 Ribot, J. C., Agrawal, A. & Larson, A. M. (2006), Recentralizing while decentralizing: How national governments re-appropriate forest resources, World Development, 34, (11), 1864-1886 Robbins, P. (2000), The rotten institution: corruption in natural resource management, Political Geography, 19, (4), 423-443 Robbins, P. (2004), Political ecology: A critical introduction, Wiley-Blackwell, Oxford, 288pp Robson, E. (1994), From teacher to taxi driver: reflections on research roles in developing areas, IN Robson, E. & Willis, K. (Eds.) Postgraduate Fieldwork in Developing Areas: a rough guide. Developing Areas Research Group of the Institute of British Geographers, p.36-59 Robson, E. & Willis, K. (Eds.) (1994) Postgraduate Fieldwork in Developing Areas: a rough guide, Developing Areas Research Group of the Institute of British Geographers, Monograph no. 8 Rocheleau, D. E. (2008), Political ecology in the key of policy: From chains of explanation to webs of relation, Geoforum, 39, 716-727 Roe, D. (2008), The origins and evolution of the conservation-poverty debate: A review of key literature, events and policy processes, Oryx, 42, (4), 491-503 Roe, D., Nelson, F. & Sandbrook, C. (2009) Community management of natural resources in Africa: Impacts, experiences and future directions, Natural Resource Issues, International lnstitute for Environment and Development, London, UK Roe, E. M. (1991), Development narratives, or making the best of blueprint development, World Development, 19, (4), 287-300 281 Rose-Ackerman, S. (1999), Corruption and government: Causes, consequences, and reform, Cambridge University Press, Cambridge, 267pp Rose, G. (1997), Situating knowledge: Positionality, reflexivity & other tactics, Progress in Human Geography, 21, (3), 305-320 Sabatier, P. A. (1988), An advocacy coalition framework of policy change and the role of policy-oriented learning therein, Policy sciences, 21, (2), 129-168 Sabatier, P. A. (1991), Toward better theories of the policy process, Political Science and Politics, 24, (2), 147-156 Sachedina, H. T. (2008), Wildlife is our oil: conservation, livelihoods and NGOs in the Tarangire ecosystem, Tanzania, PhD thesis, Schol of Geography and Environment, University of Oxford, Oxford Sahlins, M. D. (1963), Poor man, rich man, big-man, chief: political types in Melanesia and Polynesia, Comparative Studies in Society and History, 5, (3), 285-303 Sayre, N. F. (2005), Ecological and geographical scale: parallels and potential for integration, Progress in Human Geography, 29, (3), 276-290 Schafer, J. & Bell, R. (2002), The state and community-based natural resource management: The case of the Moribane forest reserve, Mozambique, Journal of Southern African studies, 28, (2), 401-421 Schlager, E. (1995), Policy making and collective action: Defining coalitions within the advocacy coalition framework, Policy sciences, 28, (3), 243-270 Schlager, E. & Ostrom, E. (1993), Property rights regimes and coastal fisheries: An empirical analysis, IN Anderson, T. L. & Simmons, R. T. (Eds.) The political economy of customs and culture: Informal solutions to the commons problem. Rowman and Littlefield Publishers, Lanham, p.13-41 Scott, J. (2001), Power, Polity Press, Cambridge, 184pp Scott, J. C. (1985), Weapons of the weak: Everyday forms of peasant resistance, Yale University Press, New Haven, 389pp Scott, J. C. (1990), Domination and the Arts of Resistance: Hidden Transcripts, Yale University Press, London, 251pp Scott, J. C. (1998), Seeing like a state: How certain schemes to improve the human condition have failed, Yale University Press, New Haven Selfa, T. & Endter-Wada, J. (2008), The politics of community-based conservation in natural resource management: a focus for international comparative research, Environment and Planning A, 40, 948-965 282 Sen, A. (1981), Ingredients of famine analysis: availability and entitlements, The Quarterly Journal of Economics, 96, (3), 433-464 Sen, A. K. & Press, O. U. (1982), Poverty and famines: an essay on entitlement and deprivation, Clarendon Press Oxford Seppälä, P. (1998) Tanzanian Local Administration: A Vehicle for Democratic Development?, Report to the Institute of Development Studies, University of Helsinki, Finnish aid in development Shackleton, C., Willis, T., Brown, K. & Polunin, N. (2010), Reflecting on the next generation of models for community-based natural resources management, Environmental Conservation, 37, 1-4 Shackleton, S. & Campbell, B. (2001) Devolution in natural resource management: institutional arrangements and power: A synthesis of case studies from southern Africa, , Report funded by USAID, CIFOR, WWF-SARPO Shanahan, E. A., Jones, M. D. & McBeth, M. K. (2011), Policy Narratives and Policy Processes, Policy Studies Journal, 39, (3), 535-561 Shauri, V. (1999) The New Wildlife Policy in Tanzania: Old Wine in a New Bottle?, Lawyers' Environmental Action Team Shiva, V. (1991), Ecology and the politics of survival: conflicts over natural resources in India, Sage, New Delhi, 365pp Shiva, V. (1993), The impoverishment of the environment: women and children last, IN Mies, M. & Shiva, V. (Eds.) Ecofeminism. Zed Books, Atlantic Highlands, NJ, Shore, C. & Wright, S. (Eds.) (1997) Anthropology of policy: Critical perspectives on governance and power, Routledge, London, 239pp Siege, L. (2001), Hunting and community based conservation in Tanzania, IN Baldus, R. D. & Siege, L. (Eds.) Experiences with community based conservation in Tanzania. Tanzania Wildlife Discussion Paper 29 Sikor, T. & Nguyen, T. Q. (2007), Why may forest devolution not benefit the rural poor? Forest entitlements in Vietnam's central highlands, World Development, 35, (11), 2010-2025 Singer, E., Frankel, M. R. & Glassman, M. B. (1983), The effect of interviewer characteristics and expectations on response, Public Opinion Quarterly, 47, (1), 68-83 Skelton, T. (2001), Cross-cultural research: issues of power, positionality and ‘race’, IN Limb, M. & Dwyer, C. (Eds.) Qualitative Methodologies for Geographers: Issues and Debates. Arnold, London, p.87-100 283 Smith, J. & Scherr, S. J. (2003), Capturing the value of forest carbon for local livelihoods, World Development, 21, (12), 2143-2160 Smith, M. W. (1995), Ethics in focus groups: a few concerns, Qualitative Health Research, 5, (4), 478-486 Smith, N. (1990), Uneven Development: Nature, capital and the production of space (2nd edition), Basil Blackwell, Oxford, 198pp Smith, N. (1992), Geography, difference and the politics of scale, IN Doherty, J., Graham, E. & Malek, M. (Eds.) Postmodernism and the social sciences. St Martin's Press, New York, p.57-79 Smith, N. (1993), Homeless/global: scaling places, IN Bird, J., Curtis, B., Putnam, T., Robertson, G. & Tickner, L. (Eds.) Mapping the futures: Local cultures, global change. Routledge, London, p.87-121 Smith, N. (2004), Scale bending and the fate of the national, IN Sheppard, E. S. & McMaster, R. B. (Eds.) Scale and geographic inquiry. Blackwell, Oxford, p.192-212 Smith, T. W. (1987), That Which We Call Welfare by Any Other Name Would Smell Sweeter an Analysis of the Impact of Question Wording on Response Patterns, Public Opinion Quarterly, 51, (1), 75-83 Smoke, P. (2003), Decentralisation in Africa: Goals, dimensions, myths and challenges, Public Administration and Development, 23, 7-16 Songorwa, A. N. (1999), Community-based wildlife management (CWM) in Tanzania: Are the communities interested?, World Development, 27, (12), 2061-2079 Songorwa, A. N. (2004), Human Population Increase and Wildlife Conservation in Tanzania: Are the Wildlife Managers Addressing the Problem or Treating Symptoms?, African Journal of Environmental Assessment and Management, 9, 49-71 Songorwa, A. N., Bührs, T. & Hughey, K. F. D. (2000), Community-based wildlife management in Africa: A critical assessment of the literature, Natural Resources Journal, (40), 604-645 Sundar, N. (2001), Is devolution democratization?, World Development, 29, (12), 2007-2023 Sunderlin, W. D., Angelsen, A., Belcher, B., Burgers, P., Nasi, R., Santoso, L. & Wunder, S. (2005), Conserving Africa's rainforests: Problems in protected areas and possible solutions, Biological Conservation, 123, 45-54 Sunseri, T. (2005), Something else to burn: Forest squatters, conservationists and the state in modern Tanzania, Journal of Modern African Studies, 43, (4), 609-640 284 Swyngedouw, E. (1997), Neither global nor local: 'glocalization' and the politics of scale, IN Cox, K. R. (Ed.) Spaces fo globalization: Reasserting the power of the local. Guildford Press, New York, p.137-166 Swyngedouw, E. (2000), Authoritarian governance, power, and the politics of rescaling, Environment and Planning D, 18, (1), 63-76 Swyngedouw, E. & Heynen, N. C. (2003), Urban political ecology, justice and the politics of scale, Antipode, 35, (5), 898-918 Tacconi, L. (2007), Decentralization, forest and livelihoods: Theory and narrative, Global Environmental Change, 17, 338-348 Taylor, R. D. (1995) Consultancy to assess the potential for community management of wildlife resources in the Lunda-Mkwambi Game COntrol Area adjacent to the Southe Eastern border of Ruaha National Park, Report to the British Development Divsion in Eastern Africa, Overseas Development Administration Terborgh, J. (1999), Requiem for nature, Island Press, Washington D.C., 235pp Theobald, R. (1990), Corruption, development and underdevelopment, Duke University Press Durham, NC, 191pp Thomas, J. W. & Grindle, M. S. (1990), After the decision: Implementing policy reforms in developing countries, World Development, 18, (8), 1163-1181 Thoms, C. A. (2008), Community control of resources and the challenge of improving local livelihoods: A critical examination of community forestry in Nepal, Geoforum, 39, 1452-1465 Threadgold, S. (2006) Habitus, governmentality and young people's engagement with September 11, Australasian Political Studies Association Conference, University of Newcastle, 25-27 September, 2006 Turner, S., Fabricius, C., Koch, E. & Magome, H. (2004), Community-based natural resource management and rural livelihoods, IN Fabricius, C., Koch, E., Magome, H. & Turner, S. (Eds.) Rights, resources and rural development: Community-based natural resource management in Southern Africa. Earthscan, London, p.44-65 Vayda, A. P. & Walters, B. B. (1999), Against political ecology, Human Ecology, 27, (1), 167-179 Véron, R., Williams, G., Corbridge, S. & Srivastava, M. (2006), Decentralized corruption or corrupt decentralization? Community monitoring of poverty-alleviation schemes in Eastern India, World Development, 34, (11), 1922-1941 285 Village of Kiwere (2002), Agreement: Forest of Kidundakiyave, produced in conjunction with Iringa District Council Virtanen, P. (2005), Community-based natural resource management in Mozambique: a critical review of the concept's applicability at local level, Sustainable Development, 13, (1), 1-12 Vyamana, V. (2009), Participatory forest management in the Eastern Arc Mountains of Tanzania: who benefits?, International Forestry Review, 11, (2), 239-253 Wainwright, C. & Wehrmeyer, W. (1998), Success in integrating conservation and development? A study from Zambia, World Development, 26, (6), 933-944 Walker, P. A. (2005), Political ecology: where is the ecology.?, Progress in Human Geography, 29, (1), 73-82 Walker, P. A. (2006), Political ecology: where is the policy?, Progress in Human Geography, 30, (3), 382 Walsh, M. (1995) The Potential for Community Management of Wildlife Resources in the Lunda-Mkwambi Game Control Area Bordering Ruaha National Park, Southern Tanzania, Report to the Overseas Development Administration Walsh, M. (1996) Assessment of the strategy for transition between REWMP (Ruaha Ecosystem Widllife Management Project) and MBOMIPA (Matumizi Bora ya Malihai Idodi na Pawaga), eport to the Overseas Develoment Association Walsh, M. (1998) Notes for MBOMIPA Project Visitors, MBOMIPA Project Report, August 1998 Walsh, M. (2003) MBOMIPA: From Project to Association and from Conservation to Poverty Reduction, MBOMIPA Final Project Report, July 2003 Walsh, M. (2012), The not-so-Great Ruaha and hidden histories of environmental panic in Tanzania, Journal of East African Studies, 6, (2), 303-335 Walsh, M. T. (2000) The development of community wildlife management in Tanzania: Lessons from the Ruaha ecosystem, paper presented at African Wildlife Management in the New Millennium, College of African Wildlife Management, Mweka, Tanzania, 13-15 December, 2000 Watts, M. (1983), Silent violence: food, famine and peasantry in northern Nigeria, University of California Press, Berkeley, 687pp Watts, M. & Peet, R. (2004), Liberating political ecology, IN Watts, M. & Peet, R. (Eds.) Liberation ecologies (2nd edition). p.3-44 286 Weible, C. M. & Sabatier, P. A. (2005), Comparing policy networks: Marine protected areas in California, Policy Studies Journal, 33, (2), 181-201 Wells, M. (1998), Institutions and incentives for biodiversity conservation, Biodiversity & Conservation, 7, 815-835 Western, D. & Wright, R. M. (1994), The background to community-based conservation, IN Western, D., Strum, S. C. & Wright, R. M. (Eds.) Natural Connections: perspectives on community-based conservation. Island Press, Washington, 1-14 Western, D., Strum, S. C. & Wright, R. M. (1994), Natural Connections: Perspectives in Community-Based Conservation, Island Press, Washington D.C., 581pp Wildlife Division (2001), Guidelines for the designation and management of wildlife management areas Wildlife Division (2003), Reference manual for implementing guidelines for the designation and managment of wildlife management area (WMAs) in Tanzania Williams, G. (2004a), Towards a repoliticization of participatory development: political capabilities and spaces of empowerment, Participation: From tyranny to transformation. 92-109 Williams, G. (2004b), Evaluating participatory development: Tyranny, power and (re)- politicisation, Third World Quarterly, 25, (3), 557-578 Williams, G., Véron, R., Corbridge, S. & Srivastava, M. (2003), Participation and power: Poor people's engagement with India's employment assurance scheme, Development and Change, 34, (1), 163-192 Wily, L. (1997) Villagers as forest managers and governments" learning to let go": the case of Duru-Haitemba and Mgori forests in Tanzania, Report to the International Institute for Environment and Development Wily, L. & Haule, O. (1995) Good news from Tanzania: village forest reserves in the making. The story of Duru-Haitemba, Forests, Trees and People Newsletter Forest and Agriculture Office, Sweden Wily, L. & Dewees, P. A. (2001a) From users to custodians: changing relations between people and the state in forest management in Tanzania, Report to the World Bank, Africa Technical Families, Environment and Social Development Unit Wily, L. A. (2000), The evolution of community based forest management in Tanzania, Proceedings of the international workshop on community forestry in Africa, Banjul, The Gambia, 26-30th April, 1999 287 Wily, L. A. (2004) Can We Really Own the Forest? A Critical Examination of Tenure Development in Community Forestry in Africa, Paper presented at the 10th biennial conference, International Association for the Study of the Commons, Oaxaca, Mexico, 9-13 August, 2004 Wily, L. A. & Dewees, P. A. (2001b) From users to custodians: changing relations between people and the state in forest management in Tanzania, Policy Research Working Paper no. 2569, World Bank Ylhäisi, J. (2003), Forest privatisation and the role of community in forests and nature protection in Tanzania, Environmental Science and Policy, 6, 279-290 Zacharia, M. & Kaihula, S. A. (2001), Community participation in the conservation and management of wildlife in Tanzania, IN Baldus, R. D. & Siege, L. (Eds.) Experiences with community based wildlife conservation in Tanzania. Tanzania Wildlife Discussion Paper 29, p.26-37 Zimmerer, K. S. (2000), The Reworking of Conservation Geographies: Nonequilibrium Landscapes and Nature-Society Hybrids, Annals of the Association of American Geographers, 90, (2), 356-369 Zimmerer, K. S. (2006), Cultural ecology: At the interface with political ecology - the new geographies of environmental conservation and globalization, Progress in Human Geography, 30, (1), 63-78 Zimmerer, K. S. & Bassett, T. J. (2003), Political ecology: an integrative approach to geography and environment-development studies, The Guilford Press, New York, 311pp Zulu, L. C. (2009), Politics of scale and community-based forest management in southern Malawi, Geoforum, 40, (4), 686-699 288 Appendix 1: Bundles of rights in natural resource management (adapted from Agrawal & Ostrom, 2001) Drawing on the common-pool resources literature, the table below outlines the continuum of different bundles of rights, made up of operational- use rights and property rights, which may be devolved to a community under CBNRM or other forms of community conservation. These bundles of rights are critical to the autonomy of the community in decision-making and are a representation of the devolution of power. Type of Right Explanation Associated Property Rights Example Withdrawal The right to enter and obtain resource units or products from the resource system (e.g. timber harvesting) Owner Proprietor Authorised claimant Authorised user The Parks and People Programme in Nepal’s Terai is a buffer zone project that permits local communities limited rights of access and withdrawal. Management decisions are retained at the level of the park authorities Management The right to regulate patterns of use and to make decisions to transform the resource (e.g. planting trees) Owner Proprietor Authorised claimant Joint Forest Management in India grants communities access and user rights. Management rights are in place but highly controversial. Little security of tenure limits property rights to that of authorized claimants Exclusion The right to determine who has access to withdrawal rights and prevent those without rights from withdrawing from the resource Owner Proprietor Community Forestry in Nepal is associated with proprietorship status for local communities. Communities make day to day management decisions governing the rules of use within their forests and also enforce sanctions upon those who break these rules Alienation The right to sell or lease the rights described above Owner Community-based forestry management projects in Tanzania are located on land owned by the communities and they have the right to lease withdrawal rights (see Blomley et al., 2008). 289 Appendix 2: Schedule of Interviews, Focus Groups and Qualitative Surveys Carried out during Fieldwork Legend: SSI Semi-Structured Interview QS Qualitative Survey FG Focus Group PO Participant Observation LA Local Assistant Report Code Category Respondent Date P1 SSI Former MEMA Project Officer 11/2010 P2 SSI Academic Researcher 15/2/2011 P3 SSI Academic Researcher 15/2/2011 P4 SSI Academic Researcher 11/2/2011 P5 SSI Academic Researcher 3/7/2010 P6 SSI Tanzania Forest Conservation Group (1) 17/6/2010 P7 SSI Academic Researcher 17/2/2011 P8 SSI Regional Wetlands Officer 6/2010 P9 SSI Regional Participatory Forest Management Officer 17/6/2010 P10 SSI District Catchment Forest Officer and Former MEMA Project Officer 24/6/2010 P11 SSI District Natural Resources Officer 02/2011 P12 Transcript District Catchment Forest Officer and Former MEMA Project Officer N/A P13 – P17 P18, P19 P20 P21, P22 P23 P24 P25 P26 P27 LA LA LA LA LA LA LA LA LA Local Assistant Reports: Benefits and Problems Village Council & VNRC relationship VNRC Unequal Benefits Wealth Poverty Scouts and Patrols Rules of the Forest Uses of CBFM Income 03/11 03/11 03/11 03/11 03/11 04/11 04/11 04/11 04/11 P28 SSI District Natural Resources Officer 15/6/2010 P29 – P33 QS 5 CBFM villages 08/10 P34 SSI Kiwere VNRC Secretary 03/11 P35 SSI Kiwere VNRC Chairman 03/11 P36 SSI Kiwere Village Chairman 03/11 P37 SSI Kiwere VNRC Assistant Secretary 03/11 P38 SSI Kiwere Village Treasurer 03/11 P39 SSI Kiwere VNRC Treasurer 03/11 P40 SSI Kiwere Village Executive Officer 03/11 P41 FG Kiwere VNRC Representatives 04/11 P42 Transcript Kiwere VNRC Secretary 03/11 P43 Transcript Kiwere Village Chairman N/A P44 Transcript Kiwere VNRC Assistant Secretary N/A P45 Transcript Kiwere Village Treasurer N/A P46 Transcript Kiwere VNRC Treasurer N/A P47 – P56 SSI Kiwere Village Residents 03/11 290 P57 FG Kiwere Village Residents 04/11 P58 P59 P60, P61 P62, P63 P64 FG FG FG FG FG Kiwere Forest User Groups: Beekeepers Charcoal Producers Tobacco farmers Women Pastoralists 03/11 03/11 03/11 03/11 03/11 P65 FG Kiwere Village Scouts 04/11 P66 PA Kiwere VNRC Responsibilities 03/11 P67 PA Kiwere Forest Patrol Systems 04/11 P68 PA Kiwere VLFR Income Systems 04/11 P69 PA Kiwere Charcoal Systems 03/11 P70 PA Kiwere Tobacco Systems 03/11 P71 PA Kiwere Beekeeping Systems 03/11 P72 PA Kiwere Wealth Ranking 03/11 P73 SSI Former MBOMIPA Project Officer 6/2/2010 P74 20/10/2010 P75 SSI Consultant (CBNRM) 08/2008 10/4/2010 P76 SSI Former PAWM Staff Member 12/2010 P77 SSI Former REWMP Project Officer 12/2010 P78 SSI Academic Researcher 17/7/2010 P79 SSI Tourist Lodge Owner 1/5/2010 P80 SSI Former Wildlife Division staff member 17/2/2011 P81 SSI Tourist Lodge Owner 02/2011 P82 SSI Wildlife Conservation Society Staff Member (1) 19/3/2010 P83 31/5/2010 P84 4/2/2011 P86 SSI Wildlife Conservation Society Staff Member (2) 21/3/2010 12/5/2011 P87 SSI District Game Officer 24/6/2010 P88 31/1/2011 P89, P90 SSI Makifu Village Chairman 8/4/11 P91, P92 SSI Makifu Village Executive Officer 13/4/11 P93, P94 SSI Makifu MBOMIPA Representative (1) 19/4/11 P95, P96 SSI Makifu Village Treasurer 21/4/11 P97, P99 SSI Makifu MBOMIPA Representative (2) 22/4/11 P100, P101 SSI Chairman of the MBOMIPA Authorised Association 26/4/2011 P102 SSI Chairman of the MBOMIPA Board of Trustees 3/2/2011 P103 SSI MBOMIPA Investor 07/2010 P104 SSI Secretary to the MBOMIPA Authorised Association 17/7/2010 P105 SSI Former Chairman to the MBOMIPA Board of Trustees 07/2010 P106 Transcript Makifu Village Chairman N/A P107 Transcript Makifu Village Executive Officer N/A P108 Transcript Makifu MBOMIPA Representative (1) N/A P109 Transcript Makifu MBOMIPA Representative (2) N/A P110 – P117 SSI Makifu Village residents (Makanisoka) 04/11 P118-P137 QS 21 MBOMIPA villages 07/2010 P138 - P145 SSI Makifu Village residents (Makifu) 4/11 P146 - P153 SSI Makifu Village residents (Makambalala B) 4/11 P154 - P159 SSI Makifu Village residents (Makambalala A) 5/11 P160, P161 FG Makifu VNRC 8/4/11 P162, P163 FG Makifu Village residents (Mahove) 05/11 P164, P165 FG Makifu Village Scouts 05/11 P166, P167 FG Makifu Village residents (Isanga) 05/11 P168 SSI Luganga Village Chairman 06/11 P169 SSI Luganga MBOMIPA Representatives 06/11 P170 SSI Mbuyuni Village Chairman 06/11 291 P171, P172 SSI Mbuyuni MBOMIPA Representatives 06/11 P173 SSI Itunundu Village Chairman 06/11 P174, P175 SSI Itunundu MBOMIPA Representatives 06/11 P178, P179 SSI Nyamahana Village Chairman 30/4/11 P180, P181 SSI Nyamahana MBOMIPA Representatives 30/4/11 P182 - P186 PO Makifu Village Meeting 4/11 P187 PO Makifu Village Meeting 03/11 P188 Transcript Nyamahana Village Chairman N/A P189 Transcript Nyamahana MBOMIPA Representatives N/A P190 PO Kiwere Village Meetings 04/11 P191 SSI Village Key informant 5/11 P192 SSI Village Key informant 5/11 P193 PA Makifu Farming Calendar 05/11 P194 PA Makifu Wealth Ranking 05/11 P195 SSI Kiwere Assistant Chairman 03/11 P196 FG Kiwere VNRC Leaders 04/11 P197 Transcript Kiwere Village Chairman N/A P198 SSI Government Advisor 09/2008 P199 SSI Government Advisor 09/2008 P200 SSI Academic Researcher 11/2010 P201 SSI Academic Researcher 07/2010 P202 SSI Tanzania Forest Conservation Group (2) 09/2008 P203 SSI Safari Tour Operator 16/6/2010 292 Appendix 3: List of Documents Collected and Analysed The documents are arranged according to their source. Government of Tanzania: • Wildlife Conservation Act (1974) • National Forest Policy of Tanzania (1998) • Wildlife Policy for Tanzania (1998) • Local Government Act (Act 6; 1999) • Land Act (1999) • Village Land Act (1999) • Forest Act (2002) • The Wildlife Conservation (Wildlife Management Areas) Regulations (2002) • Population and Housing Census: Iringa Region profile (2002) • Agricultural census: Smallholder household characteristics and access to services and natural resources (2002-3) • The Wildlife Conservation (Wildlife Management Areas) Regulations (2005) • Iringa Region: Socio-economic profile (2007) • Wildlife Policy for Tanzania (2007) • Wildlife Conservation (Non-Consumptive Wildlife Utilization) Regulations, 2007 (Amended) • Wildlife Conservation Act (2009) • Muungano wa Jumuiya Zilizoidhinishwa za Hifadhi ya Maeneo ya Wanyamapori Tanzania(Authorised Associations Consortium; 2011) Ministry of Natural Resources and Tourism: • Draft national policy for wildlife conservation and utilisation (1994) • Draft Wildlife Management Area Guidelines (2000) • Second Draft Wildlife Management Area Guidelines (2000) • Wildlife Conservation (tourist hunting) Regulations (2000) • Draft Wildlife Conservation (Wildlife Management Areas) Regulations • Wildlife Management Areas Regulations (2001) • Community-Based Forest Management Guidelines (2001) • Guidelines for the designation and management of wildlife management areas (2001) • Wildlife Conservation (tourist hunting) Regulations (2002) • Procedures for communities to enter into joint ventures in WMAs (2002) • Reference Manual for implementing guidelines for the designation and management of wildlife management area (WMAs) in Tanzania (2003) • A report on baseline information of pilot wildlife management areas in Tanzania: First draft of the Southern working group (2003) • Baseline Survey of the Eastern Pilot Wildlife Management Areas, Volume 2: Idodi- Pawaga (2003) • Participatory Forest Management: Facts and Figures (2008) 293 MBOMIPA: • MBOMIPA project induction (1998) • MBOMIPA project notes (1998) • MBOMIPA Quarterly Reports (1-6; Oct 97-Jun 99) • Roles of Village Natural Resource Committee (Sep 98) • Seminar on the Roles and Relationships between Village Governments, Village Natural Resource Committees and government departments (Nov 98) • Summary of activities to December 1998 • Report on seminars for village leaders (Feb 99) • Feasibility Report: beekeeping and honey (May 99) • Report on the workshop on participatory techniques and planning for Village Land Use Plan (May 99) • Report on the workshop on initiating Income Generating Activities (Sep, 99) • Report: Game Scout Training (Sep, 99) • Training for Game Scouts, Village Natural Resources Committees and the implementation of participation (Nov, 99) • MBOMIPA Participatory planning reports (all villages; 1999)Proposed revision of logical framework (Jan 2000) • Follow up training for Village Government and Village Natural Resource Committees (Jun 2000) • MBOMIPA review (2000) • Key issues for the MBOMIPA project (2000) • Participatory monitoring of wildlife resources (Mar 2001) • MBOMIPA constitution of registered trustees (2002) • MBOMIPA project key issues (2003) • MBOMIPA final report (2003) • Pawaga-Idodi proposed wildlife management area: Draft resource management zone plan (2006) MEMA: • Progressive report MEMA Projects 1999-2003 • MEMA Progress Report (2001) • MEMA Semi-Annual Progress Report no. 1: September 1999 – May 2000 • MEMA Semi-Annual Progress Report no. 2: May 2000 – December 2000 • Socio-Economic Baselines (Kiwere, Mfyome, Kitapilimwa, Itagutwa, Kinyang’anga; 2000) • Training Needs Assessment and Human Resource Development Plan (March 2001) • Non-wood Forest Product Baseline Survey (April 2000) • MEMA Semi-Annual Progress Report no. 3: January – June 2001 • MEMA Semi-Annual Progress Report no. 4: July – December 2001 • Review of MEMA Phase I and Formulation/Pre-appraisal of a Second Phase (September, 2001) • Assessment of Gender Issues in Participatory Forest Management (November, 2001) • MEMA Semi-Annual Progress Report no. 5: January – June 2002. 294 • MEMA Semi-Annual Progress Report no. 6: July – December 2002 • MEMA Kitapilimwa Forest Reserve. Agreement, Bylaw & Management Plan (2002) • MEMA. North Nyang’oro Woodlands Village Land Forest. Agreement, Bylaw & • Management Plan (June 2002) • MEMA South Nyang’oro Woodlands Village Land Forest. Agreement, Bylaw &Management Plan (June 2002). • MEMA: Initial learning experiences (2003) Kiwere Village Council: • Agreement: Forest of Kidundakiyave Wildlife Conservation Society Tanzania: • Community-Based Wildlife Management: MBOMIPA • Ruaha Landscape Programme Strategic Plan • Ruaha Landscape Programme Report 2008/09 Institute of Resource Assessment, University of Dar es Salaam: • Assessment and Evaluation of the Wildlife Management Areas in Tanzania (2007) Health and Development International Consultants (HDIC): • Socio-Economic Baseline Studies in Selected Wildlife Management Areas under the Financial Crisis Initiative/Cash-for-Work Program (2010)