Democratization of Wildlife Management

MISSING IMAGE

Material Information

Title:
Democratization of Wildlife Management a Comparative Study of the Community Based Natural Resource Management (CBNRM) Programs in Botswana and Zambia
Physical Description:
1 online resource (264 p.)
Language:
english
Creator:
Mupeta-Muyamwa, Patricia Chilufya
Publisher:
University of Florida
Place of Publication:
Gainesville, Fla.
Publication Date:

Thesis/Dissertation Information

Degree:
Doctorate ( Ph.D.)
Degree Grantor:
University of Florida
Degree Disciplines:
Interdisciplinary Ecology
Committee Chair:
Child, Brian
Committee Co-Chair:
Barnes, Grenville
Committee Members:
Keys, Eric
Lindberg, Staffan

Subjects

Subjects / Keywords:
cbnrm -- community -- decentralization -- democracy -- devolution -- governance
Interdisciplinary Ecology -- Dissertations, Academic -- UF
Genre:
Interdisciplinary Ecology thesis, Ph.D.
Electronic Thesis or Dissertation
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )

Notes

Abstract:
Decentralization of wildlife management in southern Africa has largely taken place through the Community Based Natural Resource Management(CBNRM) programs. Transfer of rights from state actors to local communities has been at the core of this institutional change. Following two decades of implementation however, a growing concern has been the need for establishing micro community governance systems that enable improved wildlife management,increase community incomes, provide equitable share of benefits, and promote democratic principles of governance. This study examines whether this has taken place in six communities in Botswana and Zambia. The study’s objectives include: examining the importance of Annual General Meetings as platforms for improved democratic governance; examining the factors that influence community participation in CBNRM elections and its role in improved CBNRM governance, and finally examining the macro enabling conditions that shape micro CBNRM governance and the provision of public goods. Using survey and interview data that compares single and multi-village models of governance, the study examines factors that influence community member’s participation in two democratic institutions namely Annual General Meetings (AGMs) and elections. In addition, the study examines how these six communities perform within macro policy environments that have shaped decentralization of wildlife management in the two countries of study. The findings show that single villages perform better than multi village communities in providing more equitable shares of CBNRM benefits and increasing political participation of its members at AGMs and elections. The study further reveals that despite recentralized policy directives shaping wildlife management in Botswana and Zambia, single village communities have been able to sustain their rights to manage, sell, and benefit from the resources when compared to multi villages that have experienced deterioration of these outcomes.
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Bibliography:
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis:
Thesis (Ph.D.)--University of Florida, 2012.
Local:
Adviser: Child, Brian.
Local:
Co-adviser: Barnes, Grenville.
Electronic Access:
RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2013-06-30
Statement of Responsibility:
by Patricia Chilufya Mupeta-Muyamwa.

Record Information

Source Institution:
UFRGP
Rights Management:
Applicable rights reserved.
Classification:
lcc - LD1780 2012
System ID:
UFE0044123:00001


This item is only available as the following downloads:


Full Text

PAGE 1

1 DEMOCRATIZATION OF WILDLIFE MANAGEMENT: A COMPARATIVE STUDY OF THE COMMUNITY BASED NATURAL RESOURCE MANAGEMENT (CBNRM) PROGRAMS IN BOTSWANA AND ZAMBIA By PATRICIA CHILUFYA MUPETA MUYAMWA A DISSERTATION PRESENTED TO THE GRADU ATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2012

PAGE 2

2 2012 Patricia Chilufya Mupeta Muyamwa

PAGE 3

3 To my parents C harity and David

PAGE 4

4 ACKNOWLEDGMENTS This dissertation would not have been possible without the assistance and remarkable encouragement and support of several people. I gratefully acknowledge my advisor, Professor Brian Child for being a mentor, friend a nd colleague throughout the whole process of project development and completion. Your passion and belief in the work I have undertaken over the years, has allowed me to complete this labor of love. My sincere and heartfelt thanks to all my committee member s, Professors Grenville Barnes, Eric Keys and Staffan Lindberg, for all the support and mentoring you provided. Each of you offered your expert advice and time to ensure this project was completed I would like to thank the School of Natural Resources and Environment (SNRE) and the Center for African Studies (CAS)/ African Power and Politics project for the graduate assistantship that allowed me to complete my program. Within the SNRE department I would like to thank Dr Steven Humphrey, Ms Cathie Ritchie, Ms Karen Bray and Ms Meisha Wade. Within CAS, I would like to thank Professors Leo Villaln Abraham Goldman, Ms Corinna Greene and Ms Ikeade Akinyemi. I would further like to thank the people of Khwai, Mababe, Sankuyo and Okavango Community Trust in Botswana and the Kasempa and Mulendema community members of Zambia. Your cooperation and access to your communities enabled me to conduct this research. I thank the staff at the Department of Wildlife and National Parks in Botswana and the Zambia Wildlife Authority for allowing me access to your protected areas.

PAGE 5

5 T hank you to my family and friends for the support and encouragement through this whole process. Last but not at all the least, to my husband Andrew thank you for believing in me, and always encoura ging me to a im for excellence in all I do.

PAGE 6

6 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ .......... 11 LIST OF FIGURES ................................ ................................ ................................ ........ 12 LIST OF ABBREVIATIONS ................................ ................................ ........................... 13 ABSTRACT ................................ ................................ ................................ ................... 14 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 16 CBNRM The Democratic Experiment ................................ ................................ ..... 16 The CBNRM Discourse in Southern Africa ................................ ....................... 16 Statement of the Problem ................................ ................................ ....................... 18 Literature Review ................................ ................................ ................................ .... 19 The CBNRM Debate ................................ ................................ ......................... 19 Democracy in CBNRM: Ideals o r Actuality? ................................ ..................... 22 Scholarship On Decentralization and Democracy ................................ ............ 27 What Makes Democracy Work? ................................ ................................ ....... 30 CBNRM Economic Incentives a Means for Increasing Democratic Governance ................................ ................................ ................................ ... 38 Economic measures of effective CBNRM ................................ .................. 38 Cost benefit analysis at the household level ................................ .............. 41 Cost benefit analysis at the community level ................................ ............. 42 Governance in C ommon Property Regimes ................................ ..................... 44 The common pool resource ................................ ................................ ....... 44 Common property rights institutions ................................ ........................... 45 ................................ ................................ ........... 47 Situating CBNRM Governance within Common Property Theory and Democratic Governance ................................ ................................ ............... 48 2 METHODOLOGY ................................ ................................ ................................ .... 53 The Study Area ................................ ................................ ................................ ....... 53 Botswana ................................ ................................ ................................ .......... 53 Zambi a ................................ ................................ ................................ ............. 54 Case selection Why Botswana and Zambia ................................ ..................... 57 Sample Selection ................................ ................................ ............................. 58 Sampling Design ................................ ................................ .............................. 58 Data Collection ................................ ................................ ................................ 59 Demographic Characteristics ................................ ................................ ........... 59

PAGE 7

7 Age ................................ ................................ ................................ ............. 60 Gender ................................ ................................ ................................ ....... 60 Education ................................ ................................ ................................ ... 60 Gender of household head ................................ ................................ ......... 60 Head of household ................................ ................................ ..................... 61 Period of stay in village ................................ ................................ .............. 61 Position in CBO ................................ ................................ .......................... 61 3 BRINGING DECISIONS TO THE GRASSROOTS: THE ANNUAL GENERAL MEETING ................................ ................................ ................................ ............... 66 Significance of Micro Institutions in Participatory C BNRM Governance .................. 66 Annual General Meetings ................................ ................................ ................. 67 The Institutional Importance of AGMs in CBNRM Programs in Botswana and Zambia: ................................ ................................ ................................ .. 72 Theoretical Framework ................................ ................................ ........................... 75 ................................ ............ 75 Participation in Town hall Meetings and its Importance in AGMs ..................... 79 Research Question ................................ ................................ ........................... 81 Research Sites ................................ ................................ ................................ 82 Methods ................................ ................................ ................................ .................. 84 Data Collection ................................ ................................ ................................ 84 Sampling Design ................................ ................................ .............................. 85 Results ................................ ................................ ................................ .................... 87 Attendance in CBNRM AGM ................................ ................................ ............ 88 Demographic Characteristics and AGM Attendance ................................ ........ 88 AGM attendance and age ................................ ................................ .......... 88 AGM attendance and gender ................................ ................................ ..... 89 AGM attendanc e and education ................................ ................................ 89 AGM attendance and household head ................................ ....................... 89 CBNRM Social and Economic Benefits and AGM Attendance ......................... 89 Cash benefits and AGM attendance ................................ .......................... 90 Employment benefits and AGM attendance ................................ ............... 90 S ocial benefits and AGM attendance ................................ ......................... 90 Institutional Characteristics and AGM attendance ................................ ............ 90 Age of community and AGM attendance ................................ .................... 91 Size of community and AGM attendance ................................ ................... 91 Factors Predicting AGM Participation ................................ ................................ ..... 91 Results ................................ ................................ ................................ ....... 92 Education, benefits and size as predictors of AGM participation ............... 92 Interpretations of key findings ................................ ................................ .... 93 Qualitative Data ................................ ................................ ................................ ...... 94 Community Records of AGMs Held ................................ ................................ .. 94 Factors C ontributing to Non Participation in AGMs ................................ .......... 95 Poor consultation of communities by Trust and CRB leaders .................... 96 Poor transfer of inform ation to the communities by leaders ....................... 97 Absent from AGM due to work related engagements ................................ 97

PAGE 8

8 Lack of trust in leadership ................................ ................................ .......... 98 AGMs not held ................................ ................................ ........................... 98 Discussion ................................ ................................ ................................ .............. 99 s Attendance at AGMs ................. 99 Education ................................ ................................ ................................ ... 99 CBNRM benefits ................................ ................................ ...................... 100 Size of community ................................ ................................ .................... 103 Community characteristic ................................ ................................ ......... 104 Factors that Contribute to Non Participation in AGMs ................................ .... 106 4 THE IMPORTANCE OF DEMOCRATIC ELECTIONS IN PROMOTING CBNRM GOOD GOVERNANCE ................................ ................................ ........................ 118 Role of Elections in Improving CBNRM Micro Governance ................................ .. 118 Board Elections in Botswana and Zambia ................................ ...................... 120 Theory on Participation in Elections ................................ ................................ ...... 121 Voter Turnout: Participation in Elections ................................ ......................... 123 Quality of Elections Free and Fairness ................................ ........................... 124 Factors Influencing Participation in E lections ................................ ................. 124 Methods ................................ ................................ ................................ .......... 127 Results ................................ ................................ ................................ ........... 127 Voter turnout ................................ ................................ ............................ 127 Who votes in CBNRM elections? ................................ ............................. 128 Education and voter turnout ................................ ................................ ..... 128 Gender and voter turnout ................................ ................................ ......... 129 Age and voter turnout ................................ ................................ .............. 129 Do CBNRM Benefits Influence Participation in Elections? ............................. 129 Cash benefits and voter turnout ................................ ............................... 129 Employment benefit and voter turnout ................................ ..................... 129 Social benefits and voter turnout ................................ .............................. 130 Preference of voting method ................................ ................................ .... 130 Factors Predicting Voter Turn out ................................ ................................ ... 130 Model One Does the provision of CBNRM incentives predict participation in elections? ................................ ................................ ..... 131 Summary of results ................................ ................................ .................. 132 Interpretation of results ................................ ................................ ............ 132 Model Two Does the size of the community predict participation in CBNRM elections? ................................ ................................ ............... 133 Summary of results ................................ ................................ .................. 133 Interpretation of results ................................ ................................ ............ 134 Quality of Elections ................................ ................................ ......................... 134 Discussion ................................ ................................ ................................ ............ 136 Participation in Elections ................................ ................................ ................ 136 Factor Predicting Participation in Elections ................................ .................... 140 5 DEMOCRATIC DECENTRALIZATION THE ENABLING ENVIRONMENT FOR CBNRM ................................ ................................ ................................ ................. 148

PAGE 9

9 Performance of CBNRM Programs in Southern Africa ................................ ......... 148 Theoretical Framework ................................ ................................ ......................... 150 Decentralization in CBNRM: Exploring the Concepts and Principles ............. 150 Decentralization the big picture ................................ .............................. 150 Devolution a form of decentralization ................................ ....................... 152 Decentralization and devoluti on in natural resources management ......... 154 Devolution an enabling condition of CBNRM governance ....................... 156 Institutional Changes that have Shaped CBNRN in Botswana ....................... 161 Institutional Changes that have Shaped CBNRM in Zambia .......................... 167 A Comparative Assessment of Devolution i n Botswana and Zambia ............. 173 How Have Efforts To Decentralize Wildlife Management in Botswana and Zambia, Affected Micro level Governance Performance? ................................ 177 Integrating Democratic Governance and Socioeconomic Indicators to Assess Performance of CBNRM Communities within Uncertain Macro Institutional Frameworks ................................ ................................ ............. 177 Methods an d Data Analysis ................................ ................................ ............ 178 Results ................................ ................................ ................................ ........... 180 Assessment of democratic governance performance .............................. 180 Political participation ................................ ................................ ................ 180 Political rights ................................ ................................ ........................... 180 Accountability ................................ ................................ ........................... 181 Provision of information ................................ ................................ ............ 181 Assessing socioeconomic performance ................................ ................... 182 The Zambian cases: why Kasempa comm unity has had governance success and Mulendema continues to struggle ................................ .... 182 The case of OCT and why it continues to perform poorly when compared to single village communities in Botswana ........................... 183 revenues ................................ ................................ ............................... 184 Methods and Data Collection ................................ ................................ ......... 184 Revenue trends Khwai trust (2000 2005) ................................ ............ 185 Revenue trends Mababe trust (2000 2005) ................................ ........ 185 Revenue trends Sankuyo trust (2000 2005) ................................ ....... 186 Revenue trends Mulendema CRB (2003 2008) ................................ .. 186 Revenue trends Kasem pa CRB (2005 2010) ................................ ...... 187 Discussion ................................ ................................ ................................ ............ 187 6 CONCLUSIONS ................................ ................................ ................................ .... 198 The Impo rtance of Micro and Macro Governance Institutions in Shaping CBNRM Performance ................................ ................................ ........................ 198 Role of Micro Governance Institutions and Recommendations for Improved CBNRM governance ................................ ................................ ................... 199 Macro Enabling Conditions for CBNRM Governance ................................ ..... 201 Lessons from Namibia and Zimbabwe and Future Implications for CBNRM in Botswana and Zambia ................................ ................................ ............. 202

PAGE 10

10 APPENDIX A DEMOCRATIC GOVERNANCE QUESTIONNAIRE BOTSWANA ....................... 206 B DEMOCRATIC GOVERNANCE QUESTIONNAIRE ZAMBIA .............................. 229 LIST OF REFERENCES ................................ ................................ ............................. 248 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 264

PAGE 11

11 LIST OF TABLES Table page 2 1 Demographic characteristics ................................ ................................ .............. 62 2 2 CBO characteristics ................................ ................................ ............................ 63 3 1 Typologies of participation ................................ ................................ ................ 110 3 2 CBO characteristics ................................ ................................ .......................... 111 3 3 AGM attendance variables measured ................................ .............................. 112 3 4 Crosstab AGM attendance ................................ ................................ ............... 113 3 5 Description of variables in model ................................ ................................ ...... 115 3 6 Logistic regression model on factors that predict AGM attendance .................. 115 3 7 Record of AGMs held ................................ ................................ ....................... 116 3 8 Reasons for not attending AGM ................................ ................................ ....... 117 4 1 Crosstabs of voter turnout ................................ ................................ ................ 142 4 2 Voter turnout & benefits ................................ ................................ .................... 143 4 3 Frequency of elections ................................ ................................ ..................... 143 4 4 Preference of voting method ................................ ................................ ............. 144 4 5 Variables in model ................................ ................................ ............................ 145 4 6 Binary equation: factor change in odds of voting ................................ .............. 145 4 7 Binary Equation: factor Change in Odds of Voting ................................ ........... 146 4 8 Q uality of elections ................................ ................................ ........................... 147 5 1 Devolved rights and micro governance of natural resources, income, benefits and political participation ................................ ................................ .................. 191 5 3 Indicators for CBNRM socio economic benefits ................................ ................ 192 5 4 Democratic governance performance ................................ ............................... 193 5 5 Socioeconomic pe rformance ................................ ................................ ............ 194

PAGE 12

12 LIST OF FIGURES Figure page 1 1 Institutional change: devolution and collective action in commons governance 52 1 2 Actors and institutional change ................................ ................................ ........... 52 2 1 Map of the Okavango delta (source Dr. Lin Cassidy, using GIS layers housed at the Okavango Research Instit ute) ................................ ................................ .. 64 2 2 Map of Kafue National Park & surrounding game management areas (Source: Zambia Wildlife Authority GIS Database 2011) ................................ .. 65 5 1 Khwai revenue ................................ ................................ ................................ .. 195 5 2 Mababe revenue ................................ ................................ ............................... 195 5 3 Sankuyo revenue ................................ ................................ .............................. 196 5 4 OCT revenue ................................ ................................ ................................ .... 196 5 5 Mulendema revenue ................................ ................................ ......................... 197 5 6 Kasempa revenue ................................ ................................ ............................ 197

PAGE 13

13 LIST OF ABBREVIATIONS ADMADE Administrative Management Design AGM Annual General Meeting CAMPFIRE Community Areas Management Programme for Indigenous Resources CBNRM Community Based Natural Resource Management CBO Community Based Organization CPR Common Pool Reso urces CRB Community Resource Board CSO Central Statistics Office DWNPS Department of Wildlife and National Parks GMA Game Management Area GRZ Government Republic of Zambia KNP Kafue National Park LIRDP Luangwa Integrated Resource Development Project NGO No n Governmental Organization OCT Okavango Community Trust ODMP Okavango Development Management Plan VAG Village Action Group ZAWA Zambia Wildlife Authority

PAGE 14

14 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy DEMOCRATIZATION OF WILDLIFE MANAGEMENT: A COMPARATIVE STUDY OF THE COMMUN ITY BASED NATURAL RESOURCE MANA G E MENT (CBNRM) PROGRAMS IN BOTSWANA AND ZAMBIA By Patricia Chi lufya Mupeta Muyamwa December 2012 Chair: Brian Child Cochair: Grenville Barnes Major: Interdisciplinary Ecology Decentralization of wildlife management in southern Africa has largely taken place through the Community Based Natural Resource Management ( CBNRM) programs. Transfer of rights from state actors to local communities has been at the core of this institutional change. Following two decades of implementation however, a growing concern has been the need for establishing micro community governance s ystems that enable improved wildlife management, increase community incomes, provide equitable share of benefits, and promote democratic principles of governance. This study examines whether this has taken place in six communities in Botswana and Zambia. T as platforms for improved democratic governance; examining the factors that influence community participation in CBNRM elections and its role in improved CBNRM governance, a nd finally examining the macro enabling conditions that shape micro CBNRM governance and the provision of public goods.

PAGE 15

15 Using survey and interview data that compares single and multi village models of governance, the study examines factors that influence participation in two democratic institutions namely Annual General meetings (AGMs) and elections. In addition, the study examines how these six communities perform within macro policy environments that have shaped decentralization of wi ldlife management in the two countries of study. The findings show that single villages perform better than multi village communities in providing more equitable shares of CBNRM benefits and increasing political participation of its members at AGMs and ele ctions. The study further reveals that despite recentralized policy directives shaping wildlife management in Botswana and Zambia, single village communities have been able to sustain their rights to manage, sell, and benefit from the resources when compar ed to multi villages that have experienced deterioration of these outcomes.

PAGE 16

16 CHAPTER 1 INTRODUCTION CBNRM The Democratic Experiment The CBNRM Discourse in Southern Africa The Community Based Natural Resource Management (from this point CBNRM) approa ch is a democratic experiment aimed at achieving both conservation and development objectives in common property regimes ( Anstey 2009 ) Informed by general scholarship on property rights and institutions, it is institutional change that devolves the power to manage common pool natural resources to democratically elect ed local communities ( Ostrom 1999 Murphree 2000 ) This institutional change, known as democratic decentralizat ion, has shifted the traditional role of the State from that of centralized management to more participatory democratic approaches ( Mohan 2000 Larson 2004 ) This change has been implemented against a global backdrop of change in the way conservation and development is addressed. It is a response to the many challenges that have surfaced in the conservation of c ommon pool resources in nearly all parts of the world. National governments in nearly all developing countries have turned to local level common property institutions in the past few decades as a new policy thrust to decentralize the governance of the envi ronment ( Agrawal 2001 a Agrawal 2 001 b ) resources ultimately depends on sustainable institutions that can effectively manage these resources. Institutions of collective action such as CBNRM are critical in shap ing this. Comprehensive studies have been undertaken to support this in different parts of the world which support theoretical findings of collective action and identify conditions

PAGE 17

17 under which groups of self organized users are successful in managing their commons ( Ostrom 1990 Baland and Platteau 1996 ) and ( Agrawal 2001 p. 16 50 ) Scholarship on the governance of the commons shows that this response is necessary ( Agrawal 2001p. 1649 ) and reflects on larger scholarship of the commons that have over time illustrated that resource users often create institutional arrangements and management regimes that help allocate benefits equitably over long periods of time ( McKean 1992 Ostrom and Schlager 1992 Agrawal and Gibson 1999 Acheson 2006 ) against threats of resource degradation by devel oping and maintaining self governing institutions ( McCay and Acheson 1987 Ostrom 1990 Baland and Platteau 1996 ) Often using informal strategies for to rules and subtle social sanctions, community based institutions can navigate these governance systems ( Deitz, Ostrom et al. 2003 ) However, in as m uch as there is an advocacy for such self governing community groups, it is important to acknowledge that there are limitations to this model too. Barrett and colleagues caution that community based natural resource management overemphasizes the place of local communities in tropical conservation efforts, just as the previous top down model underemphasized scales and institutional landscapes to suit the biophysical and so cioeconomic context and commonly involving distributing authority across multiple institutions rather than concentrating on one ( Barrett and Arcese 1995 Barrett, Brandon et al. 2001 p. 497 ) articulates this same argument, emphasizing the importance of having a systematic

PAGE 18

18 appro ach, which provides for congruence and connectivity in matching functional, ecological and jurisdictional scales ( Murphree 2000 p. 15 ). These scholars thus convince us not to view commons governance in isolation, and to retract from previous works that have overemphasized the role of local communities but to understand that this form of governance operates within other nested institutions ( Ostrom 1990 ) Agrawal also points out shortcomings in how commons regimes have been studied thus far ( Agrawal 2001 p. 1650 ) He argues that existing studies of sustainable institutions around common pool resources s uffer from the lack of being substantive in their single focus on local institutions and resources. These studies have not articulated well enough aspects of resource systems, some aspects of user group membership and the external social, physical and inst itutional environment that affect institutional durability and long term management ( Agrawal 2001 p. 1650 1651 ) This scholarship has come a long way and through an adaptive process institutions of governance for common property regimes are being better understood than they were over two decad es ago when this form of resource governance was first advocated. Statement of the Problem The African continent has not been excluded in the global challenges of developing democratic institutions for the management of their common pool resources ( Murphree 2000 Hutton, Adams et al. 2005 ) Africa is a vast continent, and home to countries embrace this institutional change of democratic decentralized natural resource management. Two decades of implementing this resource governance regime in

PAGE 19

19 regions of both Souther n and East Africa has scored some results in meeting both conservation and development objectives, but has also received huge criticism. Among the many criticisms of these program has been the issue of poor institutional southern Africa is the incomplete devolution of rights that allow communities to govern and benefit from their natural resources. Very few studies exist that explore the main factors that have led to this incomple te devolution of rights and how in turn this has affected the target communities. This doctoral research examines governance systems in CBNRM and their role in providing public goods. The study, presented as three coauthored papers focuses on three objecti ves: Determining the importance of Annual General Meetings as platforms for improved democratic governance; Identifying the factors that influence community participation in CBNRM elections and its role in improved CBNRM governance; Determining the macro enabling conditions that shape micro CBNRM governance and the provision of public goods. Literature Review The CBNRM Debate In the wake of failed fortress conservation strategies and policies, a new narrative vocated for over the past two decades. This new conservation narrative is based on the premise that conservation cannot and should not be pursued against the interests and wishes of local people ( Ghimire and Pimbert 1997 ) It has become so widely accepted that it is now almost everywhere dominant, the obvious answer to the dilemmas and disappointments of conservation policy particularly in the Third World ( Adams and Hulme 2001 )

PAGE 20

20 Community Based Natural Resource Management (from this point on CBNRM), which aims to achieve rural development through the use of wildlife or other natural resources in places connected to protected areas ( Adams and Hulme 2001 ) Other scholars such as ( Barrow and Murphree 2001 ) argue that CBNRM is community conservation that seeks to create an enabling legal and policy context for local people to: manage their own resources sustainably and encourage the development of wildlife off take, safari hunting and tourism in communal lands; establish institutions for effective lo cal management of natural resources; and ensure the benefits accrue on a sustainable and equitable basis. For the purpose of this research I focus on the latter two objectives of CBNRM. So regardless of these definitions and objectives of CBNRM, what is th is concept that has been promoted in both conservation and development arenas, and is gaining ground almost as a panacea of solving conservation and development issues, particularly in developing countries. CBNRM can be theoretically explained as a common property arrangement that aims to understand how humans behave when resource rights to manage common pool resources such as wildlife, fisheries, water, conditions under which users of renewable resources cooperate to achieve efficient management ( Agrawal 2003 ) Deitz and his colleagues argue that humans using resources of this type face at least two underlying incentive problems, the fi rst is the problem of overuse, congestion or even destruction and the second is the free rider problem that stems from the cost or difficulty of excluding some individuals from the

PAGE 21

21 benefits generated by the resource ( Deitz, Dolsak et al. 2002 ) Incentives are thus a crucial element in the sustainability of programs that involve CBNRM. With as many proponents as it has attracted, CBNRM has also gained extreme criticisms ( Noss 1991 Wells and Brandon 1992 Barrett and Arcese 1995 Oates 1995 ) Critics of this new conservation approach com e from two very different positions. Some come from a tradition that is highly suspicious of the principles and practices of facade designed to hide old style preservat ion, with its harsh colonial legacy of policing, eviction and misanthropy ( MacKenzie 1988 Neumann 1998 ) This view holds that, at best, community conservation is but a poor imitatio n of a genuinely democratic about and uses of nature. These critics, from a broadly human rights position, are in an alliance with other opponents of community conservatio n whose interests are very different, and who see in community conservation a fatal weakening of resolve on the part of conservationists and who fear that the preservation of species and ecosystems will be compromised by placing any measure of control in t he hands of wildlife/ resources greatest enemy local people ( Spinage 1998 ) Others have positively criticized CBNRM from within, such as ( Murphree 2000 ) whose work outlines conditions in which CBNRM m ay work, and those in which they would not. Scholar practitioners, ( Murphree 2000 Child 2004 ) propose similar conditions that are similar to ( Ostrom 1990 ) design principles of the conditions that would enable CBNRM to work most effectively. Among the factors that sustain CBNRM are strong incen tive structures for constituents involved in managing common pool

PAGE 22

22 resources. Ostrom, explores the importance of incentive structures for the user groups, in order to achieve both economic and environmental sustainability ( Ostrom 1990 ) and ( Murphree 2000 Child 2004 ) have outlined the importanc e of the link between production and benefit for the user/ community groups. Another rarely explored area that contributes to success in CBNRM programs is its resources. CBN RM governance is promoted on the principles of participatory democracy where ordinary citizens actively participate, hold their leaders accountable and are included in the management and governance of their resources. Ostrom (2007) indirectly argues that t his form of governance is a key to solving commons problems. In her experimental work on commons management, she explores how face to face interactions of group members ultimately lead to socially optimal harvesting levels. Despite the importance of these principles on democracy, there have been few studies of this kind in the context of CBNRM. In this next section, I explore the literature on the meaning of democracy in CBNRM. Democracy in CBNRM: Ideals or Actuality? What is democracy? As a starting point to understand what democracy is, I focus first on some of the work undertaken by ( Dahl 1989 ) Dahl begins by conceptualizing democra cy as both an ideal and an actuality. These are two important starting points as they help in understanding what democracy is. The ideals of democracy, as Dahl explains them, are the values people believe democracy should be able to achieve. He outlines ho w democracy creates opportunities for effective participation of citizens in political processes; contributes to equality of voting; allows for citizens to control policy agendas; and how it allows for the inclusion of adults in the process. In addition he

PAGE 23

23 understand why democracy may be an ideal. He outlines eight reasons why it is importa nt to have democracy: Democracy helps to prevent governments by cruel and vicious autocrat. Democracy is respects basic human rights Democracy guarantees its citizens a number of fundamental rights that non democratic systems do not and cannot guarantee. Democracy ensures its citizens have a broader range of personal freedoms than other governance systems. Democracy helps people to protect their own fundamental interests such as food, health, shelter etc. Democratic governments provide maximum freedom of self determination for people to live under laws of their own choosing. Democratic governments can provide a maximum opportunity for exercising moral responsibility. Democracy fosters human development more fully than any feasible alternatives ( Dahl 1971 p. 2 10 ) Dahl explains that this is an empirical assertion in principle this should be able to people who live in non democratic regimes. Only a democratic government can foster a relatively high degree of political equality. One of the most important reasons for preference of a democratic government is that it can achieve political equality among citizens to a much greater extent than any feasible alternative. As outlined above, ( Dahl 1971 p. 257 ) outlines what describes democracy by illustrating the ideals of what democracy aims to achieve in society. These attributes are thus very instrumental in di stinguishing what democracy is when this concept is being explored.

PAGE 24

24 Sartori explores the definition of democracy by explaining that democracy can be defined as a high flown name for something that does not exist. He explains that if defining democracy mere ly signifies the meaning of the word, examining the Greek ( Sartori 1962 ) This limits it only to a word definition. However, the problem of defining this word is more complicated than that and he advises that when the term is used it c learly stands for something. The question is not only: What does it mean? But also: What is the thing? Again, Sartori presents a clear distinguishing angle in terms of how democracy is conceptualized in terms of meaning. He outlines that it not only define s a descriptive or denotative function, but also a normative and persuasive function. Sartori definition as majority rule that protects minority rights he proposes th is as a correct interpretation and a working solution. In sum, he describes modern democracies as hinging on the following: Majority rule Elective mechanisms Representative transmission of power. This means that the section of people who count are above all those who constitute the victorious voting majority; that as far as the actual wielding of power goes even they count only partially; and that a series of mechanisms modifies and reduces the degree of control that is left in the hands of the governed who are further removed from the levers of authority. In his conclusions, Sartori cautions that, whenever we judge democracy by the standard of etymological definition, the inadequacies that emerge lie in the definition and not in the reality. The etymolog ical definition provides us a start. For

PAGE 25

25 and the legitimacy of power. It means that in a democracy power is legitimate only if it is attributed from below, only if it i s an emanation of popular will in other words only if it is granted freely ( Sartori 1962 ) Adcock and Collier in their work on the democracies and dichotomies, also contribute to the debate on what democracy is. They examine how scholarship on democracy has been one based on defining it as a graded versus a d ichotomous concept ( Adcock and Collier 2001 ) They examine work by scholars of democracy and ask the basic question of whether democracy is viewed a s a concept that is graded or one that is dichotomous. In using the graded approach to defining democracy, an understanding of what democracy is, is thus examined using different degrees of democracy. Therefore one democracy would be different from another based on the degree of whether minimal standards such as the ones proposed by in his work on polyarchy are met ( Dahl 1989 ) The oth er option is to define democracy in terms of dichotomies, i.e. comparing what a democracy looks like and what it does not look like. In their conclusions, they propose that when examining the meaning of democracy it is important to understand the conceptua l requirements for understanding subtypes of democracy. Schmitter and Karl are concerned with four important elements in their definition of what democracy is. These include: The contestation over policy and political competition for the office. Partici pation of citizenry through partisan forms of collective action; Accountability of rulers through mechanisms of representation and the rule of law; Protection of rights for meaningful contestation, participation and accountability. ( Schmitter and Karl 1991 p. 75 88 )

PAGE 26

26 Schmitter and Karl express that these are key attributes in defining democracy ( Schmitter and Karl 1991 ) democracy in terms of its ideals and its actuality. He points out that all definitions of democracy are a distillation of the historical trajectory and the present situation of the originating countries. He thus points out the need to study democracy across different historical/ contextual settings ( O'Donnell 1998 ) He prop oses how the emergence of democracy in different settings may generate specific characteristics. In turn, these characteristics may make it useful to distinguish among subtypes within the universe of okes a realistic definition of democracy based on polyarchy ( Dahl 1989 ) The polyarchy traits consist of: Elected officials Free a nd fair elections Inclusive suffrage. Right to run for office. Freedom of expression. Alternative information (including that) alternative sources of information exist and are protected by law. Associational autonomy. In summary, this definition of democ racy stipulates some attributes of elections (clause 1 to 4) and lists certain freedoms (clause 5 to 7) that are deemed necessary for the political process ( Dahl 1971 ) contain two things: The first ide ntifies the attributes of elections that are considered free and fair.

PAGE 27

27 These various definitions of democra cy can be confusing, particularly when attributes outlined by these scholars inform one on what a democracy looks like and what it is not ( O'Donnell 1998 ) As ( Lindberg 2006 ) democracy as a variable attribute to political systems should include an empirically oriented definition of representative liberal democracy that is general enough to apply to a variety of contexts but specific enough to discriminate against clearly nondemocratic Scholarship On Decentralization and D emocracy Democracy as outlined in the above arguments is centrally situated in terms of its ideals and its actuality. The definitional discussion of democracy expresses what it tends to achieve in terms of ensuring that collective ac tion is achieved and public goods are realized. The democratic political science theorists outline some of the important attributes that democracy has in terms of contributing to how this is achieved. A question therefore would be how do these compare or c ontrast with those of decentralization or local governance. With respect to CBNRM governance, how do micro level democratic institutions function within the broader macro institutional changes such as decentralization? CBNRM is an institutional change that is initiated through the decentralization of natural resources from the state to non state actors. A key question to explore therefore is whether the same ideals and attributes of democracy hold in decentralization of natural resources? This section explo res some of the meanings that have been put forward in terms of decentralization. Upon outlining this concept, we will then outline how this compares to what theorists of democracy have outlined.

PAGE 28

28 By definition, decentralization involves the transfer of pow er from the central government to actors and institutions at lower levels in a political administrative and territorial hierarchy ( Mawhood 1983 ) The mechanisms through which theorists believe that efficiency and equity should increase is by public decisions being brought closer and made more open and accountable to local popula tions ( Mawhood 1983 Manor 1999 Smoke 2003 Manor 2004 ) For this to happen, some authors argue that some form of downwardly accountable local representation is necessary ( Smoke 2003 ) This brings in the concept and id eology of democracy. The political process that enables this is thus one that is premised on democratic principles and ideals. Other authors such as ( Larson 2004 ) have argued that in as much as decentralization describes a tran sfer of power to actors and institutions at lower levels in a political administrative and territorial hierarchy, this definition is only convenient as it relates to the formal, legal process of decentralization. It does not include the way in which local actors make decisions without specific central authority to do so, or the way in which legitimacy of local elected authorities grows outside of any decentralization as in order to advance. Larson argues that this dynamic reality represents decentralization democracy are forged. It is then the foundation of democratic decentralization. Bard ham & Mookherjee argue that from a standpoint of politics, decentralization is viewed as an important element of participatory democracy that allows citizens to have an opportunity to communicate their preferences and views to elected officials who are

PAGE 29

29 sub sequently rendered accountable for their performance to citizens ( Bardham and Mookherjee 2007 ) Blair outlines that the devolutionary form of decentralization (in which real authority and responsibility are transferred to local bodies) is termed democratic local governance. This process involves meaningful authority devolved to local units of governance that are accessible and accountable to the local citizenry, who enjoy full political rights and liberty. He cautions that it differs from earlier efforts of decentralization, which were largely ini tiatives in public administration without any serious democratic component ( Blair 2000 p. 21 39 ) Rondinelli, one of the distinguishing scholars on the subject of decentralization, affords some distinctions within the broad term decentralization. He outlines three distinct forms of decentralization: Deconcentration seeks to shift administrative responsibilities from central ministries and departments to regional and local administrative levels. Delegation this allows na tional governments to shift management authority for specific functions to semiautonomous or parastatal ortanizations and state enterprises, regional planning and area development agencies, multi and single purpose public authorities. Devolution: where hig her level authorities recognize the formal permanent control by lower level governmental on non governmental actors ( Cheema and Rondinelli 2007 p. 2 4 ) Devolution is closely premised on the foundations of democracy. It involv es the transfer of powers and resources to local governments or actors and has been the foundation of sustainable decentralization in developing countries. Political devolution provides a legal basis for the exercise of power at the local level and enables citizens to influence local policymaking and priority setting ( Grindle 2007 )

PAGE 30

30 R ondinelli regards the decentralization/ devolution democracy connection as a hypothesis. He lists four possible democratic benefits of decentralization: It could institutionalize citizen participation in development planning and management; It could all allocation of government resources and investments; It could increase political stability and unity by giving more groups a greater stake in maintaining the political system; and It could overcome the control of local elites who may not be sensitive to the needs of the poor ( Ron dinelli 1981 p. 133 145 ) Rondinelli in his hypothesis presents us with some important points of comparison between mainstream scholarship on democracy and decentralization. The ideals of democracy as outlined by the political science scholars are imp ortant in ensuring that issues surrounding collective action in the political process are addressed. There is very little departure between what democracy is, and how it links in with the political process of decentralization. Democracy and its principles form the underlying foundation on which decentralization is exercised. It addresses the most basic need that of involving people in the political process be it through elections, policy implementation or exercise of rights and freedoms. Decentralization as a political process therefore should be exercised on the foundations of democratic principles if it is to be effective. The transfer of power from the state to the local level actors, should provide an enabling environment that ensures the attributes of d Sartori, Dahl et al., are promoted. It may not necessarily lead to democracy; however its exercise should start an institutionalization of the democratic process. What Makes Democracy W ork? To explore this q uestion, a useful starting point is to examine the conditions that enable democracy to work in terms of producing different kinds of public/ collective

PAGE 31

31 goods or solving collective action problems. This section will thus examine scholarship on democracy on the conditions that enable it to work and this will be presented to how it compares or contrasts to local governance and decentralization. Dahl outlines some broad conditions that enable democracy to work. He proposes that in order for democracy to work th ere should be democratic institutions that ensure control of military and police by elected officials; that would ensure democratic beliefs and political culture and that would ensure there is no strong foreign control hostile to democracy. These condition s focus primarily on ensuring that institutions in place are creating this enabling environment for democracy to work ( Dahl 1989 ) Secondly, Dahl (1989) proposes the importance of a modern market economy in order to enable democracy to work. He argues that a market capitalist economy, the society it produces, and the economic growth it typically engenders are all highly favorable cond itions for developing and maintaining democratic political institutions. the conditions should be such that there is a regime that has competitive elections for presidency, rotation in the presidential office, and more than one party. They caution that while some regimes are more democratic than others, unless offices are contested, they should not be considered democratic ( Przeworski, Alvarez et al. 1996 ) Putnam undertook an intensive comparative study of Ital importance of market based economies as necessary conditions for democracy.

PAGE 32

32 Overall, Putnam services the larger debate on democratization by forwarding his views on civic virtue as a motivator for the success of democratic institutions ( Putnam 1993 ) Powell examines elections as instruments of democracy, and a core attribute to the success of whether democracy works or not. He outlines how elections are critical democratic instruments that establish connection s that compel or greatly encourage the policy makers to pay attention to citizens. He reinforces this position by arguing that there is a widespread consensus that the presence of competitive elections, more than any feature, identifies a contemporary nati on state as a democratic political system ( Powell 2000 ) Other scholars that have foc used on elections and democracy such as ( Lindberg 2006 ) outline three important dimensions that are necessary instruments for the realization of self government (or government by the people). He proposes equality of political participation; freedom of political competition; and legitimacy of the idea of self g overnment. These dimensions constitute what he proposes as central democratic qualities that are examined in elections. So how then do these attributes and conditions of democracy ensure that public/ collective goods are provided for, or collective action problems are solved? Douglas North in his work on institutions provides us with some guidance. In his chapter on political and formal rules, he explains how political rules broadly define the hierarchical structure of the polity, its basic decision structu re and the explicit characteristics of agenda control. He explains that the evolution of single absolute rulers to democratic governments is typically conceived as a move towards greater political efficiency. In the sense that democratic government gives g reater and greater percentage of the

PAGE 33

33 populace access to the political decision making process, eliminates the capricious capacity of a ruler to confiscate wealth, and develops third party enforcement contracts with an independent judiciary, resulting in a move towards greater political efficiency ( North 1990 ) This greater political effic iency can be expressed as the improvement in the provision of public goods, and one where collective action problems are solved. Democracy is thus a political institution that enables this. In order for this political institution to work however, the condi tions outlined above such as societies in market economies; the role of civil society in the process; free and fair elections; political competition; political participation; legitimacy; the promotion of political freedoms; universal suffrage; etc. should be met. So how then do the attributes of democracy, contribute to the provision of public goods or solve collective action problems in local governance and decentralization? Manor, points to three essential conditions that enable democratic local governan ce/ decentralization to work. These include: A considerable amount of resources (especially financial resources) from higher levels of government; The devolution of substantial powers to local authorities and Mechanisms that ensure accountability of bur eaucrats to elected representatives and in turn the elected accountability to voters ( Manor 2004 ) These conditio ns should be enabling factors in ensuring that democratic decentralization at local level is effective. However, most of these practices are undertaken in rural areas, where the provisions of public goods are weakened by certain conditions in the systems. Some scholars of decentralization have thus challenged the concept of democracy in providing rural people public goods and its usefulness in the area.

PAGE 34

34 Johnson examines the question of whether decentralized local governance achieves goals such as poverty re duction. This relationship parallels the question on what conditions of democracy allow for the provision of public goods. Johnson explains that democracy and decentralization are often presented as necessary conditions for effective rural development ( Johnson 2001 ) Democratic decentralization it is argued, results in a state apparatus more exposed and therefore more responsive to local needs and aspirations ( Crook and Manor 19 98 ) However, he cautions that the idea that democratic decentralization will necessarily produce gains for the poor is not entirely consistent with evidence on the ground. This is explained in the disharmony that can exist between democratic decentraliz ation and rural poverty/ how public goods are provided for to the poor. Johnson considers the challenge of encouraging decentralization and democracy in rural areas where political agency and access to information are frequently limited by traditional and modern bureaucratic systems of hierarchy and control ( Johnson 2001 ) This raises concern on some the attributes of political freedoms as key standards in the concept of democracy. In rural areas, if those conditions are not there, how then can the practice of democracy within local decentralized governance be exercised? An emerging theme from a sizeable body of literature on democratic decentralization is the weak correl ation that exists between democratic decentralization and poverty reduction ( Crook and Manor 1998 Manor 1999 Blair 2000 Crook and Sverrisson 2001 ) First they suggest that even the most successful forms of democratic decentralization have been unable to overco me economic and political disparities both

PAGE 35

35 within and among regions. This highlights the problem of raising public revenue in rural areas where economic surplus is typically sparse. Secondly, these authors imply an underlying tension between rural inequal ity and local democracy. An important concern they raise is that poverty would have a debilitating effect on the ability to engage in formal political processes. A direct illustration of this is the relationship between basic literacy and political action. As Dreze and Sen have forcefully argued, the ability to obtain and understand information about laws, policies and the rights to which one is entitled is often dependent on the ability to read ( Drez and Sen 1995 ) Thirdly, is the dilemma of encouraging poor people to assume the costs of engaging in direct political action? As Moore and Putzel have argued, agrarian institutions may be structured in a way that prevents poor people from participating in political rallies and the like. Moreover the costs of political action may deter them from pursuing or sustaining coherent pol itical movements ( Moore and Putzel 1 999 ) Finally, there is the problem of local elite capture. As numerous studies have pointed out, one of the dangers of decentralization is that it may simply empower local elites and perpetuate poverty and inequality ( Drez and Sen 1995 Crook and Manor 1998 Manor 1999 Moore and Putzel 1999 Blair 2000 ) These factors outlined above may cause one to retract to whether democracy is an ideal political institution when it comes to decentralized local governance. The above scholars on decentralization pose some thought provoking studies on some of the challenges democracy as a political institution faces when being undertaken in areas with high poverty levels. Here the provision of public goods becomes primary; ho wever

PAGE 36

36 the problems of collective action are perpetuated due to conditions on the ground. When comparing how democracy works therefore in the case of decentralized local governance, it is very important to be mindful that unlike in liberal democracies where freedoms are promoted, there are: market based capitalistic economies; elections are free and fair; there is political participation and competition; self governance is legitimate. In contrast, rural areas with high poverty levels would not usually have t hose conditions. The ideals of democracy may be promoted, and should be emphasized but achieving them at local level, in decentralized forms of governance is slightly more challenging. ship in decentralization has also shown that democratic ideals in decentralization can contribute in a number of ways to enhancing rural livelihoods/ or providing public goods to rural communities. Baland and Platteau, have explored how empowerment of loca l resource user groups can improve the ways in which local people manage and use natural resources ( Baland and Platteau 1996 ) Implicit in this process of decentralization is that local people possess the knowledge, information and incentives to manage and co nserve their resources on which they depend. Democratization and empowerment of local administrative bodies can enhance participation in decision making forums, particularly among groups that have been traditionally marginalized by local political process es ( Blair 2000 Crook and Sverrisson 2001 ) Studies from Africa, Asia and Latin America have shown that the introduction of elections, systems of transparency and rights of expression and associatio n can empower poor people, enhancing their ability to participate in local decision making and

PAGE 37

37 encouraging them to hold public officials to account ( Rondinelli McCullough et al. 1989 Blair 2000 ) what makes democracy work in the context of providing collective public goods and so lving collective action problems. The definitions of democracy from contemporary scholarship were presented, with some varying degrees in the way democracy is defined. A recurring thread in the definitions is the way the different scholars characterize wha t democracy is, and the most basic minimal attributes that should be met in defining this term. The definition of decentralization in local governance was also explored, and here, democratic ideals, were highlighted as underlying measures in the decentrali zation process, particularly decentralization in the form of devolution. In the second section, we explored what makes democracy work, by focusing on some of the conditions that scholars in democracy have outlined as crucial in ensuring that democracy wor ks. As a follow, we contrasted and compared what conditions in decentralization allowed for the provision of public/ collective goods, and presented some of the studies that have been undertaken in democratic decentralization and rural poverty. Poverty all eviation is a particular focus, as it presents an excellent comparative variable, when addressing the question of how decentralization ensures that public goods for the poor are delivered. Decentralization as a political process has also been implemented m ostly in rural underdeveloped areas, as a measure of ensuring that Government gets closer to the people. In this section, we demonstrated how democratic ideals as defined and elaborated in the political science literature have been brought and practiced in the decentralization of local governance literature. These two areas

PAGE 38

38 have close comparison in terms of the use of a political institution such as democracy to provide collective public goods and solving collective action problems. However, there is some p oint of departure when it comes to the enabling conditions that ensure that democracy is achieved. In ensuring that democracy is institutionalized in decentralization process, it is therefore very important to be aware of the conditions that may or may not allow it to be pursued. CBNRM Economic Incentives a Means for I ncreasing Democratic Governance CBNRM governance is dependent on ensuring that the right incentive structures are in place and that ultimately the user groups sustainably manage the resource. In this section, we will therefore pursue two objectives. Firstly, we identify the important economic measures at both household and community level that are important in ensuring effective CBNRM governance. Secondly, we determine how these measures can be incorporated into adaptive management system that creates more effective CBNRM performance. Economic measures of effective CBNRM The assumption underlying CBNRM effectiveness is that if community/ user groups act collectively, they will ensure sustainabil ity of the resources, and they will benefit from it ( Agrawal and Ostrom 2001 Agrawal 2003 ) CBNRM effectiveness is thus ensuring that collective action is achieved in commons arrangements; however in the process community groups benefit from it economically both at their household and communal level. The important starting point in developing economic household and community measures for effective CBNRM is thus to ensure the incentive structures created in this arrangement can sustainably contribute to collective action.

PAGE 39

39 Ensuring that economic sustainability is achieved in CBNRM, involves particularly understanding four important variables. B aland and Wade in their studies on what can contribute to collective action in the commons, have outlined four main variables that should be addre ssed ( Wade 1987 Baland and Platteau 1996 ) These include: (a) an understanding of the characteristic of the resource; (b) understanding the nature of the groups (group size, heterogeneity, levels of poverty) that depend on the resource; (c) the particulars of the instit utional regimes through which resources are managed; and (d) the nature of the relationship between a group and external forces and authorities such as markets, states and technology ( Wade 1987 Baland and Platteau 1996 ) Agrawal contributes further to this, by proposing the recognition of power within the community user groups and how this will ultimately affect distribution of benefits both at household and community level ( Agrawal 2003 ) Secondly, he proposes recognition of the importance of the changing relationship between the environment and human beings who use environmenta l resources. Developing measures for effective CBNRM therefore require that one understands these important nuances that could either contribute to collective action on the commons or prevent it. Riding on this is the So what are then are the economic measures at household and community level that would contribute to effective CBNRM? We will explore this question by focusing on economic measures in wildlife as a common property resource. In Southern Africa, CBNRM has gained ground, as both a conservation and development programme, based on the transfer of property rights from the State to local communities residing around wildlife areas ( Adams and Hulme 2001 Hulme and Murphree 2001 ) High value

PAGE 40

40 resources such as wildlife and forestry are commoditized to produce revenue for both Central Governments and local communities. Pr oprietorship is central to CBNRM, and includes the rights to allocate, apportion and sell resources (e.g. set, use and sell quotas); the rights to benefit; the authority to make general management decisions and the rights of exclusion and control ( Murphree 1991 Murphree 1997 Emerton 2001 Murombedzi 2001 Child 2004 ) In CBNRM, there has been suspicion raised that material benefits resulting from tourism and consumptive wildlife use are inadequa te to compensate communities for all the costs of investing in wildlife ( Barrett and Arcese 1995 Infie ld and Namara 2001 ) Apart from a few studies ( Jansen 1990 Barnes 1995c Barnes 2001 Bond 2001 ) no rigorous analysis has been done of the financial and/ or economic benefits at the household and community levels in CBNRM. In developing measures of economi c sustainability at both these levels, it is important to begin by undertaking a simple cost benefit analysis of a program. In the case of CBNRM, the basic question is whether the costs incurred at the household/ community exceeds the benefits. If this is the case, then the incentive to manage the resource sustainably is weakened, and free riding will ultimately lead to the depletion of the resource ( Deitz, Ostrom et al. 2003 ) Emerton argues that benefit based models of CBNRM are based on incomplete understanding of economics of community conservation ( Emerton 2001 ) Secondly there is a misunderstanding on the nature of wildlife benefits. Benefit distribution is necessary, but in itself may not be a sufficient condition for communities to engage in conservation. This is dependent on: the economic costs that wildlife incurs; the form in which wildlife benefits are received; the economic activities which compete with wildlife;

PAGE 41

41 the intra inter household distribution of costs an d benefits; and on a range of external factors which all limit the extent to which communities are able to appropriate wildlife benefits as real livelihood gains ( Murombedzi 2001 ) Assessing the economic effectiveness of CBNRM will be delineated fi rst at the household and then at the community level. Cost benefit analysis at the h ousehold level For household measures, economic effectiveness can be measured using three indicators. If the costs outweigh the benefits, this is an indication of weak CB NRM effectiveness. Studies have shown that communities residing in wildlife areas usually lose crops and livestock to wildlife ( DeMotts and Hoon 2012 ) These costs often outweigh t he benefits that the household gains from wildlife common property arrangements. Cost can also be measured by assessing the opportunity costs that households incur through living in wildlife areas. Households forfeit the opportunity costs of farming when they enter into common property arrangements such as CBNRM. If this is high, this would also lead to an ineffective CBNRM program. To measure economic effectiveness at the household level, a focus on the benefits received by each household is also importan t. Benefit based approaches to community conservation are based on the economic rationale that although wildlife has a high economic value, local communities receive little and therefore have no incentive to conserve ( Emerton 2001 ) The financial impacts of such progra ms are a useful measure of benefits at the household level ( Ashley 1998 ) The key questions here are: (a) how much cash is generated locally? (b) how many local people benefit? and (c) who benefits? (particularly do the poorest and marginalized in the group benefit?). In

PAGE 42

42 her studies in Namibia, explains that th ere are three types of local cash income, which are quite different in their amount and distribution: wages of full time employees; casual earning from selling products and labour (informal sector earnings); and collective income earned by community instit utions, either from profits of community enterprises or leases and other fees paid by private enterprises ( Ashley 1998 ) Cost benefit analysis at the c ommunity level and this should be the primary benefactor in any successful CBNRM program the communi ty should also benefit collectively. Communal benefits have typically been the main focus of most CBNRM programs ( Barnes 1994 Emerton 2001 ) and this has attracted criticism as a narrow focus on communal benefits does not address distributional aspects at the household level. CBNRM programs should thus aim to simultaneously establish both household and community incenti ve based structures to be successful. Benefit based approaches have allocated substantial amounts of money to community development activities and have managed to involve local people in wildlife conservation. However, it is not self evident that sharing w ildlife revenues as development benefits will by itself lead to a net economic gain for communities in wildlife areas. These only partially address the economic issues. They do not take into account the local economic forces motivating wildlife loss. Commu nity development projects such as infrastructure construction (e.g. schools, clinics, etc.) can lead to short term improvements in conservation. However, they do not address reasons why people engage in economic activities that would destroy wildlife, whic h lies at the core of developing benefit based community incentives. It has been proposed that two main areas need to be addressed when

PAGE 43

43 examining benefits that accrue at community level ( Barnes, MacGregor et al. 2002 ) They propose a measure of financial impacts using models based on empirical data that examines financial data in CBNRM programs in Namibia. They measure fina ncial profitability (annual net income, financial rate of return, financial net present value) at the HH level as well as the economic contribution, CBNRM programs make at the community level. Economic efficiency is measured as the annual contribution to g ross and net national income, economic rate return, and economic net present. Aside from only gaining benefits, another important way of measuring economic sustainability at the community level is to assess its entrepreneurial capabilities. Antinori and Br ay, suggest that this is an important indicator of whether CBNRM is successful in the commons ( Antinori and Bray 2005 ) In their study on community forestry in Mexico, t hey identify the need, to ensure that entrepreneurial capabilities are assessed for each community. These economic measures proposed by ( Barnes, MacGregor et al. 2002 ) and ( Antinori and Bray 2005 ) are designed to directly measure economic sustainability at the community level in CBNRM. These are impo rtant indicators, and are core to the economic measures are met from a community standpoint, then this should contribute to sustaining the CBNRM commons regime and re duce the costs that most communities living with wildlife incur. It is imperative that CBNRM communities be economically efficient and for benefits outweighs costs in order for their members to meaningfully participate in democratic processes.

PAGE 44

44 Governance i n Common Property Regimes Global concerns about environmental degradation and resource depletion have stimulated the growth of scholarship on common pool resources and common property in the last two decades ( Deitz, Dolsak et al. 2002 ) Failures attributed to state management and ma rket oriented policies have made communities attractive to many policy makers as an alternatives actor to govern forests, pastures, water, fisheries and wildlife areas ( Agrawal 2003 ) Governance of wildlife commons in southe rn Africa has been practiced under the auspices of CBNRM programs. So what then are these common property arrangements, and what is the foundation of the problems they attempt to solve in resource management? Focusing on commons governance with a particula r focus on governance defined formal and informal rules that regulate the public realm ( Hyden, Court et al. 2005 ) we examine how common property literature distinguishes between rules, property rights and restrictions. Within the framework of rules and property rights, we examine why these rules in the commons are developed and what sustains them. The common pool re source Common pool resources (CPRs), such as a lake or ocean, an irrigation system, a fishing ground, a forest, is a natural or man made resource from which it is difficult to ose units unavailable to others ( Deitz, Dolsak et al. 2002 ) When the resource units produced by a common pool resource have a high value and institutional constraints do not restrict the way resou rce units are appropriated, individuals face strong incentives to appropriate resource units, leading to congestion, overuse and even the destruction of

PAGE 45

45 facing the join t users of a common pool resource is known as the appropriation problem. incentives exist in all jointly used common pool resources for individuals to appropriate more resource units when acting independently than they would if they could find some way o f coordinating their activities ( O strom 1999 ) Solving the appropriation problem is therefore a matter of solving collective action problems. Common property rights institutions The term common property implies a kind of management arrangement created by humans to manage a common pool. The diversity of property rights regimes that can be used to regulate the use of common pool resources is very large including the broad categories of government ownership, private ownership and ownership by a community ( Deitz, Dolsak et al. 2002 ) When valuable CPRs are left to an open access regime, degradation and potential destruction are potential results. The proposition in past arguments that resource users cannot themselves change from no p roperty rights (open access) to group or individual property, however can be strongly rejected on the basis of empirical evidence. Resource users through the ages have done just that ( McCay and Acheson 1987 Wade 1987 Feeny, Berkes et al. 1990 ) In our context, we focus on the property rights arrangements that hu mans devise in common pool resources held by communities. Commons theorists have identified rene wable resources cooperate to achieve efficient management or fail to do so ( Agrawal 2001 ) The property rights in common pools held by a community, are rights transferred from the state to local resource managers to manage a com mon pool resource. Property rights are instruments of society that derive their significance from

PAGE 46

46 the fact that they help some individuals form those expectations, which they can reasonably hold in dealing with others. These expectations find expressions i n the laws, customs and morals of society. An owner of property rights possesses the consent of fellowmen to allow him/her to act in a particular way ( Demsetz 1967 Schlager and Ostrom 1992 ) In common property rights therefore, these are t he rights that frame what community owners are entitled to in the management of common pool resources. For commons theorists, property rights as institutions are seen as sets of rules that define access, use, exclusion, management, monitoring, sanctioning and arbitration behaviour of users with respect to specific resources ( Schlager and Ostrom 1992 ) At the same time as such rules are significant in governing patterns of use, they are al so the principle mechanisms through which policies regarding resource management work ( Demsetz 1967 Furubotn and Pejovich 1974 ) In investigating the impact of different institutional structures on resource management, commons theorists have also shown the importance of both formal and informal institutions as they influence human behavior ( Ostrom, Gardner et al. 1994 ) In light of these arguments, understanding governance from the perspective of stewardship of both formal and informal rules, common property rights are thus the formal rules that define the management of common po ol resources. Common property rights may be articulated in the form of Government policies that explain the rights communities have access to when managing a resource. A typical example would be the common property rights that are designed in the Wildlife Act of Zambia No. 12, 1998. In this Act, the common property rights provided to local communities in Zambia are conditional and include partial rights of access to wildlife commons. The

PAGE 47

47 communities have rights to earn revenue generated from wildlife touris m. On the other hand, the communities do not have the rights to sell, manage, benefit or own this land. They therefore have only partial property rights to the wildlife commons. So once common property rights are defined, what are t he rules in a common property arrangement and why are they devised? In general, humans using common pool resources face at least two underlying incentive problems ( Ostrom, Gardner et al. 1994 ) The first subtracts from the benefits available to others. Secondly, is the free rider problem that comes from the cost or difficulty of excluding some individuals from the benefits generated by the resource? The benefits of maintaining and enforcing rules of access and exclusion go to all users regardless of whether they have paid a fair share of the costs. The institutions that humans devise to regulate the use of common pool resources must som ehow try to cope with these two basic incentive problems. They struggle with how to avoid overuse and how to ensure contributions to the strategies used to maintain both the resource and institution itself ( Deitz, Dolsak et al. 2002 ) In define: who has access to a resource; what c an be harvested from, dumped into, or engineered within a resource; who participates in key decisions about these issues and about transferring rights and duties to others ( Deitz, Dolsak et al. 2002 ) These latter rules, on who makes decisions, are the main focus of understanding governance in the commons.

PAGE 48

48 These are the political rules that decide issues of power, accountability, transparency and decision making in the commons. The study of the stew ardship of these rules, which may be both formal (e.g. local constitutions) and informal (e.g. norms of behavior) is thus where we focus our understanding of governance in common property arrangements. Scholars of the commons emphasize the importance of th ese political institutions/ rules in the commons and that institutions change mainly as a result of attempts by specific social actors, and therefore institutional change is likely to occur when relevant political actors perceive gains from such change. Th e emergence of new institutions is thus a highly political affair ( Gibson 1999 ) Further, whether new political institutions that emerge will also be efficient for a society depends on the extent to which the interest groups attempting institutional change intersect or overlap with those of the larger collective ( Agrawa l 2003 ) Situating CBNRM Governance within Common Property Theory and Democratic Governance Understanding CBNRM governance (CBNRMG) begins, by understanding the theories that define this concept. What is the causal link between Common Property and Democr atic governance, and how do these apply to CBNRMG? CBNRM is a common property regime. Understanding the CBNRMG concept using decentralization/ devolution and common property theory presents various challenges. To tease out a little more and gain some commo n denominator variables, we argue that at the core of understanding CBNRMG is the knowledge of how institutional change unfolds. How do both formal and informal institutions interact to lead to collective action on the commons? Scholars such as ( Hyden, Court et al. 2005 )

PAGE 49

49 erstanding CBNRM governance therefore involves understanding how both the formal and informal rules are used and managed on the CBNRM commons. As outlined earlier, decentralization policies create the structural institutions that shape how commons are gov erned. In particular the institutional change enabled through devolution is fundamentally built on democratic principles that should lead to collective action on the commons. This may or may not happen; therefore causation in the context of CBNRMG is not d eterministic but probabilistic. Secondly, decentralization may create institutional changes that have outcomes tha t are non linear in nature ( Figure 1 1). As shown in the diagram below, state elites begin this process by creating new formal institutions. I n the case of CBNRM in Southern Africa, this is usually decentralization of wildlife management. At the core of these institutional changes is to understand how elite state actors weigh the costs and benefits of devolving power, and depending on this, whet her they choose either to centralize or partially decentralize wildlife management to local actors. Once this institutional change has been initiated, policy directives are developed, either through administrative policy changes or new legislation, depend ing on the context. The formal institutions created in the case of CBNRM promote decentralization in the form of devolved wildlife management power to local actors. CBNRM formal institutions therefore interact with informal ones such as norms and culture t hat were practiced prior to the introduction of these formal rules. The outcome of this institutional change is a hybrid of both formal and informal institutions and this outcome varies in

PAGE 50

50 different contexts. This interaction of both formal and informal in stitutions may or may not lead to collective action on the commons. In addition to understanding CBNRMG using an institutional change lens, it is important to also identify how the actors interact and how their actions lead to certain outcomes on the Comm the importance of understanding how various actors interact within institutional constraints ( Ostrom 1990 ) Other scholars such as ( North 1990 ) emphasize the institutional changes that take place. Figure 1 2 illustrates how this interaction takes place in CBNRMG. There are four main actors that influence this institutional change on the CBNRM commons. The primary relationship is between state a ctors and local communities. Other actors who indirectly influence this institutional change include: local government; tribal authorities and special interest groups such as Non Governmental Organizations (NGOs) and the private sector. Figure 1 2 illustra tes how the actions of the various Actors (in particular state and CBNRM communities) interact, and how this could possibly lead to collective action on the Commons. As Figure 1 2 illustrates, institutional change on the CBNRM commons, can lead to three ou tcomes, namely (a) provision of CBNRM economic benefits; (b) protection of the natural resources, in this case wildlife; and (c) improved CBNRM governance. These three outcomes can broadly be defined under the term CBNRM performance as it leads to collecti ve action. A combination of these two conceptual models (Figure 1 1 & 1 2) thus provides us with the causal relationship on CBNRMG that this doctoral research examines. In

PAGE 51

51 CBNRMG both formal and informal institutions shape how collective action is achieve d on the commons. In addition, key actors shape how these institutions will evolve in the management of wildlife commons.

PAGE 52

52 Figure1 1. Institutional c hange: d evolution and c ollective a ction in c ommons g overnance Figure 1 2. Actors and institutional change

PAGE 53

53 CHAPTER 2 METHODOLOGY The Study Area Botswana The study was conducted in the north west of Botswana in and around the Okavango Delta (Figure 2 1). The Okavango Delta is a vast inland delta in the northern Kalahari that boasts a unique ecosystem and supports large and varied wildlife populations. It is an endorheic (a large internal drainage basin) in central southern Africa and is a basin that extends a cross three countries namely Angola, Botswana and Namibia ( DEA 2008 p.11 19 ) nti River which is located within the Ngamiland District of Botswana. A large majority of people living in these areas derive their livelihoods from this basin and it is economically valuable as it offers both sources of revenue from tourism and water for the region ( DEA 2 008 p. 30 ) The Delta is home to high densities of large mammals, particularly elephants and region. It is also habitat to a total of 1300 identified plant species, 7 1 fish, 33 amphibians, 64 reptiles, 444 birds and 122 mammals. The Delta has a large variation of habitat types over small distances despite it being very flat and being made up of homogeneous sand. Due to small differences in altitude of 1 2 meters, ther e are large differences in the frequency and duration of flooding that creates habitat gradients from permanent rivers and lagoons to permanent swamps with reeds and papyrus, seasonally flooded grasslands and riverine and dry woodlands ( DEA 2008 )

PAGE 54

54 Various ethnically and culturally diverse groups resid e in the Delta, and these include the San (Basarwa), Bayeyi, Bambukushu, Baherero, Batawana, Bakgalagadi and Basubiya ( DEA 2008 ) African subcontinent and in Botswana and make up to population of 1.6 million ( Taylor 2002 p. 469 ) Livelihood activities of communities in the Delta comprise tourism, livestock farming, arable agriculture, fishing, collection of veld products, crafts and pro vision of various other services ( DEA 2008 ) Th ese areas have remained areas of economic hardships, despite them also being lucrative sites of capitalist production with the tourism industry ( Taylor 2002 ) To address some of the economic hardships faced by communities on the periphery of the Delta, the Government of Botswana introduced the CBNRM program in 1989. Since its inception, the program has been implemented in a total of 19 community based organizations in the Ngamiland District ( Schuster 2007 ) Our study included 4 out of the 19 communities and these were Khwai, Mababe and Sankuyo in the north east, and Okavango Community Trust in the North West panhandle of the Delta. Sankuyo and OCT were registered as Trusts in 1995 and were among the first communities to practice CBNRM. Khwai and Mababe Trusts were established in the year 2000. These communities are sparsely populated with populations fewer than one thousand in Khwai, Mababe and San kuyo and approximately just over three thousand in OCT. The main source of revenue for CBNRM activities for all four communities is from both safari huntin g and photographic tourism ( Table 2 2 for CBO characteristics). Zambia In Zambia, the study was condu cted in the Greater Kafue National Park (KNP),

PAGE 55

55 zone areas known as Game Management Areas (GMAs). With an area size of approximately 68, 000km (22, 480 km of the Nat ional Park and 45, 406 km of the park and its GMAs extend approximately 306 km from north to south and 145km from east to west. The Kafue River is the dominant landscape, togeth er with various other tributaries. The River, has well established rapids, sandy shallows, deep pools, grassy sandbanks, palm covered islands and mature evergreen riparian woodlands on levees. The PAs relatively undisturbed ecosystem supports a vast array of various flora and fauna. It has at least 11 main vegetation types, extensive floodplains and dambos, grasslands and thickets and both Miombo and Mopane woodlands. These diverse habitats support approximately 158 species of mammals; 515 birds; 70 reptile s; 36 amphibians and 58 species of fish. The PA particularly the Park, houses among the largest antelope species (21) in Africa ( ZAWA 2011 ) Its boundaries spread across five districts and four provinces and is home to a human population of approximately 200 000 people. Over the past decade, human populations in the GMA have increased specifically settle ments and encroachments along the Park boundaries. This rise in human populations in the GMAs, has led to an increase in unsustainable harvesting of natural resources. Following a long spell of poor management and funding in the 1980s and 90s, the Park and its surrounding GMAs experienced drastic declines of wildlife populations due to high levels of both Wildlife Authority (ZAWA) received funding from the World Bank and the Norwegian Government to the improve the management of the Parks resources through both

PAGE 56

56 infrastructure and law enforcement support. Over the past eight years, this has improved the status of the park and tourism numbers have increased, as well as increased levels of wildlife populations ( Simasiku, Simwanza et al. 2008 ) Two communities around the KNP were included in our study i.e. Kasempa community in Lunga Luswishi GMA that is on the north east border of the Park and boundary (Figure 2 2 & 2 pioneer communities to implement CBNRM in the Kafue system through the Administrative Management Design (ADMADE) program. The Kasempa community is ethnically and culturally homogeneous with the predomi nant group being the Kaondes. In comparison, Mulendema community is a relatively mixed heterogeneous community with ethnic groups that include the Lenje and Kaonde Ila who are the oldest inhabitants of the area. The Lozi and (from Zimbabwe ) are ethnic groups that migrated to the area later to seek farming opportunities by taking advantage of the close proximity of the area to the capital city Lusaka ( ZAWA 2011 ) The livelihoods of both Kasempa and Mulendema residents largely include small scale agriculture, fishing, and hunting and gathering of forest products. These areas are characteri zed by high levels of poverty, food insecurity and lack of basic social services such as schools or healthcare centers. Community members previously had resource user rights through negotiated terms with their traditional leaders; these however were declar ed redundant through new institutional rules that emphasized non consumptive utilization of the Park. Despite these restrictions, the CBNRM program was developed as a strategy to provide communities in GMAs user rights of the wildlife resource, where they are supposed to

PAGE 57

57 benefit from its sale through the safari hunting industry. Under the Wildlife Act of 1998, this resource is held by Government even though the community has partial rights. The Government holds its custodial rights on behalf of the public t hrough competitive tender for hunting concessions in the GMAs. Case selection Why Botswana and Zambia Both Botswana and Zambia, have been implementing the CBNRM program since the late 1980s. In determining how to explore the relationship of CBNRM governan ce, selecting country cases that had a substantial of period of CBNRM implementation was an important criterion. These two countries, though different in various macro factors i.e. GDP, population they do however share some commonalities such as being form er British colonies, have shared boundaries, culture particularly in communities residing in the north west of Botswana and Southern parts of Zambia. These macro factors were however not the main focus of comparing the two country programs. On the basis of CBNRM, both countries began the implementation of this program 20 years ago; received initial funding through USAID; focus predominantly on wildlife as the resource (some variation in Botswana where veld products are also used, however for the purpose of this study we limited it to wildlife); target marginalized rural communities with little or no Government presence, and both programs were modeled on the CAMPFIRE program of Zimbabwe. These factors therefore provide reasonable comparative elements at the m acro level. As much as similar comparative elements are important in a study such as this, it is important that there be variation in the case selection, in order to provide a rich analytical comparative design for the study ( George and Bennett 2005 ) To gain a rich comparative design, selection of the communi ties was based on how individual communities in the study varied on the following:

PAGE 58

58 Sources of CBNRM revenues; The population size of the communities; The numbers of households; The number of years individual communities had been implementing CBNRM, R evenues earned from CBNRM tourism; Area size of the Protected Area and Finally the ethnic diversity of the communities included in the sample (Table 2 2). Sample Selection Purposive sampling was used to select the six community cases. Three multi villa ge communities were included (Kasempa, Mumbwa and Okavango Community Trust) and three single village communities (Khwai, Sankuyo and Mababe). Multi village communities consist of five villages that make up the CBNRM community. Village areas are the basic u nit of governance in the CBNRM structure and in a participatory form of governance is pivotal in decision making. Single village communities consist of households as the main units that make up the CBNRM community. The multi village communities had an aver age population of between 300 1500 households ( Zambia CSO 2010 Botswana CSO 2011 ) Single village communities had an average of appr oximately 50 households ( Schuster 2007 ) Sampling Design Questionnaire surveys were administered in a total of 526 households between May August, 2009 in Botswana and August December, 200 9 in Zambia. The survey was later administered in OCT in June 2010 and was included to improve on the comparative elements of the study. The questionnaire was initially piloted in one of the study communities. In our survey development and administration, we ensured that our questionnaire went through rigorous reliability tests, such as piloting whether the questions included in the final survey addressed the concepts in the research design.

PAGE 59

59 Due to lack of community registers or membership lists, probabil ity sampling was used. Mababe was the only village in the sample with a register however all the households in Mababe village were included in the survey. Geographical stratification for each village was first undertaken followed by random sample selection of households within the stratified areas of the village ( Kear l 1976 ) At the selected household, only one member of the household was interviewed and the criteria included those who were 18 years and above. This included both head and non heads of households. Data Collection Data was collected using three method s, democratic governance survey; review of document s; and in depth interviews ( Appendix A). Secondary/ high school graduates were trained in each of the villages in the study and they administered the final questionnaires at the household level ( Grooves, Fowler Jnr et al. 2004 ) For Botswana the survey was translated into Setswana and in Zambia they were translated into Bemba the predominantly spoken languages in the village communities. Community records included budgets, financial reports, project reports, operational reports, records of meetings held and correspondences. In depth interviews were also conducted with key informants to assess various issues around both incentives of CBNRM and partic ipatory democracy. Demographic Characteristics Demographic data collected included age; sex; gender; level of education; household heads versus non household heads; length of stay and the position in the CBO ( Table 2 1).

PAGE 60

60 Age Data was collected from rand omly selected adults of 18 years and above. The sample included adults who were 18 years and above. This was in accordance with community constitutions that stipulated that universal suffrage criteria. This has been used as age cap in other studies that ha ve examined democracy at national level (see Bratton). This was then grouped in six different age categories ( Table 2 1). The largest majority of respondents were in the 18 28 year old category (27.1%). The age group with the least number of respondents was in the 73 and above category (5.9%). Gender Gender of respondents was relatively balanced with 49.4% females and 50.6% males. Education Education grade of respondents was categorized in six groups. These ranged from those with no education to those w ith college education. The majority of respondents sampled had junior secondary education (Grade 8 10) with 28.6% of all respondents having attained at least some form of junior secondary education. 27.6% of respondents had no education and the group wit h the least number of respondents was those with college education (5.2%). Gender of household head Respondents were asked to state whether they were heads of households or not. This demographic variable is important, as household heads are instrumental pa rticipants in CBNRM activities. The premise underlying the selection of this variable is that differences in gender of household heads can influence the democratic process. 67.7% of respondents were male headed of households where as 30.5% were female

PAGE 61

61 head ed and 1.8% was female headed where the male was absent for six months or more in the year. Head of household Respondents were asked whether they were heads of households or not. This survey included both groups as stipulated in the community constitutions all members of the community regardless of status was eligible to be part of the CBNRM program. Of those that responded to this question, 52.4% were heads of households and 47.6% were not. Period of stay in village Scholarship on collective action on com mons, has articulated that community members that are born or have lived longer in the village are bound to be more actively involved in commons governance than new comers ( Ostrom 1990 ) Understanding democratic participation in CBNRM therefore requires, testing whether the length of stay in a CBNRM community affects how members participate in democracy. Respondents were therefore asked to state the number of years they had lived in the village. These were then grouped into three main categories i.e. those born in the village; those that had lived there for more than 10 years and those that had lived in the village for less than 10 years. 71% of those interviewed had been born in the village. Position in CBO 78.8% of respondents indicated that t hey were ordinary members of the community and 12.8% were members of the CBO committee. The main objective of the study was to ensure that a large percentage of respondents included ordinary members of the community as its aim is to examine how ordinary me mbers perceive CBNRM democratic governance.

PAGE 62

62 Table 2 1 Demographic c haracteristics VARIABLE FREQUENCY (N) PERCENTAGE (%) Age: 18 28 29 39 40 50 51 61 62 72 73 + 138 136 90 69 46 30 27.1 26.7 17.7 13.6 8.7 5.7 TOTAL 509 100 Education: No e ducation Primary Secondary College 139 130 208 26 27.6 25.8 41.3 5.1 TOTAL 503 Gender of head of household: Male Female Female (male missing for 6 months +) 262 118 7 68.5 30.5 1.8 TOTAL 387 Head of household : Yes No 119 250 32.2 67.8 TOT AL 369

PAGE 63

63 Table 2 2. CBO c haracteristics KHWAI MABABE SANKUYO KASEMPA MULENDEMA OCT Year Established 2000 2000 1995 2003 2003 1995 No. of Years of CBNRM 9 9 14 6 6 14 No. of Households +/ 52 55 56 1479 2500 300 No. of People +/ 600 500 700 9, 000 12, 000 3, 043 No. of Villages 1 1 1 5 5 5 No. of Ethnic Groups 1 1 3 1 5+ 3+ Protected Area Size (Km2) 1918 2181 860 13, 340 3, 091 92, 200 CBNRM Revenue Source Photogr aphic & Hunting Photographic & Hunting Photograp hic & Hunting Hunting Hunting Photographic & Hunting Ethnic groups: Khwai (Basarwa), Mababe (Basarwa), Sankuyo (Bayeyi, Basubiya & Mbukushu), Kasempa (Kaonde), Mulendema (Kaonde Ila, Lenje, Tonga, Shona & Lozi), OCT ( Basarwa, Mbukushu & Bayeyi)

PAGE 64

64 Figure 2 1 Map of the Okavango delta (source Dr. Lin Cassidy, using GIS layers housed at the Okavango Research Institute)

PAGE 65

65 Figure 2 2. Map of Kafue N ational P ark & surrounding g ame m anagement a reas (Source : Zambia Wildlife Authority GIS Databa se 2011 )

PAGE 66

66 CHAPTER 3 BRINGI NG DECISIONS TO THE GRASSROOTS: THE ANNUAL GENERAL MEETING Significance of Micro Institutions in Participatory CBNRM Governance In many areas, biodiversity and poverty co exist, as the livelihoods of the rural poor depend on the use of natural resources. While protected areas are said to be the cornerstone of conservation ( MacKinnon, MacKinnon et al. 1986 ) even with the rapid expansion of protected areas since the 1960s, only 5% of land surfaces are strictly protected. A further 7.1% are protected through IUCN Protected Area categories that imply human habitation and sustainable us e ( Jenkins and Joppa 2009 ) Thus we have no option but to simultaneously address the challenges poverty and the environment and will need to learn how to work with local people if we want to achi eve biodiversity objectives. Since the 1970s, CBNRM has emerged globally for different resources and in different regions of the world to address the failure of centralized conservation ( Pomeroy 1995 Arnold 2001 Poffenberger 2006 Charnley and P oe 2007 ) and is centered on the idea of collective action, local participation, benefit sharing and control. rules and, we would argue, financial and other decisions that affect them ( Ostrom 1990 ) and that bottom up accountability is important ( Ribot 2003 Ribot 2008 ) While these principles are clear, the operational mech anisms for achieving these outcomes are discussed far less often, including in the numerous developmental projects that have attempted to implement CBNRM ( McShane and Wells 2004 )

PAGE 67

67 In general, two local mechanisms are invoked to encourage participatory decisio n making and accountability. These include elections to the committee that manages the CBNRM program and the Annual General Meeting through which the members of the community meet to discuss and manage their affairs. In this paper we assess what factors a t both the individual and community level enhance or hinder levels of attendance at AGMs using both quantitative and qualitative data from six communities in Botswana and Zambia. Annual General Meetings CBNRM Annual General Meetings (AGMs) are the institut ion most commonly created to encourage participatory governance and accountability. Once a year 1 decisions related to CBNRM are deliberated on by a significant number of the community in a manner that should be both open and participatory. CBNRM practition ers/ scholars such as suggest that the face to face deliberation of decisions within communities is among one of the core compliance principles for effective CBNRM ( Murphree 1997 Murphree 2000 Child 2004 ) In common property theory, the principles articulated by ( Ostrom 1990 p. 180 ) the operational rules can participate in modifying the op is often not focused on rule making, it is anticipated to be the primary mechanism for making decisions about how a community allocates its scare resources. AGMS are incorporated into many CBNRM programs, at least in south ern Africa (i.e. Zambia, Zimbabwe, Botswana, Namibia), for the purpose of encouraging high 1 W e argue elsewhere that meeting only once a year is insufficient to create democracy or accountability in communities that are new to democratic practices.

PAGE 68

68 levels of participation and to achieve the goals of transparency, accountability, democracy and equity. S ome proponents of CBNRM argue that CBNRM will be effective only if decisions are made face to face at village level i.e. participatory democracy ( Child and Dalal Clayton 2004 ) However, in practice most CBNRM communities comprise several (multi ) villages and the locus of decision making is in the committee elected to represent the community, rather than the community as a whole. In such programs, the linkage between the committee and the community is operationalized primarily through the AGM where the committee theoretically seeks approval for the decisions that it recommends, and reports ba ck to the community on its performance. Despite their importance in CBNRM programs, very little is known about how AGMs actually function, and in turn how they could be managed to improve CBNRM governance. An array of studies on CBNRM have focused on bene fits ( Barrett and Arcese 1995 Bond 2001 Emerton 2001 Hulme and Murphree 2001 Infield and Namara 2001 ) and natural resource management, but few studies have examined the relationship between CBNRM perfo rmance and issues of micro level governance such as AGMs ( Anstey 2009 Child and Barnes 2010 ) and natural resource management, but few studies have examined the relationship between CBNRM performance and issues of micro level governance such as AGMs ( Anstey 2009 Child and Barnes 2010 ) The objective of this chapter, therefore, is to examine this relationship. As we have conceptualized elsewhere ( Child, Mupeta et al. In review ) CBNRM requires that we simultaneously scale up to achieve economies and ecologies of scale but we need simultaneously to scale down because equitable benefit sharing and

PAGE 69

69 genuine participation are easier to achieve with fewer people. In southern Africa, improved revenue generation and wildli fe management associated with CBNRM reflects progress in scaling up. However, the generally low level of benefit sharing and participation that we observe in CBNRM communities in southern Africa, and the problems of elite capture that are widely reported ( Charnley and Poe 2007 Fritzen 2007 Ribot 2008 ) suggest that we do not yet understand how to operationalize the process of scaling down. A major challenge facing CBNRM is to improve the distribution of benefits and levels of participation. In this chapter we examine factors that affect participation in AGMs. AGMs provide a potential mechanism 2 for deliberative participatory democracy, much in the manner of town hall democracy in the USA ( Zimmerman 1999 Bryan 2004 ) Face to face meetings like AGMS are intended to achieve participatory demo cracy, and this intention is often enshrined in the rules defining CBNRM governance even if it is practiced much less often. Borrowing several concepts from scholars of democracy such as ( Dahl 1971 ) town hall democracy ( Zimmerman 1999 Bryan 2004 ) participation ( Agrawal and Goyal 2001 ) and CBNRM ( Murphree 1991 Murombedzi 2001 Child 2005 ) we explore the fa ctors that incentivize ordinary members of CBNRM communities to participate in AGMs. This study assesses which demographic, socio economic and institutional factors predict attendance in AGMS. This study was undertaken using a comparative study of six com munities in Botswana (Khwai, Mababe, Sankuyo and Okavango Community Trust) and Zambia (Kasempa 2 This depends on how they are managed. AGMS can be used to facilitate democracy and accountability, but they can also be a smokescreen to legitimize elite capture.

PAGE 70

70 and Mulendema). A total of 526 questionnaire surveys were administered between May June, 2009 and follow ups in June, 2010. Bringing decisions to the grassroo ts So why bring decision making to the grassroots? Leading scholars of common property or CBNRM governance broadly agree that participatory models of democratic governance are important in ensuring decisions are brought close to the main constituents of t he CBNRM programs, i.e. the households ( Murphree 1997 Child 2005 Anstey 2009 ) Thus ( Murphree 2000 ) in emphasizing the importance of small local jurisdictions, suggests that the fewer members the better, the closer they live together the better, and the more they interact together on a daily basis the better. In comparison, large groups with weak resource bases or small disperse groups will have difficulty acting in cohesion. Similarly, Ostrom major impacts to face ( Ostrom 2007 ) Decisions ma de in a participatory manner allow for reduced transaction costs, controls via peer pressure are tighter and more efficient when compared to larger jurisdictions ( Murphree 2000 ) and there is the likelihood that improved decision can be made in the management of complex social ecological systems 3 While not explicitly stated, local democracy is also enshrined in the principles of CBNRM. Murphree repeatedly emphasizes the importance 3 Interestingly, Ostrom does not specifically mention democracy in her design principles for long enduring common property resource institutions (p90), though her third principle (collective choice cipate in

PAGE 71

71 of scale, face to face decision making and accountability. He suggests that there is a close inter relationship between local democracy, tenure rights and local political activism ( Murphree 1995 ) This study of micro institutions in CBNRM experiences in Southern Africa also contributes to our broader understanding of the institutions of decentralization. However, we note that the literature on decentralization usually interprets governance in terms of the delivery of services ( Rondinelli, McCullough et al. 1989 Manor 2004 ) which may differ fr om decentralization intended to empower communities to manage their resources profitably and sustainably in the manner of private game ranchers (i.e. private community proprietorship). The importance of decentralization and participation has caught on in discussion on effective local environmental decision making ( Ribot 2004 ) including the important concept of upward and downward accountability ( Ribot and Agrawal 2006 ) However, it seems to us that the details of downward accountability have not really been thought through 4 Nonetheless, both decentralized governance (aiming to deliver services), and devolved natural resource management are predicated upon the assumption of citizen empowerment, be this as voters in local government bodies or shareholders in natural resource production enterprises. 4 Scholars with as much experience as Ribot, for example, suggest that elections have some equivalency with democratic accountability Ribot, J. C. (2008). Building Local Democracy through Natural Res ource Interventions. An Environmentalist's Responsibility, World Resources Institute. Ibid. Our experience in the field suggests that elections on their own may be weak mechanisms for accountability, and even mechanisms that legitimize processes of cooption. If our purpose is to measure democracy we should d understanding, control of the agenda, inclusion of all adults Dahl, R. (1989). Democracy and It's Critics New Haven, Yale University Press. noting, incidentally, that his criteria do not mention elections. Ibid. noting, incidentally, that his criteria do not menti on elections.

PAGE 72

72 So how do supposedly understand that participatory or democratic inst itutions, including general meetings, work? Natural resources play a very important role in local and leverage over their natural resources? In the longer term we hope th at empower rural subjects democratically. Although much of the literature mentions issues like local level corruption or elite capture as serious problems in CBNRM ( Arnold 2001 Charnley and Poe 2007 ) there is surprisingly little research on these community level institutions and organization, a gap that we hope to be gin to rectify. The Institutional I mportance of AGMs in CBNRM P rograms in Botswana and Zambia: A formal requirement in Botswana and Zambia CBNRM policies ( GRZ 1998 Schuster 2007 ) is that local communities hold AGMs each year. The AGM is the mechanism through which communities are supposed to make key decisions a bout policy and resource allocation. The specifications for these meetings vary within each community, but some features are salient to CBNRM AGMs in both Botswana and Zambia. These meetings are generally supposed to be held a month after each financial ye ar. Agenda items include the annual financial report; the projected budget; annual work plans; community projects; elections of Board members; and reports on natural resources. However, the locus and mechanism of decision making is seldom clearly specified and varies, especially at village level. In ideal circumstances, these meetings are highly interactive and community members are well informed about the issues and engage in productive discussions; reports are presented; recommendations are made to

PAGE 73

73 priori tize the needs of the community; and decision making power over budgets is exercised. Two factors are critical for categorizing AGMs: the scale at which decisions are made and the type of participation by ordinary people (ranging from passive to interact ive, ( Agarwal 2001 ) We have already noted the crucial distinction between communities that are structured as single village or multi village community. With multi village communities c omprising five to eight component villages, this creates logistical challenges for all 2,000 members or so to meet face to face to deliberate at the AGM. This problem has been addressed by adopting representational forms of governance, with village partic ipation occurring either by sending recommendations to the overall CBO via elected members or by a proportion of the ordinary members attending large central AGMs 5 Participation also varies greatly in the sample communities, ranging from an elected commi ttee making decisions on behalf of the community and informing them of this decision, or even not informing them properly about decisions taken, to a highly participatory process where the whole community debates the decision together. Participation can b e seen structurally in the form of the meeting, with the former being characterized by big men standing at the front and telling communities what is going to 5 While representational governance may nominally be a form of democracy, we have concerns that scaling up may have significant costs in terms of the efficacy and equity of participation and benefit sharing. In other words, there is consi derable merit in deconstructing multi villages so that single villages are the primary units of decision making and benefit sharing (receiving the majority of the budget) while the multi level forum becomes a coordination (rather than a doing) level D alal Clayton, B. and B. Child (2003). Lessons from Luangwa. The story of the Luangwa Integrated Resource Development Project, Zambia London, International Institute for Environment and Development. Ibid. Here, for re asons of scale it is relatively easy for all the members or households within a single village to meet face to face to representational body such as the Board in multi village communities.

PAGE 74

74 happen, and the latter being more circular and discursive. However, the quality of the informatio n available to community members, the way issues are presented, the time at which they are presented, and so on also greatly affect participation. These factors are highly varied and difficult to measure. In one AGM in north east Caprivi we observed for example, the managers employed by the community ran the meeting, relegating elected representatives into the background. Despite there being considerable confusion and concern about the finances of this community, almost two days of the meeting were taken up in talking about minor issues (e.g. over an hour about a goat killed by a leopard). Just as it was getting dark on the second day and people were ready to eat and walk home, the budget was rapidly read off a typed sheet in a manner that people could n ot possibly have understood. Officially, an AGM was held and people had been involved in the budget process, and the minutes may well reflect this, but clearly this was not really the case. The exact mechanisms by which AGMs are managed vary. According t o the community constitutions in Botswana, members of the elected Board send out a written notice specifying the date, location and time of the AGMs. This announcement is published in the Kgotla (traditional village meeting place presided by the chief) not ice board 21 days before the date of the proposed AGM. It is a requirement that the Chairperson presides over the meeting and if not available his Vice Chair can undertake this role. Before the meeting is called to order, a quorum should be reached, and th ough this varies in the 4 communities sampled in this survey the average requirements is that six trustees and 40 general members of the community constitute the quorum. AGM tled

PAGE 75

75 to one vote. AGM agenda items vary but usually cover adoption of minutes, annual budget; work plans; policies on natural resources; financial reports, community projects, and elections of Board members ( Sankuyo Trust 1995 Khwai Trust 2007 ) In Zambia, AGMs are held at the level of the Village Action Group (cluster of villages made of 75 300 households) so that a multi AGM is given 14 days prior to the meeting. The meet ing agenda is announced either by word of mouth or a posting on a notice board. Similar to the case in Botswana, the meeting is overseen by the Chairperson and in his absence by the Vice Chairperson. In order for the meeting to progress it is a requirement that at least 60% of the member households or of 75 members are present. Decisions at the AGM are decided by a majority of votes. Voting is undertaken by show of hands, however if more than five members request it, voting is by secret ballot. Agenda items discussed include adoption of previous AGM minutes, sub committee reports (community projects e.g. school construction, performance of game guards etc), audit/ financial reports, work plans, budgets, bye law amendments, elections and other general busines s of the institution ( ZAWA 2003 ) Theoretical Framework wn AGMs Literat ure that examines micro liberal democracy is remarkably thin ( Zimmerman 1999 Bryan 2004 ) In CBNRM, studies of these institutions are almost non existent. Most studies of governance of CBNRM focus on macro factors such as the process of decentralization and the reluctance of central bureaucracies to really devolve rights, responsibilities and benefits

PAGE 76

76 to local communities ( Rihoy and Maguranyanga 2007 Larson and Soto 2008 Nelson and Agrawal 2008 ) There is considerable emphasis on the importa nce of the devolution of rights to use, manage and benefit from resources to the local level, but little analysis of what is required to make this local level work. Thus even the best CBNRM programmes appear to be content with representational forms of g overnance at the local level, yet our (unpublished) research from over twenty communities suggests that ordinary people are largely excluded from benefit and participation by these mechanisms ( Child and Barnes 2010 ) Nonetheless, the primary mechanism for community participation and accountability in many CBNRM programmes is the Annual General Meeting. In most guide lines and constitutions, AGMs are the primary mechanism for constituents to interact in decision making processes. Theoretically they are fundamental democratic institutions/ platforms for decision making where citizens are in sync with their governance sy stems at three levels. The first is where they are able to formulate their preferences, second are able to signify their preferences, and third they are able to have their preference weighted equally in the conduct of their governance system ( Dahl 1971 p. 2 ) stage in the genesis of a democratic culture and with less experience and capacity to use written instruments of participation like notices, minutes, financial accounts, and the press. Town hall meetings are modeled on the principles of deliberative or pa rticipatory democracy. This form of democracy arose as a response to perceived short c omings of representative democracy ( Bohman 1998 ) Deliberative democracy as it practiced in

PAGE 77

77 town hall meetings, strengthens citizen voices in governance as a consequen ce of including people of all races, classes, ages and geographies in deliberations that directly affect public decisions (Deliberative Democracy Consortium, 2003). It allows face to face deliberations on policy issues affecting both constituents and lead ers. of democracy that ensure all policy decisions are in the public interest ( Zimmerman 1999 Bryan 2004 ) No intermediaries are placed between voters an d the public decisions. In addition, these meetings maximize citizen/ constituent participation, allow ordinary voters to hold administrative officers directly accountable, provide psychological benefits to attendees, preserve local customs and perform ci tizen education and community building functions ( Zimmerman 1999 p. ix ) Critics of participatory democracy, present the counter arguments that an assembly of voters can become an ochlocracy with fears of mob rule ( Madison 1787 p. 342 ) tyranny of the majority and the rule of passion over reason. Others suggest that employ competent people ( Overdevest 2000 ) however have challenged the reasoning that this promotes mob rule or tyranny of the majority, arguing that the effectiveness of these town hall meetings is dependent on the actual quality of t he meetings ( Zimmerman 1999 ) This argument can be extended to the CBNRM debate. AGMs are instrumental in maximizing citizen participation, allowing ordinary voters to hold administrative officers directly accountable, providing psychological benefits to attendees, preserving local customs and performing citizen education and community building functions. This is enhanced by two factors: people

PAGE 78

78 meeting together to discuss issues, and clear procedural guidelines for governing debate and decision making. of citizenry it is therefore important to understand what factors actually enhance these forms of participation. The first level involves the demographic characteristics of the people attending. According to ( Zimmerman 1999 Bryan 2004 ) there is varia tion in issue on the agenda, there is variation in attendance from different gender groups. Different education levels also predict which members of the community would be more likely to attend or not. The general assumption is that members with higher levels of education are more likely to attend. Finally, age has been shown to be another younger members of the community is generally higher than older members. motivate individual to attend. CBNRM, likewise, anticipates that providing incentives to community mem bers in the form of social and economic benefits should increase interest in the CBNRM program and ultimately encourage the protection of natural resources ( Murphree 2000 Child 2005 ) Thus, incentives are a key factor in determining who attends AGMs, so we would expect members of communities who receive more social and economic benefits to attend meetings more enthusiastically than where benefits are fewer. Finally, institutional characteristics such as size, history and character of the s. In CBNRM, we would expect

PAGE 79

79 influence on the meeting, or levels trust in decision making and information, to increase participation. These factors are often subtle and complex, but smaller communities are easier to organize and collective action can therefore be achieved more effectively ( Olson 1971 ) Participation in Town hall M eetings and its I mportance in AGMs CBNRM in Southern Africa uses two primary models of governance. In single village communities AGMs are held at the village level, and all decisions are deliberated in one forum. This is described for Chikwarakwara Village in Beitbridge ( Child and Peterson 1991 ) and also for the Luangwa Valley in Zambia ( Dalal Clayton and Child 2003 ) In contrast, multi village communities hold their AGMs in a representational manner. This also takes two primary forms. In the Okavango Community Trust, for example, each of the f ive component villages holds its own AGM, and recommendations are taken forward to the Trust (meso level) where elected known CBNRM Conservancy programme, considerable logistical effort is investe d in bringing a selection of people from each constituent village to a single, central AGM where matters of community management are discussed, usually for two days. The way AGMs are run also affects outcomes and participation, but differences are subtle, we were not able to attend enough AGMs to measure them. For example, the way people place their chairs, merits more prominence. This ranges from communities sitting in a circle in which people are more equal, to the situation where the leadership has con siderable prominence at the front of a meeting that enhances the authority of the leadership relative to the followership. Another factor, which receives

PAGE 80

80 relatively, little attention, is the content of participation, including the agenda, the timing of age nda items, and the way information is presented. In some cases we have observed, considerable time is allocated to financial matters, and great care is taken to explain budgets and expenditure using flipcharts and even role plays. In other AGMs, the fina nces have been read off a piece of paper rapidly and incoherently, and in a great rush as the sun sets after two days of meeting. Scholars of participation have argued that participation in AGMs shoul d not only be measured in the passive form, i.e. numbers attending, but should include who is attending ( Pretty 1995 ) as well as their level of active participation though, again we did not measure this complex variable 6 The level of active pa rticipation provides insight into the q uality of these meetings ( Table 3 1). 6 Some authors argue that interactive participation is instrumental and allows members of the group at Narayan, D. (1996). The contribution of people's participation: Evidence from 21 rural water supply proje cts. Occassional Paper Series No. 1 Environmentally Sustainable Development Washington DC, World Bank. White, S. (1996). "Depoliticising development: The uses and abuses of participation." Development in Practice 6 (1): 6 15. and Pretty, J. N. (1995). "Participatory learning for sustainable agriculture." World Development 23 (8): 1247 1263. Agarwal, 2001, expanding on earlier work by Arnstein, S. R. (1969). "A Ladder of Citizen Participation." American Institute of Planners Journal 35 (4): 216 224, Pimbert, M. P. and J. N. Pretty (1995). Parks, People and Professionals: Putting `Participation' into Protected Area Management. Discussion Paper No 57, February 1995. Geneva, UNITED NATIONS RESEARCH INSTITUTE FOR SOCIAL DEVELOPMENT, INTERNATIONAL INSTITUTE FOR ENVIRONMENT AND DEVELOPMENT, WORLD WIDE FUND FOR NATURE. provides a useful broad typology on participation ranging from nominal, passive, consultative, activity specific, through to active and interactive participation Some authors argue that interactive participation is instrumental and allows members of the group at meetings to have voice Narayan, D. (1996). The contribution of people's participation: Evidence from 21 rural water supply projects. Occassional Paper Series No. 1 Environmentally Sustainable Developmen t Washington DC, World Bank. White, S. (1996). "Depoliticising development: The uses and abuses of part icipation." Development in Practice 6 (1): 6 15. and Pretty, J. N. (1995). "Participatory learning for sustainable agriculture." World Development 23 (8): 1247 1263. Agarwal, 2001, expanding on earlier work by Arnstein, S. R. (1969). "A Ladder of Citizen Participation." American Institute of Planners Journal 35 (4): 216 224, Pimbert, M. P. and J. N. Pretty (1995). Parks, People and Professionals: Putting `Participa tion' into Protected Area Management. Discussion Paper No 57, February 1995. Geneva, UNITED NATIONS RESEARCH INSTITUTE FOR SOCIAL DEVELOPMENT, INTERNATIONAL INSTITUTE FOR ENVIRONMENT AND DEVELOPMENT, WORLD WIDE FUND FOR

PAGE 81

81 Research on associations, interest groups, and collective action also emphasizes the importance of taking costs and benefits of participation into account ( Olson 1971 Olson 1996 Olson 1998 p. 39 50 Moe 2005 ) In addition, some of the earliest education, and organizational membership have a strong effect on the propensity to participate in political activities ( Almond and Verba 1963 ) and ( Dahl 1971 ) Other studies are concerned with questions of equity and distribution in participation ( Agarwal 2001 Botchway 2001 ) and ( Teklu and Asefa 1999 ) Some argue that participation can also be nominal only requiring members to be part of the group but not requiring active participat ion in debates or meeting agendas ( Chopra, Kadekodi et al. 1990 Molinas 1998 ) Thus participation is a complex issue. It is a basic element of collective action, but just as important are the degrees of participation, as they provide the basis to move beyond the simple act of participating to participation becoming more instrumental and allowing for marginalized groups in a society to have voice, decision making power and sanctions over the actions of the implementing committee. Measuring participation is even more difficul t it requires understanding the factors that influence how individuals participate. Research Question Notwithstanding the complexity of the issue of participation, the purpose of our study is to understand: Which demographic groupings in CBNRM communitie s attend AGMs? Is AGM participation related to the scale of individual or collective benefits? NATURE. provides a useful broad typology on participation ranging from nominal, passive, consultative, activity specific, through to active and interactive participation

PAGE 82

82 Does the size of the community affect attendance? What factors impede participation in AGMs? Research Sites We conducted our research in three small single vil lages (Khwai, Mababe and Sankuyo) and one large multi village (Okavango Community Trust) in Botswana. In Zambia, the study was conducted in two large multi village communities (Kasempa and Mulendema). To the east of the Okavango Delta in Botswana lie three small villages, which make up the Sankuyo Tshwaragano Management Trust of predominantly Bayei ethnicity, and Mababe Zokosana Community Trust and Khwai Development Trust where the people say they are proud to call themselves Bushmen. These communities wer e given the rights to the benefits from wildlife through the granting of a Head Lease. Wildlife now provides at least half of the livelihood of these villages. Most households have one or more members employed by the hunting and tourism outfitters, and a lso by their Trusts where they are employed as game guards or to manage campsites. These communities each receive some P2 million ($300,000) annually which they use variably to cover management costs, for community projects, for employment and sometimes f or cash benefits, with most of this being in the form of pensions for the elderly or indigent 100 other animals are killed annually, and this meat is delivered to the commu nities. The governance of these villages varies with time and circumstances, but in general people are moderately to well satisfied with their collective affairs. The Okavango Community Trust (OCT) comprises of some 5, 000 people in five villages located in northwest of the Okavango Delta. Although there are some 14,000

PAGE 83

83 elephants in this area, hunting is no longer practiced, and the community receives its revenues (Pula 2.5m annually) from a nearby tourism concession in the Okavango Delta. This money is managed centrally, and few benefits get to ordinary people or even the five villages, so people are generally either unaware or dissatisfied with their Trust. Mulendema Community Resource Board (CRB) is located in Mumbwa GMA, and is one of the nine GMAs ( buffer zones) surrounding the Kafue National Park in Zambia. It comprises some 1500 households in 5 Village Action Groups. The area was traditionally settled by the Kaonda ila ethnic groups, but due to its close proximity to the capital city, it has had numerous migrations from city retirees of different ethnic backgrounds. It is thus comprised of a highly heterogeneous community that has been organized to implement CBNRM. Wildlife in this area is threatened by poaching, and also by considerable recent in migration for the purpose of agriculture. The hunting is managed by the Zambia Wildlife Authority, which is supposed to return 50% of trophy fees and 20% of concession fees to the CRB, though records to support this are not available. Although there hav e been a large number of donor and NGO CBNRM initiatives in this area, progress has been negligible and people are largely unaware of or disinterested in CBRNM (Simasiku et al., 2008) Kasempa CRB is a very remote community and is located in the Lunga Luswi shi GMA that lies to the north east of Kafue National Park. It comprises approximately 700 households in 5 VAGs. Kasempa community is ethnically homogeneous and is wi ldlife population that support significant safari hunting revenues though, as in

PAGE 84

84 Mulendama, this is collected by ZAWA and far less is returned to the community than in neighbouring countries. Although Kasempa has generally not benefitted from donor suppor t, it appears to have been lucky in the quality of government officials from both ZAWA and Forestry, and procedural support of CBNR M may be a positive factor ( Table 3 2). Methods Data Collection Literature on participation and natural resource management suggest several factors that influence this (see theory section). These factors include social status of individuals; education; organizational membership ( Almond and Verba 1963 ) ; incentives/benefits for participating; household socioeconomic factors; institutional and structural factors ( Verba, S chlozman et al. 1995 ) ; costs and benefits of participation ( Olson 1971 Moe 2005 ) ; issues of equity and distribution ( Teklu and Asefa 1999 Agarwal 2001 Botchway 2001 ) ; incentives/benefits for participating, and institutional and structural factors ( Verba, Schlozman et al. 1995 ) Based on the theoretical framework on participation in democratic institutions three key factors wer e identified that could most likely predict participation in AGMs. These include demographic characteristics, the incentives that motivate individuals to attend, and the structure and size of the community. The overall hypothesis tested is that the likelih ood of participation in AGMs is higher for individuals who: Have higher levels of education Enjoy a greater level of both social and economic CBNRM benefits; Are from single village governance units

PAGE 85

85 Notwithstanding the complexity of participation ( Agarwal 2001 ) AGM participation (Table 3 2) ( Agarwal 2001 ) The independent variables describe three broad factors: demographic characteristics, CBNRM incentives, and institutional characteristics of the community, especially size. To examine factors influencing CBNRM AGM participation, b oth quantitative and qualitative data was collected. The first part of data collection and analysis includes survey data gathered using questionnaires. Survey data was collected from six communities where CBNRM has been implemented in both Botswana and Zam bia. Questionnaires were pilot tested in each community prior to the survey, in order to contextualize them. Purposive sampling was used to select the six community cases. This involved purposely selecting single village and multi village communities in t he two countries. In addition, only communities that had been practicing CBNRM for over ten years were selected to provide a reasonable timeline for implementation and to assess whether CBNRM had been institutionalized. Qualitative data was further collect ed in the form of secondary data (community reports, correspondence, meeting minutes), in depth interviews and personal observation. Sampling D esign A total of 526 questionnaire interviews were conducted between May December 2009 (Khwai, Sankuyo, Mabab e, Kasempa & Mulendema communities) and May June 2010 (OCT). Due to lack of community registers or membership lists, probability sampling of households was undertaken (quota sampling but with random or systematic

PAGE 86

86 selection within groups was used) ( Kearl 1976 p.27 30 ) and ( Grooves, Fowler Jnr et al. 2004 p. 92 93 ) For the multi village communities, geographical stratification for each village was first undertaken and then a random household sample was taken of one member per household who was 18 years or older. This included both head and non heads of households. For the single village communities, interviews were conducted in all the households. A nalysis of survey/ questionnaire data Data collected was analyzed using SPSS 16 and STATA 9. Analysis included both descriptive and logis tical regression statistics ( Table 3 14, for variables included in analysis). The first stage of the analysis includ ed an assessment using the chi square test of association using SPSS 16. This statistical test is used to test the null hypothesis that variables are independent. It compares the observed data to a model that distributes the data according to the expectati on that the variables are independent. This method only tests the probability (p value) of independence of distribution of data, however it does not tell one specific details about the relationships between variables ( Pallant 2007 ) Chi square tests were therefore conducted for this study to examine wheth er there was an association between AGM attendance and the three categories of independent variables (demography; CBNRM benefits and institutional characteristics of the community). As outlined above, the chi square test of association is a useful prelimi nary statistic; however it is limited in providing details about the relationship between variables. For further analysis in identifying the interaction of the variables that were related or had association, logistic regression tests were conducted using S TATA 9

PAGE 87

87 software. This method of analysis determines the impact of multiple independent variables presented simultaneously to predict membership of one or the other of two dependent variables. Logistic regression has several assumptions among which are that it does not assume a linear relationship between the dependent and independent variables; the dependent variable must be a dichotomy (e.g. Yes/ No) responses; the independent variables are not interval, nor normally distributed data; the categories are mu tually exclusive and the sample size has a minimum of 50 cases per predictor ( Pallant 2007 Wuensch 2009 ) In our analysis of factors that predict AGM pa rticipation all the predictor variables were categorical ( Table 3 2). Interpreting the odds ratio assessed the relative importance of each predictor variable or the interaction among predictor variables. The odds ratio is a statistic that compares whether the probability of a certain event is the same for two groups. The odds of an event or outcome happening are the probability that the event will happen divided by the probability that the event will not happen ( Westergren, Karlsson et al. 2001 ) It is calculated by using the regression coefficient of the predictor as the exponent or exp. Furthermore; it allows the calculation of the sensitivity and specificity of the model and the positive and negative predictive values ( Pallant 2007 p. 169 ) For a logistic regression, the predicted dependen t variable in our case, AGM participation is a function of the probability that a particular subject will be in one of the categories ( Wuensch 2009 ) Results

PAGE 88

88 Attendance in CBNRM AGM There was higher attendance by respondents in all three single villages (Khwai 96.2%; Mababe 82.7% and Sankuy o 76.5%), whereas in multi village communities fewer respondents attended the meeting (Kasempa 46.5%; Mulendema 9.9% and OCT 45%) (Table 3 3). At the 5% significance level, the percentage of respondents who attended the AGM differed amongst the different v indicating a statistically measurable difference in AGM attendance between the six villages. Demographic Characteristics and AGM A ttendance CBNRM programs are said to favor more affluent elite members of the co mmunity while sidelining disadvantaged groups ( Agarwal 2001 ) To test whether this was the case in our sample, demographic characteristics of who attended were analyzed to determine what groups of people in the community were attending AGMs ( Bryan 2004 p. 107 138 ) We therefore conducted a chi square test of association to determine if there was any significant difference in attendance at AGMs associated with four demographic variables i.e. gender; age; education and whether they were head of households or not. These characteristics have been used in other studies that examine democratic participation (Table 3 3). AGM attendance and a ge Age was categorized int o six different groups (18 28; 29 39; 40 50; 51 61; 62 72; and above 73). Results show the lowest attendance at the AGM was within the 73 years and above age group (44.4%) and the highest attendance was within the 62 72 year old age group (64.4 %), but at the 5% significance level, the percentage of

PAGE 89

89 0.23. AGM attendance and gender Out of the 244 male respondents, 121 (49%) attended the AGM, in comparison to 12 5 (51%) out of the 245 females. Similarly, at the 5% significance level, the percentage 0.75. AGM attendance and education Education was categorized into four groups ( no education, primary, secondary, and college). 62% of respondents with no education attended the AGM, compared to 41% with primary education, 49% with secondary education and 32% with college education. At the 5% level of significance, the percentage o f respondents attending the AGM to that expected from the literature a higher percentage of respondents with no education attended AGMs when compared those with so me education. AGM attendance and household h ead Of the 196 heads of household interviewed, 92 attended the AGM (47%) compared to 59% of interviewees who were not a head of household. Attendance at AGMs by heads of households was lower at the 5% significa 5.66, p = 0.02, a result that is unexpected and difficult to explain CBNRM Social and Economic Benefits and AGM A ttendance Next we assessed if there was an association between attendance at AGMs and the provision of CBNRM bene fits, namely cash, employment, and social benefits in

PAGE 90

90 varied forms including meat, community projects, orphanage funds, assistance to old people or destitute and so on. Cash b enefits and AGM attendance Of 144 respondents who got cash benefits, 116 (80.5%) attended the AGM. Of the 349 respondents who did not get cash benefits, 113 (32%) attended the AGM. This difference is highly significant at t Respondents receiving cash benefits were much more likely to attend AGMs. However, we know that cash benefits are also strongly associated with three single villages where attendance was also far higher (K hwai, Mababe and Sankuyo). Employment benefits and AGM attendance Similarly, of the 162 respondents who reported getting employment benefits, 122 (75%) attended the AGM, compared to only 32% (i.e. 97 out of 295) who did not get benefits. This is also sign Again, high employment and attendance is also associated with the three single villages (Khwai, Mababe and Sankuyo). Social benefits and AGM attendance Results for social benefits are similar but not as stark with 72% (i.e. 125 out of 173 respondents) of people who got social benefits attended the AGM, compared to 61% significant at the 0.1% level and associa ted wi 497) = 52.93, p = 0.000). Institutional C haracteristics and AGM attendance According to ( Ostrom 1990 ) the age and size of the community are important factors to consider when examining the success of go vernance on the commons, while

PAGE 91

91 institutional theorists such as ( Olson 1971 ) argue that the smaller the community the easier it is to achieve collective action. We therefore categorized the institutional characteristics of each community according to age of the CBNRM programme (> 10 years old versus and < 10 years old) and whether the co mmunity was either a single or multi village community. Age of community and AGM attendance In CBNRM programmes that were older than ten years, 68% of respondents (199 out of 292) attended the AGM compared to 27% (57 out of 210) in newer programmes, a di older communities are also the smaller communities, and we suspect size and history are more important variables than age. Size of community and AGM attendance In sin gle village communities, 84% of respondents (145 out of 172) attended AGMs compared to 33.6% (111 out of 330) in multi village communities. This difference Factors Predicting AGM Parti cipation Based on the theory of participation in democratic institutions (see theoretical framework) and the results of the chi square tests of association, logistical regression statistics was used to identify variables that predicted attendance at AGMs in the six communities. The logistic regression model analyzed the predictive values of three categories of variables Demographic characteristics; age, education and gender Benefits: cash, employment and social Institutional and Structural characteristics : size of community ( Table 3 5).

PAGE 92

92 Results The dependent variable, AGM participation a binary categorical variable was coded as 1 = Yes and 0 = No. The predictor variables included age (continuous), education (categorical variable with 4 levels, with t he reference category = 1 = no education), sex (binary 1 = Yes and 0 = No), cash, employment and social benefits categorized as (1 = Yes and No = ), community effect ( categorical variable with 1 = Mulendema and 0 = the rest of the five villages 7 ) and siz e (1 = single and 0 = multi village). Education, benefits and s ize as predictors of AGM participation The following Hypothesis was tested, The likelihood of participation in AGMs is greater for respondents with higher levels of education, who enjoy a gre ater level of both economic and social CBNRM benefits and are from single village communities The logistic regression model reflects the results of the simple Chi2 tests (Table 3 3). The likelihood ratio chi square of 103.08 with a p value of 0.000 tel ls us that our model as a whole fits significantly better than an empty model (i.e. model with no predictors). From the model (Table 3 5) eight of the variables were statistically significant. These included education ( primary secondary and college), whic h had a moderate predictive effect, CBNRM benefits (cash employment and social) and finally institutional characteristics (community effect and size) Of the demographic variables, age and sex did not predict AGM participation. The three CBNRM benefits ( c ash, 7 Mulendema was used as reference as it had the lowest levels of participation and proved to be the outlier in the sam ple of six communities. The community effect compares results of Mulendema respondents with other respondents from all five communities.

PAGE 93

93 employment and social ) showed strong statistical significance. Cash effect had a substantially large predictive effect when compared to the others. Finally both community effect and size indicated statistical significance in predicting participation i n AGMs. Interpretations of key findings The results show that if respondents had a primary, secondary or college education, they were less likely to participate in the AGM when compared to those who had no education. When one has a primary education, the o dds of attending the AGM decrease by 58%. The odds for those with a secondary education decrease by 60% and finally odds decrease by 68% for those with a college education holding all other variables constant. Regards to CBNRM benefits, the results show that the odds of participating in the AGM increase by 215% for those who received cash benefits. It increases by 65% for those who received employment benefits. Finally it shows that for those who received any social benefits, the odds increase by a facto r of 106%. With respect to the community effect, the model demonstrates that the odds for participating in the AGM decreased by a factor of 91% if respondents were from Mulendema community when compared to those who were from Khwai, Mababe, Sankuyo, Kasem pa and OCT communities. Finally, the model shows that the odds of participating in multi village community AGMs decreases by a factor of 82% when compared to those held in single village communities.

PAGE 94

94 Overall, the results show, that the predictor variable with the strongest effect on participation at AGMs is cash benefits. Attendance at AGMs therefore is more likely to be influenced if members received some cash benefit from the CBNRM program. Qualitative Data Community Records of AGMs Held To further unde rstand the descriptive statistics and the regression models, qualitative data was collected from community records of five communities (Khwai, Mababe, Sankuyo, Kasempa and Mulendema). Retrieving community records for OCT was not possible. For the Botswana communities, data was accessed from the Department of Wildlife and National Parks (DWNPS). Access to the community files in Zambia was obtained from the Zambia Wildlife Authority (ZAWA). This source of data is more likely to be objective and is particularl y important for cross checking survey data as it is the record of what occurred. Survey data has various limitations, among them perceptions of respondents maybe bias. Subjective phenomena such as opinions, values and knowledge when compared to objective m easures have the larger of what the survey mostly measure ( Grooves, Fowler Jnr et al. 2004 ) Each community is required to keep a record of the meeting agenda, minutes of the AGM and number of peo ple who attended. This is important for monitoring aspects of community governance and whether institutional rules are being followed. According to community records, the Khwai, Mababe and Sankuyo communities had held AGMs each year, according to their co nstitutional requirements. Records for Khwai community indicate that AGMs were held each year from 2005 2009. The records further indicated the number of people who had participated for some on the AGMs. In 2005 for example a total of 92 people had parti cipated and in 2006 the

PAGE 95

95 indicate the number of people that had attended, however both communities had held the meetings each year from 2005 2008. In the case of Kasempa, the r ecords indicate the last AGM was held in 2006 in all of the five VAGs in the area. Each VAG had a record of between 68 374 participant s attending The records for Mulendema show there were no AGMs held from 2002 2008. The only meetings that had been he ld on a quarterly basis were the Board meetings. There was a higher frequency of these meetings in some years when compared to others, for example 3 Board meetings were h eld in 2005 The statistical analysis on perceptions of AGM participation corroborates with the records kept for AGMs held. There were a larger percentage of respondents in the three small communities (Khwai, Mababe and Sankuyo) who indicated they had attended the AGMs. This is supported by the records kept by each of these communities. Fa ctors Contributing to Non Participation in AGMs The first part of the analysis involved understanding factors that predict participation in AGMs. This was analyzed using both descriptive statistics, logistic regression models and examining community record s of AGMs held. The overall conclusion in this first part of the analysis indicates that education, CBNRM benefits and both institutional and structural elements of the community predict AGM participation. This part of the analysis focused on respondents t hat participated, however it falls short in informing the study on reasons why other respondents did not participate. Apart from understanding factors that contribute to AGM participation, an equally important rstanding the factors that undermine

PAGE 96

96 these institutions provide insights on how democratic governance in rural communities managing natural resources can be improved upon. The focus of this section of analysis was to understand the reasons why some respon AGM. All individual responses for the five communities (Khwai, Mababe, Sankuyo, Kasempa a nd Mulendema) were analyzed. Data for the OCT community was not available. The data was organized by identifying consistencies and differences of each listing reasons why memb ers of the community were less likely to attend the AGMs. A number of themes emerged from the initial categorization process. Five themes were cited as the major reasons for non participation in AGMs. These included poor consultation of the community by Tr ust and CRB leaders, poor transfer of information to the communities by leaders, absenteeism due to work related commitments, lack of trust in leadership and AGMs not being held so community members could not participate. Poor consultation of communities b y Trust and CRB leaders This theme emerged from the majority of responses ( N = 22) from two of the single village co mmunities and from Kasempa ( Table 3 7 ). It was evident among the responses, that leaders from Mababe, Sankuyo and Kasempa were not adequat ely consulting the community prior and during the AGMs. As a consequence of this poor consultation by leaders, some community members did not see the need to attend AGMs. None of the respondents from Khwai identified this as a reason, so it is evident that community leaders in Khwai have some form of dialogue with their constituents.

PAGE 97

97 A few quotes from the open ended responses are presented below. Sankuyo: I always raise my concern but nothing is done leading me to giving up and seeing no use of attending t he meeting Kasempa : We are not informed by the members of the CRB, so that I can know that there are meetings Kasempa : interest; themselves only Though this theme emerged among the non participati ng respondents in the single villages, the majority did however attend the AGM (quantitative data), so this is evidently not a major problem in single village communities and in the Kasempa community. Poor transfer of information to the communities by lea ders Poor communication by Trust/ CRB leaders was also highlighted as another important theme among the Kasempa respondents, and a few from Mababe community. Mababe : They take long to call a meeting, and let them know about how the situation in the Trust Mababe : People were not told the date when the meeting would be held and that is Kasempa : I was not notified Kasempa : I am not informed Kasempa : The C RB does not inform people when and what time the community will have the meeting. We just heard that there was a meeting, and we are not informed Absent from AGM due to work related engagements Non participation due to work related engagements was cited a s one of the major reasons. This reason was referred to in 24 responses from the single villages, and 3 from Kasempa. Among the single villages, Sankuyo had the largest number of these responses. AGM meetings, in the single villages are usually held during peak tourism

PAGE 98

98 season, and since the majority of residents are employed in this industry, they are usually absent during the AGMs. This was also corroborated with participatory observation data that was gathered during an AGM held in Sankuyo in June 2009 AG M. The majority of participants were the older and younger members of the community who were not in any formal. Lack of trust in leadership Lack of trust in leaders, was provided as a reason by respondents from both Kasempa and Mulendema. This was not refe rred to among any of the responses from the single villages and indicates a general trust in the leadership in these communities when compared to the two multi villages. Some of the responses include: Kasempa : Because the members of the CRB are not trust anybody else to be free to do what they want They take long to call a meeting, and let them know about how the situation in the Trust Kasempa : AGM Mulendema : Becau se the leaders are selfish. They just give to their relatives Mulendema : Members are selfish The lack of trust in leadership shows governance systems that are less participatory and more ali gned towards the top leaders in the CRB. In Zambia, this has emerged as a major factor that hinders participatory forms of CRB governance. AGMs not held The majority of Mulendema residents indicated that no AGM had been held in Mulendema and this was prim

PAGE 99

99 time these meetings were held in their individual VAGs. What is evident is that CRB leaders have been holdin g Board meetings each year ( Table 3 6). A total of 7 board meetings had been held from 2002 2008, however no record of an AGM was found. Discussion Factors that I nfluence C ommu nity M A ttendance at AGMs Participatory models of democratic governance suggest that forums such as AGMs are an important mechanism for involving members in the governance of CBNRM programs ( Murphree 1997 Deitz, Dolsak et al. 2002 Child 2005 Anstey 2 009 ) Understanding how these micro institutions work, contributes to an improved understanding of the mechanics of decentralization and bottom up approaches to development ( Mohan 2000 Larson 2004 ) In CBNRM, AGMs are the institution created to encourage participatory governance and accountability. Despite their stated importance for ensuring equitable benefit distribution and participation at the micro level, very little research shows how they actually function. Our work contributes to this scholarship by examining the factors that affect levels of participation in AGM s. Our research suggests that participation in democratic institutions such as AGMs is determined by demographic characteristics such as education, the individual and collective CBNRM benefits, the size of the community and the specific characteristic of the community. Education Our results show that the level of education an individual has attained is a moderate predictor of participation, but that less educated people are more likely to attend meetings than more educated people. The effect of an indiv such as education, on participation in associations or meetings has been examined in

PAGE 100

100 earlier studies of participation ( Almond and Verba 1963 Agarwal 2001 ) Studies of ge neral assumption is that members with high levels of education are more likely to attend when compared to those with little education ( Bryan 2004 ) While, intuitively, participation in community AGMs, favors those with higher levels of education that are capable of accessing information, and questioning their leaders the empirical evidence i n our study indicates this is not the case. Agarwal, emphasizes that meaningful participation in natural resource management at the village level should include marginalized groups such as women and the less educated ( Agarwal 2001 ) Our results show that this may be happening in AGMs, but also support the concern that CBNRM institutions are undermined because more educated people tend to leave these communities for better opportunities. CBNRM b enefits Participation in collective action, including attendance at AGMs, depends on the incentives, including costs and benefits, that motivate individuals to attend or not ( Ostrom 1990 Murphree 1991 Adhikari 2005 ) It is the assumptions that communities are prepared to invest in the collective m anagement of local resources because the benefits derived through collective management outweighs the transaction costs of collective action ( Adhikari 2003 ) In addition, individual benefits that accrue at the household level also determine whether communities participate in governance or not ( Adhikari 2005 ) Our results indicate that the scale of individual and collective benefits community members receive has a powerful influence on whether they attend or not. A higher percentage of respondents getting cash benefits participated in meetings (80.6%) when compared to those who attended but got no cash (38.1%). Similar results

PAGE 101

101 were obtained for those who had employment benefits (75.35 versus 32.9%). There was however a much smaller difference between those that had participated in t he meeting and had received social benefits and those that had attended and received none (72.3% versus 61.4%). These differences were all statistically significant. This relationship is also indicative in the logistic regression analysis which suggests t hat the likelihood of a member of the community participating in the AGM increases by a factor of 215% if they got cash, by 101% if they got social benefits (such as meat, funding of orphans, old people or other in kind benefits) and finally by a factor of 81% if they were employed in any of the CBNRM ventures such as tourism, or administrative duties for the community organization. Theorists ( Child 2006 ) argue that the provision of cash as a direct benefit t o households captures part of the value of natural resources at the private level, and symbolizes local ownership and discretionary rights to wild resources, and is therefore a critical factor in promoting sustainable natural resource management in ways th at treating wild resources entirely as a social good cannot reach. The option of cash provides a direct incentive to households to participate in community governance. One of the main agenda items discussed in AGMs is the distribution of benefits and in pa rticular cash. As our results show, cash benefits are an important predictor of whether one attends AGMs or not, even though the level of cash benefits in the study area has been on the whole quite low compared to sites where it is made quite explicit that individuals have the primary right to use wildlife incomes for cash ( Child and Pet erson 1991 Dalal Clayton and Child 2003 ) If members have received cash benefits in previous AGMs, it is highly likely that they would attend the AGMs for this purpose.

PAGE 102

102 Thus, even if cash ben efits are small, they have a big impact in encouraging attendance. We can also hypothesize that where participants have genuine discretionary choice to allocate benefits between the options of cash as well as to projects, administration and natural resour ce management, the debate on resource allocation will be rigorous. Attendance at AGMs increases by a factor of 101% if the respondents received any form of social benefits. The allocation and prioritization of funding to social benefits such as pensions f or old people or orphans, or community projects, is usually done at the AGM. The qualitative data gathered from community records suggests that this is much more participatory in the three single villages, where the benefits are clear to see water stand pipes, toilets, houses provided to old and destitute people. Where community members have received social benefits at previous AGMs, and particularly where the magnitude of these and the debate over these is significant, it increases the likelihood that respondents will attend AGMs. Finally, the likelihood of attending the AGM if one received any form of employment from CBNRM activities increases by a factor of 81%. Both push and pull factors are affecting this variable. We expect that, like cash and so cial benefits, employment encourages people to attend AGMs. But we have also observed that community members who are employed in CBNRM activities often poorly attend AGMs. AGMs take place during peak tourism periods, so those employed in the industry are restricted from attending because they are at places of work, and in tourism this often implies being located in remote areas. This argument may also be the reason why there is slightly lower attendance of working age people (Table 3 3). Employment is

PAGE 103

103 st atistically significant in predicting attendance, but is a weaker predictor compared to cash and social benefits presumably because of these influences. Size of community The size and characteristic of the community are both important institutional and st ructural predictors of community governance. According to commons theorists the size and homogeneity of the community are among some of the factors that increase governance success on the commons ( Wade 1987 Baland and Platteau 1996 ) This is reflected in CBNRM governanc e where, ideally, the deliberation of decisions within smaller producer communities permits members to participate in a face to face manner ( Murphree 2000 Child and Barnes 2010 ) Out of our sample of six communities, the results s how that there was generally larger attendance in the three small sized communities (Khwai, Mababe and Sankuyo) when compared to the multi village communities (Kasempa, Mulendema and OCT). The odds of participating in the AGM decreased by a factor of 81.6% if respondents were from multivillages when compared to those from single village. Organizing meetings in multi village communities has huge transaction costs primarily due to the large numbers of households and distances members have to travel. Meetings in some communities last for two days with the larger part of the discussion being driven by the leaders in the community. For communities that have not held meetings on a regular annual basis, when they do meet, a number of issues have to be deliberated o n. These are usually emotional debates that involve members accusing their leaders of various atrocities. This usually derails the agenda and at the end of two days, very little constructive deliberation would have taken place.

PAGE 104

104 Community characteristic In multi village communities, particularly where leaders have things to hide, not holding AGMs is a way of avoiding being accountable to their constituents. The cases of Mulendema and OCT communities particularly reflect this. In these communities leaders ha ve deliberately not held AGMs in many years; however they are able to hold Board meetings and gain allowances from them. These leaders manipulate these democratic institutions that are supposed to be open and participatory for own personal gains. Both thes e communities are relatively large, and heterogeneous for ethnic and other reasons. CBNRM in these communities has been organized by bringing together villages that consist of different tribal groups, and the assumption is that these will work together. In Mulendema, retired farmers who have formed an elite group that dictate how the program is managed have dominated the leadership in the CBNRM programs. This has left out the earlier settlers of the area who are Kaonde Ila and have lived in the area since t he establishment of the National Park. In OCT, members of the Seronga community have dominated CBNRM governance, and hindered members from other villages from participating. In both communities, few benefits get to the constituent villages, and there is s uspicion about the management of funds. Community characteristics such as population of the community, size of the area, number of villages, revenues earned and type of traditional authority are other institutional factors that i nfluence AGM participation ( Table 3 2). All the single villages had low human populations (approximately 500 700 residents) and households ranging from 52 for Khwai, 55 Mababe and 56 for Sankuyo. These are mostly members of the same families so social networks in these

PAGE 105

105 communiti es are very strong and it is evident that distribution of CBNRM benefits is fairly equitable. The size of the area of the Controlled Hunting area where the single villages are located is not a huge expanse of land, and most of the villages are located wit hin 5 20 km stretches. This therefore makes it easier for residents to easily congregate at AGMs The traditional leaders/ Chiefs for all three single villages in the sample were all relatively progressive. The Khwai Chief can be characterized as more passive, and though he partakes in CBNRM activities, he has left the majority of decisions to the community. The Chiefs in Mababe and Sankuyo were more actively involved with CBNRM affairs, and in particular the Sa nkuyo Chief has always been a big player during AGM meetings. Chief Timex of the Sankuyo community has played a key role in ensuring community leaders are accountable to the residents and this has positively impacted the CBNRM governance structures and dec isions. The Mababe Chief is young and western educated and interviews with him indicated that he was pro indigenous rights and very supportive of the community programs. Another community characteristic that enhanced participation at the AGM was the quali ty of the Manager the community had employed. The employment of Trust Managers in Botswana was a need identified following huge mismanagement of community funds. Trust Managers are professional staff employed to manage operations and finances on behalf of the community and are required to report to the community Board. In Zambia, the position of Community Extension staff has been piloted in a number of CRB communities. Kasempa was one of the communities where

PAGE 106

106 this model is currently being practiced. An offi cer from the Zambia Wildlife Authority has been employed to provide extension services, similar to what the Trust Manager in Botswana does. These Managers/ Extension officers have provided support to communities particularly in organization of AGMs and oth er meetings. Our results show that all the communities (single villages and Kasempa) that had higher levels of participation at the AGMs, also had good professional Trust Managers /Extension staff facilitating these processes. Employing these staff members has enabled some level of professionalism in how community meetings are organized. Trust Managers/ Extension staff prepare annual reports that provide an objective view of status of affairs in the community, unlike the traditional way the Board presents t hese reports. They also provide a neutral platform as they are not members of the community, so are not necessarily involved in village politics that sometimes drives the AGMs. Finally, they are professionally trained staff and are key in helping the commu nity strategically work through their day to day operations, and ensure that more households are included in the process. The characteristics of the three single villages outlined above, therefore positively reinforce why there has been higher participati on at the AGMs when compared to the multi village communities. These factors should therefore be considered as important elements for improving CBNRM governance and project implementation. Factors that C ontribute to N on P articipation in AGMs Aside from un derstanding factors that influenced AGM participation, the study further examined factors that hindered participation. Understanding these factors is as equally important as it contributes to understanding the contextual factors that would not be normally captured using questionnaire data.

PAGE 107

107 Poor community consultation and transfer of information by CRB/ Trust leaders were among the reasons given for non participation in AGMs. This indicates that the community leaders were making decisions without informing the communities. These concerns were raised by respondents from both single and multi villages Studies of other CBNRM communities have shown that this is a common trend in other programs in the southern Africa ( Collomb, Mupeta et al. 2010 ) Child et al., unpublished/ Dashboard). Leaders rarely consult or transfer the necessary information that is needed by their constituents. This can be attributed to poor capacity of leaders in understanding their roles and responsibilities. Future initiatives that support improved CBNRM governance would therefore need to target how these bottlenecks that hinder consultation and information transfer and ultimately deter members from attending could be addressed. The lack of trus t in leadership was another major reason identified as reasons for non partic ipation in AGMs. This was referred to by majority of respondents from both Kasempa and Mulendema community. A general perception among communities in the CBNRM programs in Zambia is that leaders cannot be trusted. This is particularly evident when it comes to financial matters. CRB board members, are the primarily agents that handle financial matters in the communities. This includes signing on tenders given to safari hunting companies, accessing revenues earned from hunting (CRB chairperson is a signatory to community account and a ZAWA officer is a second signatory), distributing and paying salaries of wildlife village scouts and making decisions on how funds for community projects are spent. Over the past decade, these roles have become the prestige of CR B leaders and communities therefore are rarely

PAGE 108

108 privy to information on how their finances are managed or utilized. Based on these prevailing circumstances, communities have grown to distrust their leaders. As a result of this, the need to participate in in stitutions such as AGMs therefore proves to be futile. This is a major contributing factor in a number of the communities in Zambia and in order to reverse this trend, strict measures on how funds should be managed and distributed in CBNRM programs needs t o be addressed. Common property theorists and CBNRM practitioners broadly agree that participatory models of democratic governance are important in ensuring decisions are brought close to the main constituents of the CBNRM programs ( Ostrom 1990 Murphree 1997 Murphree 2000 Anstey 2009 Child and Barnes 2010 ) The past two decades of CBNRM implementation in Southern Africa, has shown that top down models of governance have translated poorly at the gr assroots. Households that are meant to be the main target beneficiaries o f these programs have received very few benefits particularly in Zambia and the governance systems have been eroded to the point of where only 10 Board members manage and make decisio ns with little no consultation of their constituents. This study examined the factors that either enhanced or hindered participation in AGMs in six communities in Botswana and Zambia. The broader goal was to gain some understanding on how performance of m icro democratic institutions such as AGMs could be improved upon. In order to see more successful CBNRM programs, participatory models of governance need to explored and replicated in other programs. The study shows that the small size of the community and higher levels of cash, social and employment benefits enhance participation in community governance. Participation

PAGE 109

109 is further enhanced by stronger facilitation of democratic process as we have seen in Kasempa, and which we also observe in Khwai and Sanku yo (good mangers) compared to Mababe (where managers are absent or changing). It is weakened where communities are large, organized using multi village governance structures, and where organizational and financial management is weak. However, we also nee d to point out that where an organization appears to be running quite smoothly, as in the case of Sankuyo, levels of participation may fall slightly, whereas we suspect that high levels of participation in Khwai and Mababe might also be linked to internal conflicts and contestation within these communities.

PAGE 110

110 Table 3 1. Typologies of participation Forms/ Level of Participation Characteristic features 1. Nominal Participation 2. Passive Participation 3. Consultative Participation 4. Activity speci fic Participation 5. Active Participation 6. Interactive (empowering) participation 1. Membership in the group 2. Being informed of decisions ex post facto, or attending meetings and listening in on decision making, without speaking. 3. Being asked an opinion in specific matters without guarantee of influencing decisions. 4. Being asked to (or volunteering to) undertake specific tasks 5. Expressing opinions, whether or not solicited, or taking initiatives of other sorts. 6. Having voice and influence in the decisions. Source: ( Agarwal 2001 p.1624 )

PAGE 111

111 Table 3 2. CBO characteristics KHWAI MABABE SANKUYO KASEMPA MULENDEMA OCT Year Established 2000 2000 1995 2003 2003 1995 Number of Households +/ 52 55 56 1479 2500 300 Number of People +/ 600 500 700 9, 000 12, 000 3, 043 Size of Area (Km2) 1918 2181 860 13, 340 3091 92,200 Number of Villages 1 1 1 5 5 5 CBNRM Revenue Source Photographic & Hunting Photographic & Hunting Photographic & Hunting Hunting Hunting Photographic & Hunting

PAGE 112

112 Table 3 3. AGM attendance variables measured VARIABLE VARIABLE TYPE TYPE AGM attendance Yes No Dependent Passive part icipation (Being informed of decisions ex post facto, or attending meetings and listening in on decision making, without speaking. Agarwal, 2001) Demographic characteristics: Age Education Gender Household Head Gender Household head # of years in village Independent Measures of individual and household characteristics that influence participation. ( Almond & Verba, 1963; Dahl, 1961 ) CBNRM Benefits: Cash Jobs Social Independent Measure of incentives to participate in AGM ( Moe, 1980; Olson, 1965) Institu tional characteristics: Size of CBO Age of CBO Independent Measure of structural design of CBO (Ostrom, 1990; Mu r phree 2000)

PAGE 113

113 Table 3 4. Crosstab AGM attendance Cross tabulations on AGM Attendance Did you attend AGM % OF AGM Chi 2 YES (N) NO (N) TOTAL (N) ATTENDEES (p<) VARIABLE COMMUNITY Khwai 50 2 52 96.2 158.52 Mababe 43 9 52 82.7 0.000 Sankuyo 52 16 68 76.5 Kasempa 46 53 99 46.5 Mulendema 11 100 111 9.9 OCT 54 66 120 45.0 TOTAL (N) 502 AGE 18 28 75 60 135 55.6 6.71 0.243 29 39 63 65 128 49.2 40 50 39 46 85 45.9 51 61 30 36 66 45.5 62 72 29 16 45 64.4 73+ 12 15 27 44.4 TOTAL (N) 486 GENDER Male 121 123 244 49.6 0.100 Female 125 120 245 51.0 0.752 TOTAL (N) 489 HEAD OF HOUSEHOLD Yes 92 104 196 46.9 5.66 No 106 73 179 59.2 0.017 TOTAL (N) 375 EDUCATION No Education 82 50 132 62.1 15.935 Primary 51 72 123 41.5 0.007 Secondary 98 100 198 49.5 College 8 17 25 32.0 TOTAL (N) 478

PAGE 114

114 Table 3 4. Continued CBNRM Cash Benefits Yes 116 28 144 80.6 73.47 No 133 216 349 38.1 0.000 TOTAL (N) 493 CBNRM Employment Benefi ts Yes 122 40 162 75.3 75.43 No 97 198 295 32.9 0.000 TOTAL (N) 457 CBNRM Social Benefits Yes 125 48 173 72.3 52.93 No 199 125 324 61.4 0.000 TOTAL (N) 497 AGE OF CBNRM PROGRAM < 10 years ol d 57 153 210 27.1 82.19 > 10 years old 199 93 292 68.2 0.000 TOTAL (N) 502 SIZE OF CBNRM COMMUNITY Single 145 27 172 84.3 116.15 Multi 111 219 330 33.6 0.000 TOTAL (N) 502

PAGE 115

115 T able 3 5. Description of variables in model Variable Descript ion Dependent: Participation_AGM Did you attend the last AGM? Independent: age How old are you? primary Primary education secondary Secondary education college College education Gender Gender of respondent Cbnrm_Cash CBNRM cash benefits Cbnrm_E mployment CBNRM employment benefits Cbnrm_Social CBNRM social benefits kasemp Kasempa community khwai Khwai community mababe Mababe community sankuyo Sankuyo community OCT OCT community cbo_scale Size of community Table 3 6. Logistic regression m odel on factors that predict AGM attendance Participation_AGM b P>z e^b** % Demographic age 0.009 0.219 0.99 0.9 primary 0.879 0.003 0.42 58.5 secondary 0.920 0.002 0.40 60.1 college 1.136 0.034 0.32 6 7.9 sexf 0.061 0.767 1.06 6.2 Benefits cash_benefit 1.147 0.000 3.15 214.8 employment 0.498 0.052 1.65 64.6 social_benefit 0.724 0.002 2.06 106.3 Institutional characteristics cbo_scale 1.694 0.000 0.18 81.6 mumbw 2.435 0.000 0.09 9 1.2 ** e^b = exp(b) = factor change in odds for unit increase in X percentage chan ge in odds for unit increase in X No. of observations = 492, ***p<0.01, **p<0.05 & *p<0.1

PAGE 116

116 Table 3 7. Record of AGMs held CBO Record of AGMs, General Meetings, Spe cial General Meetings and Board Meetings Khwai AGM: 1 st of April, 2005 (92 people attended. AGM: 27 th 28 th April, 2006 (74 people attended). Special General Meeting: 12 th November, 2006 AGM: 2007 AGM: 2008 Special General Meeting: 8 th April 2009 Maba be AGM: 10 th 11 th March 2005 AGM: 24 th 25 th March 2006 Special General Meeting: 25 th January 2007 AGM: 15 th 16 th March 2007 AGM: 28 th 29 th February 2008 Sankuyo AGM: 14 th 15 th February 2005 AGM: 23 rd August 2006 AGM: 26 th 27 th February 2007 Kasempa Shimuka VAG meeting: 25 th July 2006 (92 men and 127 women attended) Kamfwe VAG meeting: 24 th July 2006 (374 people attended). Kamakechi VAG meeting: 28 th July 2006 (287 people attended). Lunga VAG meeting: 29 th July 2006 (68 members attended) Mule ndema Board Meeting: 25 th November 2002 Interim Board meeting: 13 th January 2003 Extraordinary Board meeting: 6 th August 2003 Board Meeting: 30 th March 2005 Board Meeting: 13 th July 2005 Board Meeting: 21 st December 2005 Board Meeting: 4 th January 2008 OC T Records not accessed.

PAGE 117

117 Table 3 8. Reasons for not attending AGM CATEGORY No. SOME RESPONSES Not consulted by Leaders 1 6 and up and seeing no use of attending the meetin of the topic to forbid me to ask and again they carry on with other Absent: Work 2 4 Absent: Sick 1 4 Absent: Age 3 the work hard and make everything perfect as I trust them, they grew in for the Absent: Personal reasons 1 8 Lack of consultation by leadership 6 Lack of trust in leadership 4 anybody else to be free to do what th Poor communicat ion of meeting 5 will have the meeting. We just heard that there was a meeting, and we Absent: Not interested 2 Absent: Sick 2 Absent: Age 4 Absent: Work 3 Absent: No benefits 3 you can get from there unless y

PAGE 118

118 CHAPTER 4 THE IMPORTANCE OF DEMOCRATIC ELECTIONS IN PROMOTING CBNRM GOOD GOVERNANCE Role of Elections in Improving CBNRM Micro Governance The AGM as outlined in the earlier chapter is the democratic institution in CBNRM that supp orts direct deliberative democracy. This general assembly is the platform where ALL major decisions on CBNRM are discussed. One of the activities that take place during this meeting is the election of board members. Elections in CBNRM are important indicat ( Child and Dalal Clayton 2004 ) Enshrining regular elections in community constitutions allows for change of leadership, within constitutional rules. Despite their importance as instruments for participatory democracy in micro CBNRM institutions, elections are an understudied area. The importance of elections as democratic institutions has been studied by political scientists in macro elections ( Kaminski 2002 Hyden 2005 Lindberg 2006 ) These studies include aspects of what motivates voter turnout, whether incentives drive higher participation or not or, which members of society are more likely to vote, etc. These studies provide analytical tools that can be used to understand natural res ource governance and examine which factors drive participation in CBNRM elections. The importance of elections as democratic institutions has been broadly studied at the macro level ( Kaminski 2002 Hyden 2005 Lindberg 2006 ) included aspects of what motivates voter turnout, whether incentives drive higher participation or not or, which members of society are more likely to vote, etc. These studies provide empirical

PAGE 119

119 evidence that can be used to understand what factors drive par ticipation in CBNRM elections. Over the past two decades, CBNRM programs in Southern Africa have promoted elections as the instrument for change of leadership and leadership accountability, but there is lack of empirical studies that assess what actually characterizes participation in elections in communities practicing CBNRM. For example, do elections in smaller communities elicit higher levels of participation when compared to those in larger ones? Secondly, do communities that are receiving direct benef its such as cash, employment or social benefits have higher participation in elections when compared to those receiving fewer benefits? Finally, which members of the community are more likely to participate in elections? By studying elections as an instit ution that enables good governance in CBNRM, we begin to understand how leadership in local community institutions responsible for governing and managing natural resource can be more accountable to their constituents and consequently allow for more equitab le distribution of benefits attained from CBNRM. This section examines the importance of elections in CBNRM governance. Using empirical data from household surveys conducted in six CBNRM communities, the study examines factors that influence household mem The study argues that in CBNRM, participation of the majority of community members in elections is an essential part of improving democratic governance. The study further argues that in CBNRM, elections are the instruments that enable change of leadership and allow constituents the privilege of choosing which members of the community represent their interests.

PAGE 120

120 Board Elections in Botswana and Zambia Elections of board members in CBNRM programs are postulated in community c onstitutions, and may vary based on context. They do however have salient features that are practiced in most CBNRM communities in Southern Africa. In Botswana, each CBO elects Boards of Trustees who are given the responsibility to make decisions on behal f of the community and provide guidance for the overall management of CBNRM projects ( Schuster 2007 ) A study conducted on CBOs in Botswana showed that a majority of these Boards consist of 10 members who are elected for terms of office that range from one to three years depending on the community. The constitutions of almost all the CBOs allowed for re election of Board members, and that at least 64 board members had been re elected in active c ommunities in two years (2004 2006). This study concluded that even though re election of Board members was an important tool for retaining institutional memory in the Board of Trustees, there was a danger of power accumulation in only a few people of th e community, a rise that was pronounced in communities where the constitution did not restrict the number of terms per board members, which was the case for 50% of the active CBOs ( Schuster 2007 p. 44 ) In Zambia, the Wildlife Act of 1998 No. 12 upholds the importance of elections in CRBs governance. The Act stipulates that each CRB hold democratic elections to select their Board members. These elections are primarily held at the VAG level, with each VAG electing a representative to a ten member board. Employees of the Zambia Wildlife Authority such as senior wildlife or extension officers oversee the electoral process. According to the manual on CRB formation, this process is divided in two categories. The first st age begins by holding of elections at the VAG level. A team composed of ZAWA officials; teachers or a representative of the local district council

PAGE 121

121 facilitates this process. Members at the village level select leaders to represent them at the main CRB board elections. Secret ballot is the method of voting used and this works more transparently in the communities where the marginalized may not be free enough to vote by show of hands. At the second level the 10 board members elect from amongst themselves the c hairperson and vice, the secretary and vice, the treasurer and five board members. Theoretically, both countries emphasize the importance of the institution of elections in CBNRM governance. In Botswana, the importance of CBNRM elections is promoted so t hat powerful members of the community do not hold onto power. In Zambia, in practice, CRBs elections have almost become non existent due to poor institutional and financial support for holding elections. CRB board members have used the excuse of poor flow of financial revenues from ZAWA for not holding elections. This has thus created powerful elites who have held onto power and has created non participatory governance systems that do not allow for members to actively participate in the process. Theory on Participation in Elections Dahl in his work on democracy, argues that the meaning of democracy as self government requires not only equality of political participation but also free political competition ( Dahl 1971 ) Dahl, measures democracy using two key attributes i.e. participation and contestation. This, he argues, requires eight institutional guarantees i.e. freedom of o rganization; freedom of expression; the right to vote; broad eligibility for public office; the right to compete for support and votes; the availability of alternative sources of information; free and fair elections and the dependence of public policies on

PAGE 122

122 ( Dahl 1961 Dahl 1971 ) To understand the theory of participation in CBNRM community elections, this study uses three of these institutional guarantees free and fair elections. These three attributes should be guaranteed in order to ensur e that the two key attributes of Dahlian democracy, i.e. participation and contestation, are achieved. Dahl provides us with some useful starting points for understanding participation in elections. The institutional guarantees of democracy can thus be em pirically measured in any form of elections to understand whether they are met or not. Do citizens have the right to vote? Do leaders have the right to compete for support and votes? Finally, if elections are held, are they held in a free and fair manner? A measure of these three institutions would provide empirical evidence that allows examining the degree of democratic practice in different communities in CBNRM. Various political science scholars have broadly studied the role elections play in democratic practice. Most studies concern broader national elections and democratization, but the concept is the same, and can be applied to CBNRM communities. Within the two attributes of participation and competition that Dahl argued for, several scholars have mad e strides in studying the importance of elections in achieving these attributes. Proponents of political participation such as Dahl, argue that in a democracy, the people should rule over themselves, that is, the people are sovereign ( Dahl 1989 p. 85 ) same vain, scholars s uch as ( Ake 1993 ) have noted that it is the equal opportunity to be part of the decision making process rather than the approval of each substantial

PAGE 123

123 decision by everyone, which sat isfies the right to self government. Other scholars such as Bratton and van de Walle argue that elections alone are not sufficient to make a democracy, however no other institution heralds participatory, competitive and legitimate elections in instrumental importance for self government ( Bratton and Van de Walle 1997 ) Regardless of the many views of what democracy is or should be though, there is one common denominator among modern democracies, and that is elections ( Lindberg 2006 p. 1 ) Lindberg has argued that elec tions are an institutionalized attempt to actualize the essence of democracy: rule of the people, by the people ( Lindberg 2006 p. 1 ) The institution of elections is important for selecting leadership and disposing of old governments in a political system ( Lindberg 2006 p. 1 ) Lindberg furthers argues that elections are concrete occurrences with institutional expressions, rules and regulations. They can be observed directly and are concrete in the form of indelible ink, ballot papers, boxes, election officials and vote rs. They are thus a necessary component for any aspiring representative democratic system. Further, they guarantee by law equality of participation and free competition. In essence, therefore, both voters and political elites should guarantee the holding o f legitimate periodic elections ( Lindberg 2006 p. 1 ) Voter Turnout: Participation in E lections Participation in the voting process is a useful starting point in understanding the role elections plays in enhancing democracy. In a representative system, popular participation is achieved through voting. This provides legal guarantee of universal suffrage, which could be limited by citizenship, age or mental sanity ( Lindberg 2006 p. 1 ) The percentage of an electorate that actually participates is used an indicator of t he comprehension of equal popular participation. Further, participatory elections imply that

PAGE 124

124 individuals have equal rights to participate and to field candidates. Voter turnout is therefore considered a classical indicator of popular participation that is considered an important dimension of the quality of the democracy of a system ( Altman and Perez Linan 2002 ) A higher turnout of voters is therefore a preferred vantage point of democratic quality as the realization of equal political participation ( Lindberg 2006 p. 37 ) voters actually turning up for the poll ( Lindberg 2006 p. 37 ) Quality of Elections Free and F airness Understanding what elections are and what role they play in promoting improving democratic governance is a first step. Elections usher in new leadership th rough the voice of the people, but it is paramount to also understand the quality of the elections themselves. The fundamental question to be asked is whether the elections were free and fair. This particular variable is crucial in ensuring an important di stinction is made between whether processes are acceptable or not ( Lindberg 2006 p. 37 ) Factors Influencing Participation in E lections Measuring the quality of elections is not limited to voter turnout or whether or not they are free and fair. The role of elections in enhancing a democratic system can also be understood by examining the factors that influence voters to participate. Blair examines how in large national elections voter turnout is correlated with the expected closeness of the election ( Blair 2000 ) He further argues that a sense of duty plays an many people voting is not only a right, it is also a duty. And the bel ief that in a democracy every citizen should feel obliged to vote induces many people to vote in

PAGE 125

125 almost all elections. Everyone does not share that sense of duty. It may vary from one ( Blaise 2000 p. 113 ) Riker and Ordeshook examine how the cost and benefits of participation can affect turn out. They argue that small increases in costs to vote can have a significant impact on turnout levels ( Riker and Ordeshook 1968 ) Other scholars s argue that high l evels of democratic participation are associated with more equal distributions of income ( Mueller and Stratmann 2003 ) This builds from a larger debate on effects of participation in election s and whether this improves the quality of the democratic process. Some scholars have opposed the broadening of the franchise of participation on more below the highest ( Lijphart 1997 p. 1 ) argues that low participation in elections is detrimental in that it leads to inequality o f representation and that this influence is not randomly distributed but systematically biased in favor of more privileged citizens such as the those with higher incomes, greater wealth, and better education and against less advantaged citizens. Both schol ars recognize that suffrage and citizen participation have similar consequences in their effects on the composition of the active electorate, however they draw opposite inferences about the desirability of greater citizen participation ( Mueller and Stratmann 2003 ) Lijphart, takes on an elitist higher participation from an electorate representing various members of society ( Lijphart 1997 ) Mueller & Stratmann expand on this argument by examining the correlation between participation in elections and the economic status of the individuals vo ting in

PAGE 126

126 national elections ( Mueller and Stratmann 2003 ) They illustrate how, as the fraction of voters with low incomes increases, government policies shift in favor of lower income groups. However, they also show that when the fraction of uninformed or uneducated voters increases, government policies worsen. They show a strong association between education and voting, and argue that better educated people gather more information about govern ment policies and candidates during work and leisure time, so the cost of becoming informed and voting are lower for better educated citizens ( Filer, Kenny et al. 1993 ) The scholarship on elections is broad and has largely been studied in macro level political systems. Many writings, like those presented, identify the importance of elections in the democratic process 1 To bring us back to Dahlian democracy, they show how elections are a fundamental component in achieving both participation and CBNRM promotes elections at the micro level. CBNRM policies explicitly argue for selection of leaders through a democratic process i.e. elections. Understanding CBNRM governance therefore requires understanding three key attributes encapsulated within the concept of elections i.e. voter turnout, factors that influence participation in elections, an d the quality of elections. This study pursues three objectives: Assessing voter turnout by examining the numbers of community members who voted in previous CBNRM elections, their frequency of voting and their understanding of voting methods according to their community constitutions. 1 Interestingly, Dahl does not mention elections in his five principles of democracy. This implies that elections are a n outcome of processes that incorporate effective participation, voting equality, enlightened understanding, control of the agenda and inclusion of all adults. Elections may, on the one hand, measure the extent to which these process are in place, but we also know that elections, or apparent elections, on their own are no guarantee of democracy and can even be used as a tool to legitimize elite capture.

PAGE 127

127 Examining the factors both at individual household and community level that influence participation in elections. Examining the quality of elections, by assessing how free and fair elections held have been. Methods The st udy uses quantitative data from 514 household surveys conducted in six CBNRM communities in Botswana and Zambia (See full method description in methods chapter). Data analysis. To examine participation in elections (voter turnout) and factors influencing this, two levels of analysis were undertaken. We used descriptive statistics (using SPSS 16) to describe who voted, and whether this was statistically different from anal yze the factors that influence participation in elections. SPSS was used to finally Results Voter turnout To assess the variable of voter turnout, respondents were asked whether they h ad voted in the last CBO election or not. According to the survey results, the community with the highest voter turnout was Khwai (89.1%), followed by Kasempa (76.8%), Sankuyo (59.4%) and the lowest voter turnout was in Mulendema (42.5%), differences that were statistically significant at the 1% level (Table 4 1). To further examine voter turnout, respondents were asked the number of times they had voted in CBNRM elections. Frequency of voting is used as an indicator of whether communities were holding regu lar elections or not. The premise for this assessment is that communities that have held regular elections should have a large

PAGE 128

128 percentage of respondents with higher frequency of voting when compared to those that have not held regular elections. Responses were categorical and included options of every three years. Within the sample, Sankuyo a nd OCT were the oldest communities, established in 1995. These two communities should therefore have held a total of five elections. In comparison, the other four communities were all established after 2000 so on average should have held at least three ele ctions. The results show that the community with the highest frequency of voting was Khwai where 90.9% of respondents had voted three times or more; followed by Mababe 75.5; Sankuyo 69.6%; Kasempa 38.3%; OCT 23.6 and finally Mulendema 6.4%. The results in dicate that voting frequency for the three multi villages was lower when compared to the single villages. This also indicates that elections are held more frequently in the single village communities. Who votes in CBNRM elections? Aside from understanding the numbers who turn out for voting in CBNRM elections, we examined what groups of people vote using demographic characteristics of respondents including education levels, gender, and age of the respondents, and conducted chi square tests of association t o determine if differences were statistically significant. These demographic variables have been used in previous studies on elections ( Filer, Kenny et al. 1993 ) Educati on and v oter turnout The results indicate that out of 132 respondents with no education, 56.3% voted; compared to 60.5% with primary education, and 51.1% and 40% with secondary and

PAGE 129

129 college education respectively. These differences were not statistically s (1, N = 469) = 9.37, p = 0.497). Gender and v oter turnout Out of the 248 males, 54% of voted, and out of the 244 females, 56.6% voted, Age and v oter turnout Within the four age categories (18 28; 29 39; 40 50; 51 61; 62 72 and 73 years and above), the 40 50 year old category had the highest voter turn (72.9%) and the group with the lowest voter turnout was people over 73 years in age (36.7%) suggesting that there is a relationship between voting in CBNRM elections and the age of the respondents. Do CBNRM Benefits Influence Participation in E lections? Sim ilar to AGM attendance analysis, (chapter 3) the relationship between provision of CBNRM benefits and voting was examined. Cash benefits and v oter turnout Of the 279 respondents who got cash benefits, 99 (35.5%) voted in the CBNRM elections. Of the 205 who = 0.001), confirming a strong association between respondents receiving cash and voting in CBNRM elec tions. Employment benefit and v oter turnout Similarly, of the 252 respondents who had some form of employment within the CBNRM programs, 101(40.1%) voted in the elections, compared to only 30% (i.e.58 out

PAGE 130

130 ams. This difference was nearly Social benefits an d v oter turnout Of the 278 respondents who received some form of social benefits 42.4% had voted, compared to 25% (53 out of 209 ) who had not received social benefits but still = 487) = 15.4, p = 0.000). Preference of voting method In addition to assessing the frequency of voting, respondent s were asked to rank the method of voting they preferred (Table 4 3). The majority of respondents from Khwai (92.3%), Mababe (68%) and Sankuyo (77.1%) preferred voting by show of hands. The majority of respondents from Kasempa (85.7%) and Mulendema (85.1%) preferred voting by secret ballot. Responses from OCT community were not included in this part of the analysis as this data was not collected in the OCT case study. These results show that the majority of respondents had knowledge on the voting methods as articulated in their community constitutions. Very few respondents ranked appointment by Headman and Wildlife authority as a method of voting. This is another indication of the acceptance of the election method for the communities practicing CBNRM. Fa ctors Predicting Voter Turn out Studies on macro level elections show that various factors predict voter turnout at elections. For this study, two models were developed to gain an understanding on what factors predict participation in CBNRM elections. Thes e factors are examined at both the community and the individual level using logistic regression analysis.

PAGE 131

131 Model One Does the provision of CBNRM incentive s predict participation in elections? This model was developed with the objective of understanding w hether CBNRM incentives given at the individual household are predictors of participation in CBNRM elections. Similar to the analysis of AGM attendance, CBNRM benefits of cash, employment and social were included in the model as predictors. According to ( Mueller and Stratmann 2003 ) high levels of participation in elections are associated with more equal distribution of income. According to ( Mueller and Stratmann 2003 ) high levels of participation in elections are associated with more equal distribution of income. Based on this theoretical argument therefore, respondents who receive any form of CBNRM household benefits should there fore participate in CBRNM elections. Secondly there ought to be higher levels of participation in CBNRM elections among community members with higher socioeconomic status (i.e. higher education) when compared to those with little or no education. The two H ypotheses for the first model therefore included: Hypothesis 1 : Community members who receive either social or economic benefits from CBNRM are more likely to participate in elections. Hypothesis 2 : Community members with higher education (higher soci oeconomic status) are more likely to participate in elections when compared to members with little or no education (lower socioeconomic status). 4). Independent variables included the demographic variables of age, gender, education levels and three variables for CBNRM benefits (cash, employment and social).

PAGE 132

132 Summary of results Results in the model (Table 4 5) show five of the variables are statistically significant. These include age (40 50 & 51 61 year olds), education ( secondary) and CBNRM cash and social benefits. The variables of gender, several age groups (29 39, 62 73 years and above) and CBN RM employment benefits are not statistically significant. The effect of being in the 40 50 year old age group was substantively higher when compared to the effect of other independent variables. Interpretation of results The results of the model show t hat respondents from age groups between 40 50 & 51 61 years old are more likely to vote in CBNRM elections when compared to respondents who are between the ages of 18 28. Respondents between 40 51 years old are 126% more likely to vote in CBNRM el ections when compared to those between the ages of 18 28 years old. Further respondents between the ages of 51 61 years old are 111% more likely to vote in CBNRM elections when compared to those between the ages of 18 28 years old. Regards to educa tion, the results show that respondents with secondary education are less likely to vote in CBNRM elections when compared to those with no education. The results show that the odds of respondents with secondary education voting decrease by 43% when compare d to those with no education. Finally the results indicate that the odds of respondents who receive cash and social benefits voting, increases by 99% and 96% respectively. These results show that based on our hypothesis, CBNRM incentives are important ind icators of predicting voter turnout in CBNRM elections. Specifically they disclose that cash and social benefits are

PAGE 133

133 significant incentives that predict voter turnout. In contrast, employment benefits do not have an effect on predicting voter turnout. Mo del Two Does the size of the community predict participation in CBNRM elections? Riker and Ordeshook, argue that the cost of participating in elections is an important predictor of voter turnout. Small increases in the cost of voting can have significant impact on voter turnout levels ( Riker and Ordeshook 1968 ) In our sample, the single village communities are smaller than the component villages in the multi village communities. Not only is it therefore l ogistically easier to meet, but we also know that single villages had much higher levels of benefit so elections were more meaningful. In our sample therefore, the cost of participating in elections in single villages should be lower when compared to that of participating in multi village communities. Based on this premise we proposed the following hypothesis: Hypothesis 3 : Single village CBNRM communities are more likely to have higher levels of participation in elections when compared to multi village CBNRM communities. The second model tests whether size and community characteristics of the villages add explanatory power to the first model. This model predicts participation in elections from size of the village (a dichotomy variable), and community ch aracteristics (a categorical variable with 5 levels, with the reference variable = Mulendema). The likelihood ratio chi square of 94.71 with a p value of 0.000 tells us that our model as a whole fits significantly better than an empty model. Summary of re sults Results show that election participation of members between the ages of 40 50, and 51 61 are statistically significant when compared to those in the 18 28 age

PAGE 134

134 groups Further the model shows that participation in elections of respondents from K asempa, Khwai and Sankuyo community is statistically significant when compared to participation in elections of respondents from Mulendema community. There is no statistical difference between election participation of respondents from Mababe and OCT when compared to those from Mulendema. Further there is no Interpretation of results Regards to age and voter turnout, the model illustrates, that the odds of voting increases by 1 51.5% and 280% respectively for respondents who are between the ages of 40 50 and 51 61. This shows that participation in CBNRM elections is higher for members of the community who are in the 40 61 year old age range. The results show that younger me mbers of the community are not participating as much as the middle aged members. With respect to the CBNRM community and voter turnout, the results show that the odds of participating in elections increases by 151% for respondents from Sankuyo; 131% for Khwai and 56% for Kasempa residents when compared to those voting in Mulendema community (reference community). Among the CBNRM communities included in the sample, Mulendema had the lowest voter turnout. Quality of Elections In addition to understanding v oter turnout and factors influencing it, the quality of the elections is another important attribute of democratic elections. In this study, quality of elections was operationalized as how free and fair the last CBNRM elections were. This particular variab le is key in ensuring an important distinction is made between whether processes are acceptable or not ( Lindberg 2006 p. 35 ) Respondents had the

PAGE 135

135 elections (Table 4 8). The results show (Table 4 8) that a larger percentage of respondents from Khwai (61.8%), Sankuyo (61.4%) and Kasempa (59%) community chose the option of elections being very free and fair i.e. they were able to choose their candidates with freedom. In comparison, fewer respondents from Mababe (17%) and Mulendema (12.5%) respond ed that the elections were free and fair. candidate well), 27.3% of Khwai respondents, 15.1% from Sankuyo, 15% from Mababe, 10.9% from OCT, 4.3% from Kasempa and 0.8% from Mul endema chose this option. In further supports the responses that only 42.5% of respondents had voted in the elections. This data suggests complex underlying factors. Khwai h as a bimodal distribution of satisfaction with election, probably reflecting the two factions vying for power. At the time of the surveys, Mababe was struggling with leadership issues and the AGM had broken up because of conflicts, though people were sat isfied with general meetings. Sankuyo is coasting along nicely, and we observed that Kasempa was well supported OCT and Mulendema, which probably accounts for lower levels of satisfaction in election, and some 40% or more of respondents not ranking elections at all.

PAGE 136

136 D iscussion According to Dahl three institutional attributes enhance participation and contestation in ( Dahl 1971 ) Our study examined these attributes by assessing participation in CBNRM elections, factors that influenced this participation, and finally the quality of CBNRM elections. Participation in Elections Our result s show that participation in elections was good (above 60%), in Khwai and Sankuyo (singe villages) and Kasempa (multi) community. On average, these three communities should have had at least three elections since their establishment, and supporting seconda ry documentation suggests that these communities have held regular elections according to their constitutions and members in the community show a strong indication that they are participating in these elections. Leadership positions in Khwai are particular ly highly contested, with community records and interviews confirming four changes of leadership since its establishment. Khwai community had one powerful family in CBNRM leadership from its inception (2002 2006). In 2007 at a special general meeting, co mmunity members used a vote of no confidence to vote out the family that had ties to the Ward councilor and one of the more influential Head men in the community. A new Board was elected in 2007, and at the time of this research in 2009, community members were generally confident in this Board 2 This backdrop of 2 However, in 2011, this board was entirely replaced, and interviews with the Manager revealed concerns that the new board was comprised of young people who may be radical and inexperienced in CBNRM management. Indeed, while the managers of Khwai, Sankuyo and Mababe generally supported the

PAGE 137

137 contestation among community members in Khwai provides further understanding of why there was higher participation in elections in Khwai when compared to the other communities in the sample. This co mmunity, through its own internal village dynamics, has managed to create a system of governance that allows members the right to vote and chose their leaders. They have further learned to use democratic institutions such as elections to hold their leaders out of office. However, there are also signs that in the absence of oversight and support from NGOs and other CBNRM support organizations there is a risk of ochlocracy, and the rule of passion over reaso n ( Madison 1787 p. 342 ) 3 For Kasempa, interviews show that this community has been having regular elections at the village level. Since its esta blishment in 2003, this community has held a total of three elections. According to interviews, these elections have been highly provision of buckets of maize, cooking oil an d seeds (pers comm). Interviews in all the five village area groups in Kasempa indicate that elections held at the village level are highly contested and, in turn, large numbers of community members vote in these elections. Unfortunately due to lack of rec ords of voter numbers, it was difficult to verify this. The survey results are therefore our main source of information on levels of rights of communities to elect their leaders, they also noted how h ard it was to work in such small communities where outcomes were often associated with bickering and decisions affected by personality differences. Specifically, managers noted that conflicts of interest were almost unavoidable in small communities. They employed community members as a social benefit, but it was extremely hard to discipline these community employees who might have significant power in (or to influence) the board, 3 For instance, when com munities were first empowered to partner with the private sector to manage hunting and tourism, they often replaced partners willy nilly for ad hoc or personal reasons, creating significant turbulence in the sector without any clear or logical benefits.

PAGE 138

138 participation in CBNRM elections; in future; it would be useful to assess this data using more objective data such as a lis t of voters. Results for Sankuyo indicate that approximately 59.4% of respondents voted in CBNRM elections. Within the sample, Sankuyo was one of the oldest communities, established in 1995. Since its inception, this community should have held a total of over five elections. The results however indicate that voter turnout was a bit mediocre. As part of this research, Sankuyo elections were observed in June 2009. Participation at these elections was relatively low and consisted of mostly elderly women and m en (above 50 years old), a few young men and women, the Kgosi (Chief), and National Parks staff (who facilitated the elections) and some local district council representatives. Elections were included as part of a broader meeting agenda that included discu ssion of a marketing strategy, new policies and procedures and a new land use plan. Elections for new Trust leaders were the last agenda item. National parks staff facilitated the elections and members voted by show of hands. One of the notable features of this election was that a large group of the members of the working age (30 50) was not present. Through interviews it was discovered that a larger number of households were not represented as the heads were working in tourism camps as this was peak tour ism period. Both results from the survey and observation of an actual election in Sankuyo demonstrate that though this community is having elections, participation is relatively poor, with mainly the elderly and young unemployed members of the community pa with CBNRM communities in the region by the University of Florida research group suggest that people in Sankuyo are generally satisfied with their programme, and more

PAGE 139

139 sat isfied than any other community that was surveyed. Sankuyo is generally well managed, money is accounted for, and people are benefiting widely, suggesting that in such circumstances AGMs and elections become routine, and that moderate levels of satisfied voters may indicate an optimal position. Voter participation in elections in Mababe (46%), Mulendema (42%) and OCT (43.9%) was relatively low. This was particularly interesting for the Mababe community which was established in 2000. This community should h ave held an average of more than three elections since its inception. According to community records a total of five Boards had been in office since 2000. The only period with a stable Board according to constitutional rules was between February 2004 and 2 006. At the time of the research in June 2009, a new Board had just been voted in. Interviews indicated that this had been done through a vote of no confidence in the 2006 2009 Board. Participation in Mababe elections demonstrates that there is some leve l of contestation for leadership positions. This community has held more elections than they are required to constitutionally. This in turn should reflect the levels of community members voting, but our survey results indicate that in the last five years o nly 46.5% of the respondents had voted. Other records suggest that this community is not well organized, but is nonetheless benefiting from a lot of jobs and meat from hunting. Finally, the results for Mulendema and OCT reflect the poor governance syst ems in both communities. Both Mulendema and OCT are multi village communities where governance has been hijacked by a few elite members who have deliberately manipulated democratic institutions such as elections. Both Mulendema and OCT have had the same Bo

PAGE 140

140 Factor Predicting P articipation in Elections In addition to assessing participation in CBNRM elections, our study examined the factors that influence participation using two logistic re gression models. Results of the chi square tests of association and the logistic regression show a strong statistical significance in how CBNRM benefits (cash and social) influence voting. Mueller and Stratmann argue that participation in democratic instit utions such as macro elections are associated with more equal distribution of incomes or revenues ( Mueller and Stratmann 2003 ) In the sample of 492 respondents, the odds of voting in CBNRM e lections by respondents are much higher for respondents who had received cash, and social benefits. This implies that community participation in democratic institutions increases if they are economically meaningful. Improving democratic governance in CBNRM communities therefore is linked to the overall level of benefit, the influence of participatory process on the distribution of these benefits, and the accrual or perception of benefits at the household level. Communities that are distributing benefits mor e effectively are more likely to have constituents engage in selecting their leaders and ultimately improving accountability in these systems. Finally, another factor that has influenced participation in macro elections is the cost of participation (Riker and Ordeshook, 1968). We assessed this by comparing participation in voting between single and multi villages on the premise that the cost of participating in elections in single villages is lower. Our results indicate there is no statistical difference b etween voting in single and multi villages However there is a statistical difference in individual community effects. Both Khwai and Sankuyo show a higher percentage of voting when compared to the multi village communities in the sample.

PAGE 141

141 Participation in elections tends to be higher in small villages where the benefits of participation are higher and the costs of participation are lower. However, sound organization can increase the level of participation above what our models predict (e.g. Kasempa), and conversely poor organization can reduce the level below what we hope (e.g. Mababe). Very high levels of participation may be linked to conflict or contestation, and perhaps even ochlocracy (e.g. Khwai) and smooth running democracies may be associated wit h good but not exceptional levels of participation in elections (e.g. Sankuyo). However, democratic institutions such as elections are more prone to being manipulated by leaders, and this risk of elite capture appears to be higher in multi village communit ies, occurring in two of our three multi villages but not being sustained in any of the three single village communities. Such risks can be addressed by encouraging more decentralized forms of governance right at the household and village level. In conclu sion, participation in elections increases where democratic institutions become more meaningful to the community and are trusted, at least to some extent. This occurs where they are associated with the management and distribution of higher levels of benef its, as we see with cash, social benefits and, to some extent, employment. It also occurs where participation itself is meaningful through either the small size of the community, situations where sound democratic processes enhance participation, or both.

PAGE 142

142 T able 4 1. Crosstabs of v oter turnout Did you vote in CBO elections % VOTED YES (N) NO (N) TOTAL (N) ATTENDEES Chi2 (p<) VARIABLE COMMUNITY Khwai 49 6 55 89.1 75.33 Mababe 23 27 50 46.0 0.00 Sankuyo 41 28 69 59.4 Kasempa 73 22 95 76.8 Mulendema 48 65 113 42.5 OCT 54 69 123 43.9 EDUCATION No Education 76 56 132 56.3 9.37 Primary 75 44 119 60.5 0.497 Secondary 101 92 193 51.1 College 10 15 25 40.0 GENDER Male 134 105 248 54.03 1.99 Female 138 102 244 56.6 0.37 AGE 18 28 70 59 129 52.6 22.16 29 39 66 60 126 51.6 0.014 40 50 62 23 85 72.9 51 61 38 26 64 57.6 62 72 27 16 43 58.7 73+ 11 17 30 36.7

PAGE 143

143 Table 4 2. Voter turnout & benefits Cross tabulations on Voting in CBO elections Did you vote in CBO elections? % VOTED YES (N) NO (N) TOTAL (N) ATTENDEES Chi2 (p<) CBNRM Cash Benefits Yes 99 180 279 35.4 14.13 No 42 163 205 0.001 TOTAL (N) 141 343 CBNRM Employ ment Benefits Yes 101 151 252 40.1 5.13 No 58 136 194 0.077 CBNRM Social Benefits Yes 118 160 278 42.4 15.4 No 53 156 209 0.000 Table 4 3. Frequency of e lections How many times have you voted in CBO Elections? Khwai Mababe Sankuyo K asempa Mulendema OCT Once (N) 1 7 7 29 25 40 Twice (N) 4 5 8 29 44 13 Three times or more (N) 50 37 48 36 7 29 Never voted (N) 0 0 6 0 32 41 Total 55 49 69 94 109 123 % Voted more than four times 90.9 75.5 69.6 38.3 6.4 23.6

PAGE 144

144 Tab le 4 4. Prefere nce of voting method Please rank the method of voting you prefer, starting with your highest preference first Khwai (%) Mababe (%) Sankuyo (%) Kasempa (%) Mulendema (%) Appointed by Headman 0.0 0.0 1.4 3.1 0.0 Appointed by Wildlife department 0.0 4.0 7.1 2.0 0.0 Vote by hand 92.3 68.0 77.1 9.2 0.0 Vote by secret ballot 7.7 28.0 14.3 85.7 85.1 Don't know 0.0 0.0 0.0 0.0 14.2 N 52 50 70 98 94

PAGE 145

145 Table 4 5. Variables in model Variable Description Dependent : didyouvote~w Did you vote in the last CBO elections? (Yes or No) Independent: Age Age1 Age 2 Age 3 Age 4 Age 5 Age 6 How old are you? 18 28 29 39 40 50 51 61 62 72 73+ primary Primary education secondary Secondary education college College education sexf Gender of respondent Cbn rm_cash CBNRM cash benefits Cbnrm_employment CBNRM employment benefits Cbnrm_social Kasemp Khwai Mababe Sankuyo OCT Cbo_scale CBNRM social benefits Kasempa Khwai mababe Sankuyo OCT Size of village Table 4 6. Binary equation: factor change in odds of v o ting didyouvote b P>z e^b** % Age: 29 39 0.295 0.317 1.221 34 Age: 40 50 0.817 0.015 0.202 126 Age: 51 61 0.749 0.043 0.289 111 Age: 62 72 0.032 0.937 3.442 3 Age: 73+ 0.369 0.377 0.997 31 Education: Primary 0.048 0.87 3.037 5 Education: Secondary 0.555 0.055 0.589 43 Education: College 0.550 0.285 0.598 42 Gender (Male = 1) 0.013 0.945 4.343 1 Benefit_Cash 0.688 0.014 0.374 99 Benefit_Employment 0.239 0.357 1.431 21 Benefit_Social 0.672 0.005 0.397 96 ** e^b = exp(b) = factor chan ge in odds for unit increase in X percentage change in odds for unit increase in X No. of observations = 492, ***p<0.01, **p<0.05 & *p<0.1

PAGE 146

146 Table 4 7 Binary Equation: f actor Change in Odds of Voting didyouvote b P>z e^b** % Age: 29 39 0.333 0.271 1.099 41 Age: 40 50 0.864 0.011*** 0.146 151 Age: 51 61 1.224 0.003*** 0.202 280 Age: 62 72 0.113 0.798 2.18 17 Age: 73+ 0.096 0.975 2.343 1 Education: Primary 0.336 0.810 1.091 8 Education: S econdary 0.491 0.494 0.711 20 Education: College 0.133 0.891 2.017 7 Gender (Male = 1) 0.078 0.855 2.551 4 Benefit_Cash 0.825 0.019 0.192 717 Benefit_Employment 0.016 0.960 4.135 523 Benefit_Social 0.342 0.211 1.073 6 cbo_scale 0.075 0.9 42 2.59 80 kasemp 1.822 0.000*** 0.5999 56 khwai 1.465 0.035*** 0.382 131 mababe 0.231 0.271 1.465 41 sankuyo 0.506 0.011*** 0.681 151 OCT 0.184 0.575 1.693 280 ** e^b = exp(b) = factor change in odds for unit increase in X percentage change in odds for unit increase in X No. of observations = 492, ***p<0.01, **p<0.05 & *p<0.1

PAGE 147

147 Table 4 8. Quality of elections Do you think the elections were free and fair? Khwai (%) Mababe (%) Sankuyo (%) Kasempa (%) Mulendema (%) OCT (%) Very fa ir (was able to choose my candidate with freedom) 61.8 17.0 61.4 59.0 12.5 38.8 Somewhat fair (free to choose my candidate ok) 9.1 34.0 11.4 5.0 25.8 10.1 Not fair (not able to choose my candidate well) 27.3 15.1 4.3 15.0 0.8 10.9 Bad (only a few peop le participated) 7.5 1.4 7.0 3.1 Not interested 1.8 17.0 20.0 6.0 8.3 17.1 Don't know 9.4 1.4 8.0 52.5 20.2 N 55 53 70 100 120 129

PAGE 148

148 CHAPTER 5 DEMOCRATIC DECENTRALIZATION THE ENABLING ENVIRONMENT FOR CBNRM Performance of CBNRM Prog rams in S outhern Africa The performance of CBNRM programs in Southern Africa has received criticism in its poor delivery of social, economic and conservation measures ( Bond 2001 Emerton 2001 ) The basis of this criticism ranges from its lack of downward accountability to its intended for constituents ( Shackleton, Campbell et al. 200 2 Dalal Clayton 2003 Jones and Murphree 2004 Murphree 2004 ) its limitation to achi eve market based solutions for improved efficiency in the management of natural resources ( Dressler, Buscher et al. 2010 ) ; and its failure to deliver to both to communities and the environment ( Blaike 2006 p. 1947 ) These criticisms broadly address the challenges of implementing CBNRM and in recent years, scholarship has grown in understanding these challenges through adaptive learning processes ( Armitage 2005 ) These macro challenges have been examined primarily through the processes of natural resource decentralization ( Agrawal and Ribot 1999 Agrawal 2001 ) These studies focus on factors that motivate states t o devolve power to the periphery and the outcomes of these institutional changes and motivations ( Agrawal and Ribot 1999 Blair 2000 Cheema 2007 Grindle 2007 ) These studies in their aim to understand the transfer of resource rights from the state to local c ommunities fall short in understanding how micro level community institutions to which power is decentralized work. Ostrom advocates for a broader understanding of the interaction between macro and micro institutions involved in this institutional transfer of resource rights. She argues that understanding how social ecological systems function begins by understanding the multi level nested framework that examines interactions and outcomes of four

PAGE 149

149 subsystems (resource systems, resource units, governance sys tems and users) ( Ostrom 2009 p. 419 422 ) Understanding the interaction and outcomes of governance systems in CBNRM therefore requires a clear examination of the relationship between macro, meso and micro level proc esses and institutions ( Child and Barnes 2010 ) In the case of CBNRM, this requires not only understanding what mot ivates states to decentralize resource rights to local communities, but further to understand how these communities perform within these institutional changes. In Southern Africa, assessment of CBNRM programs, rarely takes into consideration, how communiti es organize or govern themselves and the enabling environment that creates these conditions. First and second generation CBNRM programs mainly focused on macro level factors such as the process of decentralization and ensuring the communities were provided with economic incentives in order to achieve conservation outcomes. This approach was particularly evident during the donor funded initial projects in southern Africa, and there are few studies during this period that actually examined how these micro com munity institutions of governance perform within broader macro conditions. In this chapter, we broadly examine the institutional changes that have taken place in CBNRM in Botswana and Zambia. We ask the question: Has CBNRM created the institutional change The first section of this chapter explores the question of decentralization as it relates to Community Based Natural Resource Management (CBNRM) performance and the macro political factors th at have shaped the performance of CBNRM in the two countries of study i.e. Botswana and Zambia. This is undertaken by examining factors that influenced policy directives that decentralized power from the state to local

PAGE 150

150 communities in wildlife areas. Final ly the paper examines how these policy directives have affected the performance of six CBNRM communities in Botswana and Zambia on measures of democratic governance and distribution of CBNRM benefits. Theoretical Framework Decentralization in CBNRM: Expl oring the Concepts and Principles In CBNRM decentralization, the devolution of authority over and responsibility for wildlife resources outside the national parks to land owners and occupiers was an imperative which had to be achieved in its formative stag es. Coupled with long term economic incentives this was aimed at improving habitat and species status ( Hulme and Murphree 2001 ) The extent to which devolution in CBNRM has delivered its intended goals is understudied. Is devolution simply rhetorical in CBNRM, or does it contribute to CBNRM governance and how does it shape itself in proprietorship communities? This section is a literature review that explores the scholarship undertaken, on decentralization in the broader sense; on devolution; democratic governance and democratic decent ralization in natural resource management. The primary objective is to gain a clearer understanding of these concepts; to develop hypotheses in the context of CBNRM governance; and in turn identify indicators for testing these hypotheses. This is undertake n by asking two questions: What is the purpose of decentralization in CBNRM? What type of devolution has been undertaken in CBNRM in the countries of study, and what are the key indicator variables that can be used to measure its performance? Decentrali zation the big p icture Decentralization is a broad concept. As a means it can improve efficiency and effectiveness of public services. As an end itself, it can promote the basic principles of

PAGE 151

151 democratic governance ( Cheema 2007 ) Most authors agree that decentralization within the state involves the transfer of authority to perform some services from an individual or an agency in c entral government to some other individual or agency which is closer to the public to be served. The basis for transfer is often territorial however it can also be made functionally: that is transferring authority to an agency that is functionally speciali zed. Such transfers are of three main types: First is when the delegation is within formal political structures (for example when the central government delegates additional authority to local government). Second is the transfer within the public adminis trative or parastatal structures. Third is when the transfer is from an institution of the state to a non state state agency ( Turner and Hulme 1997 ) This criterion is similar to that earlier proposed by Rondinelli. Rondinelli distinguishes the concept of decentralization into three types, namely: ich central government personnel are assigned to localities away from the governmental centre; Delegation: through which control is granted to state enterprises, lower level governmental actors, or non governmental actors, but the higher level authorities retain the discretion to reassert their control; Devolution: through which higher level authorities grant or recognize the formal permanent control by lower level governmental or nongovernmental actors ( Rondinelli 1981 p. 133 145 ) It is important to note here, that in the case of decentralization in (CB) NRM, the third criterion proposed by Turn er and Hulme in 1997, and Rondenelli in 1981 is what has taken place. The state institution i.e. the Department of Wildlife and National Parks enabled the transfer of wildlife management to local communities residing on wildlife estates. This process has t aken place in several Southern African countries, beginning with Zimbabwe under the CAMPFIRE program, and more recently in Namibia. These three categories of decentralization provide useful analytical tools for assessing (CB) NRM outcomes. Using the above criterion, one could then take a country by country

PAGE 152

152 case study and examine the forms of decentralization that have been undertaken and the outcomes that result as a consequence of this process. Decentralization where implemented, however does not always op erate in the same manner. Grindle in her work on decentralization and local governments in Latin America confirms that decentralization is a process that unfolds over time; more important it is neither a linear process nor one that necessarily results in s imilar outcomes ( Grindle 2007 ) In her work on the effects of decentralization on thirty Mexican municipalities, she examines how local governments respond when they are assigned new responsibilities and resources. In the 1980s, the government of Mexico transferred control over property taxes, and the responsibility of services such as sewage and solid waste management to municipal governments. Grindle shows how with this transfer of financial power, there was increasing competition among opposition candidates to contest for positions in the municipal offices, particularly for local el ites. Grindle argues that decentralization increased opportunities for new groups to gain environment for good governance; instead it resulted in higher competition among inter est groups that stalled development. Devolution a form of d ecentralization Devolution is the ideal form of decentralization that is advocated for in (CB) NRM in southern Africa. Devolution of powers and resources to local governments and actors has been t he foundation for promoting sustainable decentralization in developing countries. Advocates argue that local governments and actors with decision making power, authority and resources play a more catalytic role in economic and social development. The premi se is that citizens are more likely to participate actively in local

PAGE 153

153 political processes where local governments are perceived to have the capacity to make political and financial decisions affecting their economic and social welfare. Transferring power an d authority from the central government to sub national administrative and local government units and opening the political process to widespread participation provide an institutional framework for local autonomy and empower communities to pursue local as pirations ( Grindle 2007 ) Devolution is aimed at creating or strengthening indepe ndent units of government or non state actors by devolving functions and authority. Its fundamental characteristics are the autonomous nature of local units of government, legally recognized geographical boundaries within which they exercise their authorit y, and the power of local governments to mobilize resources to perform their tasks ( Cheema 2007 ) The classical model of devolved l ocal government was advocated in the 1960s as the blueprint that newly independent countries should pursue ( Hicks and Hicks 1961 Maddick 1963 ) It had five main features: e from central government and responsible for a range of significant local services. It should have its own treasury, budget and accounts along with subnational authority to raise its own revenue. It should employ its own competent staff who it can hire, f ire and promote. A majority elected council, operating on party lines, should decide policy and determine internal procedures. Central government administrators should serve purely as external advisors and inspectors and have no role within the local autho rity ( Mawhood 1983 p. 12 ) It can be said t hat these principles are still being used to define devolution at local government level. These five features maybe several decades old, however, they are still relevant to the devolution process. What would be strikingly interesting is to assess if these measures would also apply to devolution of natural resources in CBNRM.

PAGE 154

154 Two forms of devolution in local governments have been implemented in developing countries. The first is political devolution which provides a legal basis for the exercise of power at t he local level and enables citizens to influence local policy making and priority setting. The second is fiscal devolution which assigns functions and revenues to subnational and local governments the resources by which to implement local policies and prog rams ( Grindle 2007 ) .There are however many challenges that hinder effective devol ution of financial and political authority to local governments. Both supporters and opponents of devolution agree that without adequate mechanisms of accountability and for combating corruption at the local level, devolution could lead to inefficiencies, ( Cheema 2007 ) Decentralization and devolution in n atural resources m anagement So how then are these principles elaborated in natural resource management? Scholars of decentralization and natural resources argue that decentralization through the lens of natural resources provide sharp optic insights into d ecentralization writ large not just into natural resources ( Ribot and Larson 2004 ) They elaborate that this optic is particularly powerful since natural resources differ from other sectors in ways tha t democratization and development. Natural resources are at once critical for local livelihoods (subsistence and income generation) and are also the basis of signific ant wealth for governments and national elites. As such they have historically been a point of struggle between rural people and these elites. With decentralization, natural resource transfers is a great opportunity for increasing the relevance of local au thorities

PAGE 155

155 to local people, yet is simultaneously a threat to central authorities and elites who fear loss of income or patronage resources. Other scholars focusing on governance of natural resources on the commons elaborate on the importance of institution s for governing the commons. They outline that in the absence of effective governance institutions at the appropriate scale, natural resources and the environment are in peril from increasing human population, consumption, and development of advanced techn ology for resource use, all of which have reached unprecedented levels ( Deitz, Ostrom et al. 2003 ) In natural resource management, democratic decentralization is premised on new local institutions that are re presentative of and accountable to local populations and have secure autonomous domains of powers to make and implement meaningful decisions ( Ribot 2002 ) This should be the most effective and efficient way of encouraging the local authorities to serve the needs and desires of their constituents. With regard t o natural resources however democratic decentralization has proven difficult to find and the results of existing policies are highly varied ( Ribot 2004 ) Measuring the impact of decentralization policies is particularly hard, however ( Agrawal and Ribot 1999 ) provide a framework called the actor, powers and accountability framework for analyzing the type and extent of decentralization that has taken place in a specific country. In particular the framework considers: T he powers and accompanying resources actually transferred to lower level actors to determine an autonomous domain of decision making actually exists around issues of local significance. The local level entities receiving powers and their relation to the po pulation in order to understand the extent to which these are both representative of and downwardly accountable to local peoples.

PAGE 156

156 Other authors such as ( Ascher 2007 ) emphasize the continued rol e of government following devolution of natural resource management to local communities. Ascher argues that devolution by no means implies that Government should withdraw completely, despite the temptation for the Government to reduce accountability and c osts. The abandonment of Government responsibilities can easily undermine the reestablishment of sound community should be maintained to uphold appropriate boundaries and the user rights of the community, making it clear that the Government is recognizing user rights rather than granting such rights; enforces rules against spillover damage, when communities cannot resolve these problems; increase the availability of market rate credit; and support research and de velopment, provide technical assistance and training ( Ascher 2007 ) Devolution an enabling condition of CBNRM g overnance The success of CBNRM is highly dependent on several factors, key among wh ich is the importance of a macro level enabling environment. In CBNRM, this includes policy and legislative changes that support devolution of the governance and management of natural resources to the local level, the building of local level capacities to take on these responsibilities, and the tracking of performance (to support adaptive management) and conformance with CBNRM principles (to protect the conditions in which CBNRM is theorized to be most successful). Ideally, local control over decisions and sources of revenue is intended to increase the autonomy of local communities from central government and other powerful actors, to empower communities economically, and to engender local resourcefulness. The decentralization of natural resource management in southern Africa is particularly evident in the wildlife sector. Ideally these

PAGE 157

157 conditions are achieved by devolving rights, benefits, authority and responsibility from the State to local non state actors ( Murphree 2000 ) This has rarely be en achieved in Southern Africa. Conceptually, southern African scholar practitioners have defined the ideal end point as a situation where the rights to use, manage and benefit from natural thus replicating the devolution of rights to use wildlife to private landholders ( Martin 1986 ) The second challenge is to develop systems of micro governance whereby these rights are managed to provide public benefits in the form of improved nat ural resource management, increased income, equitable benefit sharing and political participation. The only place where devolution has fulfilled the first set of conditions is on private land in Zimbabwe through the bold Parks and Wild Life Act of 1975, a s even in South Africa and Namibia the state places considerably more restrictions on private wildlife management than on the management of livestock. The intention to replicate the success of private land conservation in southern Africa on communal lands out to implement the theoretically ideal conditions outlined above in key government documents ( Martin 1986 ) While the wildlife agency fully intended to replicate full devolution as had occurred on private land on communal lands, its power was limited by the need to cooperate with the powerful ministry of local government. The Parks and Wild Life Act was amended in 1982 to allow full devolution to district councils, with an understanding that th ese rights would be withdrawn if the district councils failed to explained in lucid detail by ( Murphree 2005 Murphre e 2005 ) Working through district

PAGE 158

158 councils was a two edged sword. On the one hand, it provided long term capacity support still evident in some districts today where community revenue distribution and quota setting still takes place more than ten years after donor and NGO support was withdrawn. On the other, the wildlife agency made a crucial mistake right at the beginning of CAMPFIRE when it issued a letter stating that at least 50% of revenues should be devolved to communities, and subsequent attempt s to increase this amount to 80% were strongly resisted. These lessons were shared with Namibia and Botswana in particular. Thus, when both of these programmes were developed they bypassed district government structures. Like Zimbabwe, Namibia encased i ts CBNRM intentions then retained 100% of the income from wildlife tourism and hunting. Initially, there were strong concerns that the ministry was undermining CBNRM, for example, by retaining programme was neither initiated nor protected through national legislation. Rather, communities were given 15 year Head Leases through the Land Boards which allowed them to retain most income from wildlife. The wildlife department sets quotas and has not devolved any responsibility for wildlife management such as patrolling, and it also charges the standard (albeit low) permit fees for hunting that take s place in communal lands. In the case of e lephants this is currently P20, 000 compared to a trophy fee for elephants that varies around P90, 000. Nelson and Agrawal claim that the only CBNRM program that comes remotely close to achieving devolution is N amibia where a legislative transfer of rights to local communities has taken place ( Nelson and Agrawal 2008 ) but this is not strictly true as CAMPFIRE was based on legislative rights in the

PAGE 159

159 Parks and Wild Life Act of 1975 (as amended in 1982). Indeed, it was players like Rowan Martin in Zimbabwe that convinced the Namibians of the necessity of legislation who advocated the power of the legis lative pen. Scholar practitioners in CBNRM in southern Africa strongly advocate full devolution of proprietorship of wildlife to local communities. Anticipating ( Schlager and O strom 1992 ) this is regularly conceptualized as the rights to (1) retain all benefit from wildlife (2) to allocate and sell wildlife to best advantage, and (3) to have the right to manage it, and even to replace it should this be in the interests of the community. Thus enabling conditions should, ideally, legislate for these rights. In addition, there was strong recognition that the capacity of communities to manage and take advantage of tory technology ( Goredema, Bond et al. 2006 Bond, Goredema et al. 2007 Stuart Hill, Diggle et al. 2007 ) This included natural resource management and institutional design and development, though with most support being pro vided by conservation agencies the latter tended to be neglected, with the main exception being CAMPFIRE. Finally, CBNRM was designed as a policy experiment, and the strong emphasis on adaptive management required macro level monitoring of performance (na tural resource management, benefits, etc.) and conformance to the principles of CBNRM including revenue sharing. With its emphasis on democratization and institutional development, this was most developed in the CAMPFIRE programme as we can see from vario us summary reports ( Child 1996 Bond 2001 CAM PFIRE Association 2007 Taylor 2009 )

PAGE 160

160 reporting systems ( NACSO 2008 ) This suggests that we can define the enab ling environment using the following variables: The devolution of rights and these can be devolved legislatively, administratively or de facto and should include: Right to benefit (measured by per cent of revenue getting to community Right to allocate and sell resources (can the community own hunting and tourism joint ventures, or is this done by the government on their behalf?) Right to manage (can the community set quotas, exclude others from its resources through game guards and patrolling, and so on ?) Capacity building including for: wildlife management (e.g. monitoring, quota setting, managing safari hunting) Selling wildlife (e.g. marketing joint ventures themselves) Institutional development (e.g. constitutions, administrative and democratic p rocesses, managing projects, etc.) Performance and conformance monitoring including: Monitoring wildlife populations, income, community satisfaction, etc. Monitoring conformance including that budgets are set democratically by community, that finances and reported back to and controlled by the community on a regular basis, and that elections are held regularly. Assessing the macro factors that enable this process of devolution is important in the broader scheme of measuring CBNRM performance ( Child and Barnes 2010 ) This can be assessed at the country level, through the lenses described above. In the following sectio n we describe the evolution of CBNRM programs in Botswana and Namibia with an emphasis on the macro factors that have shaped the devolution of CBNRM authority and responsibility, or the lack of it, to local communities.

PAGE 161

161 Institutional Changes that have S hap ed CBNRN in Botswana Following the early successes of CAMPFIRE, CBNRM was rapidly adopted in southern Africa. In 1989, the first CBNRM program was implemented under the Natural Resources Management Program (NRMP) in Botswana 1 and this was facilitated in p art by the USAID funded Natural Resource Management Project and played a particularly important role. This was largely managed by what USAID calls an Institutional Contractor, albeit in conjunction with the Department of Wildlife and National Parks (DWNPS ). Through this project, the first concept of community trusts was implemented, and in line with the Wildlife Policy of 1986 communities could legally acquire benefits from the wildlife resource and manage them. This project focused on developing joint v entures between communities and the private sector as had been pioneered in Zimbabwe and Namibia ( Child and Weaver 2006 ) and resulted in significant increases in revenues, especially from hunting. However, the importance of institutional development and processes was apparently not recognized, and to this day the absence of clear guidelines and capacity building related to community parti cipation and benefit sharing are a serious weakness in the Botswana CBNRM programme. With US$25 million funding, this project found an environment conducive for CBNRM in Botswana. It took a total of 10 years for CBNRM to evolve in the approach that is now functioning in Botswana. The NRMP was instrumental in shaping the policy directives 1 In Zimbabwe, USAID submitted a nave proposal to support CAMPFIRE in 1989. This was rejected by the wildlife agency, and then rewritten overnight by the agency as the basis of $30m support for the next 13 years. By 1986, it was noted that USAID supported districts tended to perform less well than those not getting aid, and USAID insisted that the mid term review get re written when the observation was made that top reliance more than it was he lping the programme. Namibia was similarly assertive when it came to their dealings with USAID and CBNRM, taking funding only strictly on their own terms.

PAGE 162

162 that defined the CBNRM landscape. Policy development during the NRMP included development of the Wildlife Conservation and National Parks Act in 1992, Joint Venture Guidel ines 1996/ 1999, resource user leases, CBNRM implementation guidelines and the draft CBNRM policy ( Rozemeijer 2003 ) Despite the enactment of various policies in tourism and natural resource management, the main transform ative policy that can be said to have contributed to the evolution of CBNRM in Botswana was the Wildlife Policy of 1986. The early CBNRM programme in Botswana was based on spirit of this1986 policy which reflected similar trends in the region regarding usi ng wildlife to generate revenues and devolving wildlife management to communities. This can be regarded as the pioneer policy that created the first directive that allowed local communities to benefit and manage wildlife in Botswana. The Department of Wil dlife and National Parks (DWNPS) through this policy established 20 percent of the country as Wildlife Management Areas where local people could obtain a wildlife quota and consume or sell if for profit ( Cassidy 2000 ) It was designed to encourage local conse rvation practices by providing people benefits from wildlife and other resources. At this time, the spirit of devolution and community management was strong, and was quickly reflected in other policies that also emphasized the involvement of local communi ties in wildlife and natural resource use and management. These included the National Conservation Strategy of 1990; the Tourism Policy of 1990; the Tourism Act of 1992 and the Wildlife and Conservation National Parks Act of 1992. These policies and acts indirectly created the institutional change that decentralized some form of power on natural resource management to the local communities. Despite their inclusion of the

PAGE 163

163 ain any specific provisions on how greater community involvement was to be achieved. In 1995, the Government of Botswana started to make more progressive steps to achieve greater community involvement in wildlife management. As noted, the ort to decentralize wildlife management to local communities was based on administrative procedures rather than legislation and was facilitated by the Ministry of Commerce and Industry which at the time oversaw the Department of Wildlife and National Parks Towards the end of 1995, the Government began a consultative process with the aim of developing a CBNRM policy. This was intended to begin to formalize the decentralization process, shifting from reliance on administrative decentralization towards entren ching the rights of communities in policy and legislation. From the onset of this institutional change, fiscal decentralization was a consistent directive. It was accepted and promoted by official and civil participants in the process, and indeed fiscal de volution has been a consistent feature of CBNRM and wildlife management in Botswana. As noted, the Government had used administrative decentralization to enable local communities to retain 100% of revenues from lease of land and wildlife quotas from the in dustry (less permit fees). Through Head Leases, communities acquired autonomy to generate revenue from tourism activities and full discretion over how to use this money. Indeed, problematically communities had been told it was their money to spend as they wished and, in the absence of guidelines or principles of participation, committees spent money freely, which eventually led to accusations that money was being wasted, misused and that ordinary people were not benefiting.

PAGE 164

164 In 1997, the Government introduc ed the community based strategy for rural development, within the National Policy for Rural Development. This strategy was driven by the ministry of local government and rural development (not the wildlife agency) and reflected a key shift in terms of Gove decentralization to more devolution of rights to communities. The new approach to rural development was grounded in the principles of empowerment and community led development. This policy was later revised in 2002, and CBNRM was identified as an engine for rural development. In 2003, the Government developed the Poverty Reduction Strategy, and within this directive CBNRM was again expressed as a key strategy for reducing poverty in rural areas, with tourism potential. Wh ile this was going on, the considerable stakeholder participation in developing a CBNRM policy had not yet led to this policy being formally adopted. Following 10 years of consultation and implementation of CBNRM, the Government finally revisited the CBNR M policy and made efforts to harmonize its goals. When the draft policy was initiated in 1996, devolution of natural resources was the agreed upon strategy by all stakeholders (IUCN, 2004). However, at this time Botswana was under going political change w ith far reaching implications for the wildlife sector. Ian Khama, with a long term interest in wildlife and the patrol of the Kalahari Conservation Society, had been elected Vice President from 1998 to 2008 and succeeded Festus Mogae as President in April 2008. Khama is rumored to have interests in photographic tourism operations, is anti hunting, and was also impatient with the misuse of money by communities. A long term military commander, he quickly installed one of his military colleagues, Onkokame M okaila, as Minister of Environment, Wildlife and Tourism. This transfer of power was

PAGE 165

165 occurring at the same time as the finalizing of the CBNRM policy, with both the President and Minister actively participating. In 2007, the CBNRM Policy of Botswana wa s enacted with the principle goal of integrating and creating an enabling environment for CBNRM. The 2007 policy links CBNRM to rural development ( Arntzen 2003 ) includes a range of resources including wildlife, veld products, fisheries and wood resources, and clearly addresses the aspect of community benefits. This policy reaffirmed the legal basis of community management on one hand. But on the other it made provision for the diversion of fully 65% of the revenues generated in community ma nagement areas into a National Environment fund. Interviews with stakeholders involved in these discussions indicate that the President, perhaps influenced by his friends on the photo graphic tourism industry, was fed up (legitimately) with the wastage of money by CBNRM Trusts. Even within some communities, many people supported this move because they were not seeing any of these benefits anyway and were resentful that leaders were squandering or stealing money. We were told that the initial intention w as to take money away from discretionary statement was added to the 65% clause, allowing but not insisting that this be implemented. In other words, the Minister could al low a community to retain 100% of revenues if he saw fit. This policy has been viewed as one that reversed the process of decentralization. Interviews suggest that this reversal is perceived to be highly detrimental by the increasingly educated and ca pacitated cadre of officials and agents supporting

PAGE 166

166 NGO support agencies and is perceived to reflect the interests of two men, the President and the Minister, and the largely w hite photo tourism sector. Poteete & Ribot contribute to this debate and have argued that this policy replaces more or less unconditional revenue flows with conditional grants and represents a significant partial re centralization (Poteete & Ribot, 2011). Poteete, 2009 has attributed this partial recentralization to the rhetorical linking of natural resource policies with political identity. Poteete argues that the rights over natural resources gained by local communities in the CBNRM program created incon diamonds and therefore created an opening for politically charged questions ( Poteete 2009 p 281 305 ) if communities that live with wildlife can retain all the revenue from wildlife, what about communities who live with diamonds? Intellectually, there are clear differences between the revenue sharing policies associated with a renewable res ource like wildlife that has opportunity costs to local people and non renewable below ground diamonds. This is clearly understood by many local people one of who encapsulated the hether wildlife could be treated as a national resource like diamonds, with the broader society gaining benefits from it, was used to support the position that led the Government of Botswana to partially re centralize rights to wildlife resources through t he CBNRM Policy of 2007. The consensus among CBNRM practitioners in Botswana is that the misuse of money by CBNRM Trusts is worrying, but that the blame should be placed on government and other agencies (the NRMP project is sometime mentioned) for not

PAGE 167

167 sup porting the development of sound local institutional arrangements. These commentators likewise see the solution as strengthened local institutions, and feel that recentralizing will only undermine progress that has been made and return Botswana to the situ ation in the 1980s when local people were replacing wildlife with other land uses because it was a government asset. Analytically, Botswana devolved the rights to benefit and negotiate joint ventures to community (through administrative arrangements) but n ot the right to manage, with the consequence that quotas are set centrally and communities are not protecting their resources. Capacity building support to communities is very limited in Botswana and although government agents attend many meetings little e ffort has been placed in supporting wildlife and institutional management systems. Similarly, both performance and conformance monitoring are sporadic at best. Consequently, multi village CBNRM communities are faring badly with very limited community par ticipation of benefit sharing (e.g. OCT, Chobe Enclave, KALIPA), but interestingly the three single village Trusts (Mambwe, Khwai, Sankuyo) are muddling along reasonably well within the same macro environment. Institutional Changes that have S haped CBNRM i n Zambia Efforts to enable greater community participation in wildlife management in Zambia emerged in the early 1980s ( Gibson 1999 ) These efforts were highly driven as national parks, the South Luangwa and Kafue National Parks. In a period of 20 years (1973 19 93), Zambia lost a total of 20, 000 rhinos (every last one) and elephant p opulations were reduced from 31,000 7, 000 in the South Luangwa National Park alone ( Gibson 1999 ) and from 100,000 to about 10,000 in the Luangwa valley generally.

PAGE 168

168 The underlying factor that led to such a vast decline in the elephant and rhino populations has been attributed to inadequate w ildlife conservation policies inherited from the colonial Government ( Gibson and Marks 1995 ) but the real reasons are more complex and link ed to organized crime, regional conflicts, a weak state combined with opportunistic elites, and even poverty. As a response to this downward spiral of wildlife populations, the Zambian Government, together with two donors, the United States Agency for Int ernational Development (USAID) and Norwegian Agency for Development Cooperation (NORAD), and a few dedicated conservationists, initiated two parallel programs to enable community participation in wildlife management. The USAID community project called ADMA DE (Administrative Management Design) was a national program which channeled revenues from safari hunting to local communities across much of the country ( Gibson 1999 ) ADMADE focused largely on protecting wildlife through a village community development fund controlled through local chiefs and their Integrated Resource Development Committees. Officially 40% of revenues were used to support GMA management (i.e. Village Scouts) and a further 25% retained by National Parks and Wildlife Service to oversee this activity. However, w hat is seldom mentioned is that these percentages were only calculated after half the money had already been retained centrally, and further un transparent marketing of hunting concessions generated far less revenue per unit area than in neighboring countr ies, by a factor of as much as ten. The NORAD funded program was the Luangwa Integrated Resources and Development Program (LIRDP) later known as South Luangwa Area Management Unit

PAGE 169

169 (SLAMU). The LIRDP/ SLAMU project was implemented in the Luangwa Valley and received funding from the Norwegian government. Initially, under LIRDP 40 % of hunting revenues and income from South Luangwa National Park were returned to the communities, through the same structures as described for ADMADE. When this did not work, LIRD revenues to 43 Village Action Groups and insisting on high levels of participation, transparency and accountability. This quickly resolved serious issues of financial mismanagement, resulted i n over 200 projects being constructed, and over 21,000 people receiving cash benefits each year ( Dalal Clayton 2003 ) to local communities. The ADMADE program granted local communities 35 40 percent of safari hunting revenues, while LIRDP directed 100 percent of hunting concession revenues to locals ( Gibson 1999 Child and Barnes 2010 ) The latter exemplifies a more deliberate effort at fiscally devolving power to communities. Despite these effort, fiscal dec entralization (ADMADE) and devolution (LIRDP/SLAMU) was not supported by any statutory rights to wildlife or decision making authority over wildlife uses ( Nelson and Agrawal 2008 p. 570 571 ) LIRDP initially decentralized revenues based on an administrative agreement between NORAD and GRZ embedded in project documents decisions. In 1996, the devolu tionary approach was agreed by a meeting between the Project, five Permanent Secretaries and NORAD. This agreement was protected by a clause in the bilateral agreement between d

PAGE 170

170 it was not enforced by their replacements, and so ZAWA was able to recentralize the majority of the finances ( Lubilo and Child 2010 ) Gibson and Marks have argued that instead of devolving managerial authority to benefits from wildlife ( Gibson and Marks 1995 ) This h as been a major constraint and critics argue that these programs had little positive impact in terms of encouraging community investments in sustaining wildlife populations in Zambia (Gibson an d Marks 1995 ) Proponents of both ADMADE and LIRDP are likely to make the counter claim that wildlife indeed was increasingly protected and began to recover (Child and Dalal Clayton 2004) The moderate successes of ADMADE and the significant success of LIRDP led to CBNRM being incorpo rated in legislation for the first time, with the irony that this same legislation was (mis) used to undermine CBNRM. The first formal recognition of the role of communities in wildlife management was the National Parks and Wildlife Act No, 12 of 1998. Th is followed a decade of implementing the two community programs, and was a result of a combination of detailed consultation with stakeholders in the wildlife industry. The Wildlife Act of 1998 No. 12 is the first legislative recognition of communities in d ecentralized wildlife management. However, the Wildlife Act had several flawed objectives. It reflected structural adjustment policies that Government would seek to transform all government departments whose activities had economic potential to function a long commercial lines. Thus the Act was driven primarily by the intention of establishing the Zambia Wildlife Authority as a semi autonomous agency (parastatal) and devolution was clearly not the primary motive that drove this

PAGE 171

171 institutional change. Indeed, the Act is quite confused about CBNRM. It focuses on establishing in some detail the structures for managing CBNRM (which should rightly be operandi, the Act did not specify which rights would be devolved to communities. Rather it established a system of representational governance based on Community Resource Boards, and ZAWA later he ld meetings at which it steamrollered communities to meetings, ZAWA did not mention that it had decided to maintain 80% of hunting concession fees for itself, and there fore that the real share to communities would vary around 30%. ZAWA continued to be the primary agency allocating hunting concession and this was done centrally and all at one time with participation of one person from each CRB. While not ideal, the hun ting allocation process was more competitive and transparent and prices increased significantly 2 ZAWA continued to set quotas, often building and performance monitoring was extremely limited; ZAWA, with expenditure of $5 7 million and income of $1 3 million, was expected to transform itself overnight into a financially self sufficient organization. It often could not even pay salaries, and the temptation to retain money gene rated in community areas was strong. Consequently, a 2 According to Child, the absence of clearly defined rights had previously led to allocation of hu nting concessions that are highly centralized and have been the subject of political manipulation. The use of transparent circumstances was also a mechanism used for patronage and the subject of much concern amongst Zambia conservationists Child, B. (2004). Parks in transition: Biodiversity, rural development and the bottom line UK, Earthscan, ibid.

PAGE 172

172 situation arose where the poorest people in Zambia, those in the most remote wildlife areas, were effectively subsidizing a struggling parastatal structure in Lusaka. In the past decade, CBNRM in Zambia has continued to deteriorate and has experienced resistance from powerful stakeholders with vested interests. This has led to the current status quo and impasse which is characterized by capture of community benefits by local elites, government, political officials and private sector actors (Mwape, 2003). Wynter Kabimba a Zambian lawyer who has been a keen proponent of community rights since the inception of CBNRM summarizes this status as follows Kabimba: I f you compare the functions of ZAWA and its relat ionship with CRBs you legislation. The CRB still works at the beck and call of ZAWA. Hunting companies have to pay ZAWA the money and then ZAWA decides when to pay the CRBs. CRBs have no way of verifying whether the amount of money paid to them is equivalent to the amount of money they should have been given. They are not given any audit reports by ZAWA unless payment to CRBs, apart from the slight mention of Chiefs as patron, there is no other structure regarding the CRB. There is also no form of regulation governing the CRB operations. The Act provides that ZAWA shall maintain a register of all CRBs that record is not at ZAWA. whole thing is operating in an informal way, as if the local communities are begging from ZAWA and ultimately the State regarding the benefits. This conversely has made it difficult for the communities to appreciate why they should look after thes e natural resources. They really do not see how they are linked with the natural resources they are supposed to protect A decade after the enactment of the Wildlife Act, the Word Bank commissioned a study to evaluate the economic, social and ecological impacts of the wildlife policies on communities. The study explicitly attributed the failures of the CBNRM program in the communal areas to the wildlife management policy. Its recommendations highlighted the need for a policy environment that devolves res ource rights to local communities. In particular, the study suggested that community participation should be formalized

PAGE 173

173 through legal entities, and rights to all natural resources in the area should be granted upon fulfilling laid out conditions to partici pate in the program 3 The study argues that devolved rights should include adequate authority and responsibility for the management, benefit and disposal of resources, as well as the right to exclude others (Simasiku, Simwanza et al. 2008) This study challenges the existing Act and its failures to fully devolve resource rights to local communities. A Comparative Assessment of D evolution in Botswana and Zambia The CBNRM programs in both Botswana and Zambia were initiated in the late 1980s with support from donor partners. These programs were promoted under the auspices of decentralization of natural resource management to rur al communities. The principle of devolution was the rhetoric of the programs, even though two decades later the outcomes are contrary. Both countries initiated CBNRM through donor funded projects funded and during this period it is clear in the objectives of the programs that both political and fiscal devolution was the principle of decentralization that was sought for rural communities managing wildlife. In Botswana the USAID funded CBNRM NRMP atural resources (Rozemeijer 2003 p. 3 4) In Zambia the USAID funded ADMADE program instituted some form of devolution of both political 3 This reflects the catch 22 situation that Martin and others warn about. Thus, M urphree's Principle of Rights requisite for responsible management and should not be held out as a reward ially through the affairs, so they cannot develop or demonstrate their management capacity. This is rather like saying to someone that they can be granted the right to drive a car, once they can demonstrate that they are safe drivers hence the reference to a catch 22 situation.

PAGE 174

174 and fiscal devolution through transfer of 34 40% tourist revenues t o local communities (Gibson 1999) This was even further devolved in the Norwegian funded LIRDIP/ SLAMU project where the state devolved both political and fiscal power to communities. Communities derived 100% of all benefits from wildlife (Child and Dalal Clayton 2004) Interestingly in both countries, decentralization or devolution occurred through administrative means, but when it came to formalizing these gains in policy or f it) were used for a pioneer act that transferred some wildlife user rights to local communities living in to shape the legal framework for devolution through CBNRM, the outcomes are characterized by increased recentralization by the state. Other scholars have observed that management of rights given to communities by the Wildlife Act of 1998 are not specifie d (Jones 2007) Another factor that h as contributed to recentralization outcomes is the fragmented sectoral legislation on how natural resources are managed. State actors have continued to centralize power of resources such as forests, using the weaknesses in the decentralization of wildlife as an excuse. This has reinforced state control over natural resources and undermined processes that effectively devolve power to the periphery. Despite the Act being interpreted as one providing strong devolution to communities, in the absence of specific provisions detailing community rights, government officials have manipulated its intent to avoid fiscal devolution and political empowerment. This is further corroborated not

PAGE 175

175 systematically matched by political will to pursue devolved resource management with vigor, (Child and Dalal Clayton 2004) Child argues that the Zambia CBNRM approach relies on a model of revenue sharing that leaves only 50% (at best) of income with contr ibution to national poverty reduction goals (Child and Dalal Clayton 2004) s Wildlife Act of 1998 No. 12 can be regarded as a legal instrument that inadequately addressed the question of devolution of community rights in the management of wildlife. As noted, the Act is poorly cast in that it gets into the detail of implementation without outlining principles of devolution and revenue sharing. While such shortcomings in Acts can be addressed through subsidiary policies that give more detailed prescriptive mechanisms on implementation, this was never done. As happened also in Tan zania (Nelson 2007) the enlightened intentions of progressive CBNRM policies and experiences were undermined by formal legislation and by the (mis) interpretation of Acts and policies developed with considerable stakeho lder input. The Wildlife Policy of 1998 was developed to support the Wildlife Act of 1998. This policy is progressive regarding community rights and participation. The Policy advocates for increased rights through the creation of Integrated Resources De velopment Boards (IRDBs 4 ), and advocates political and fiscal devolution of powers to local communities in GMAs. The policy advocates for political devolution through granting of legal rights to conservation, sustainable use and equitable distribution of b enefits earned by wildlife outside the national parks. The policy explicitly advocates for full fiscal devolution by outlining that ALL revenues from resource use in the area would 4 These are referred to as Community Resource Boards (CRBs) in the Wildlife Act of 1998.

PAGE 176

176 go to CRBs for socio e move towards devolution, it has never been fully utilized and is rarely referred to. Like Zambia, Botswana initiated CBNRM administratively by promoting political and administrative devolution, and the intent to initiate devolution was noted in the Wild life Conservation Policy of 1986; the National Conservation Policy of 1990; and the Tourism Policy of 1990. However, when the CBNRM Policy was finally adopted in 2007, and although it advocated rhetorically for devolved rights to communities through specif ied tenure rights and benefits, the clause allowing central retention of 65% of wildlife revenues does the complete opposite, and promoted partial recentralization compared to previous informal policies before 2007 (Poteete 2009) This partial recentralization is characterized by a reduction of fiscal powers to the communities. Where local communities formerly retained 100% of all revenues obtained from wildlife and other natural resource sales, the CBNRM 2007 policy splits this in to two streams i.e. 35% goes to c ommunities and 65% is retained in a National Environment Fund (Ndlovu, 2007). The 65% revenues are therefore only accessed through an application process, and can be applied for by any community not necessarily the one from which the benefits derived. The revenues contradicts the premise of CBNRM, weakens incentives for conservation, and ultimately disempowers the communities (Poteete 2007 p. 292) Note also that the Botswana Governments intent to ban hunting within 25km of protected areas is likely to greatly reduce the flow of revenues and meat to CBNRM communities. In Sankuyo for example, we estimated that average household incomes would decline by 40% for the period in which it took tourism to replace hunting, even assuming that this was possible.

PAGE 177

177 Meat from some 22 elephants and roughly 100 other animals is provided to communities like Mababe, Khwai and Sankuyo each year, and is another important livelihood be nefit especially for communities with origins in hunting and gathering. The intent to devolve power to local communities in wildlife rich areas in Botswana and Zambia have taken over two decades and is still ongoing. At the onset of the programs in both co untries partnerships between local conservationists and external agents (donors), plus the positive experience of CAMPFIRE in Zimbabwe, convinced the state that this was the best option for achieving both conservation and development goals. In both countri es, once donor funding was withdrawn CBNRM was deliberately recentralized even though the majority of stakeholders were opposed to this. Under these macro factors in both countries, it is highly unlikely to find communities that perform well within the cir cumstances. How Have Efforts To Decentralize Wildlife M ana gement in Botswana and Zambia, Affected Micro level Governance P erformance? To further examine the efforts made towards decentralization of wildlife management, I assess the performance of communi ties in Botswana and Zambia. The study assesses the performance of by answering the following question: How do current efforts to decentralize wildlife management in Botswana and Zambia affect the democratic governance and socioeconomic performance of CB Integrating Democratic Governance and Socioeconomic Indicators to Assess Performance of CBNRM Communities within Uncertain Macro Institutional Frameworks To assess the performance of each of the six communities, indices were developed on measures of democratic governance (conformance) and the provision of CBNRM socioeconomic benefits. These two variables are examined on the premise that

PAGE 178

178 communities whose members perceive higher social and economic benefits are more likely to practices the principles of democratic governance when compared to those that have poor democratic governance scores. Methods and Data A nalysis To obtain overall measures of the latent concepts of democratic governance and socioeconomic performance, composite indicator s were developed. These latent concepts cannot be measured through direct observations and are better estimated using a combination of multiple indicators (Bernard 2002, DeVellis 2003) They are based on sub indicators that have no common meaningful unit of measure and there is no obvious way of weighing these sub indicators. These composite indic ators can be used to summarize complex or multidimensional issues such as those of socioeconomic incentives as they relate to CBNRM governance. In addition, they reduce the size of a list of indicators to include more information within the existing size. The use of composite indicators also has certain weaknesses that may send misleading, non robust policy message if they are poorly constructed or misinterpreted. To avoid this, lpha. This measures an acceptable in t lpha of anything above 0.80 is acceptable. Selection of indicators. Indicators for measuring democratic governance were increased pol itical participation and competition (Dahl 1971, Coppedge 2005) The indicators for measures of CBNRM social and economic benefits were adapted from other previous studies that have examined the relationship of economic incentives in CBNRM (Emerton 2001, Barnes, MacGregor et al. 2002) We confirmed that these indicators measured a similar c oncept by calculating Cronbach Alpha as a measure of

PAGE 179

179 nternal consistency (Bernard 2002) The indicators were grouped into their respective index, following a simple averaging procedur e, such as that used for the well being index (Cahyat, Gonner et al. 2007, Miller, Holmes et al. 2010) This approach assumes equal w eighting for all indicators, which was adopted because of the ease of implementation and potential replication within community based monitoring schemes and the absence of additional data to determine appropriate weights. For democratic governance, the ind ex consisted of six items, including two on political participation, political rights, accountability, provision of information and satisfaction with democracy. For political participation, the two items measured whether respondents had attended the AGM an d elections. For political rights, respondents were asked whether they perceived a right to seven items. A measure of accountability was assessed, on whether respondents demanded accountability of their leaders or not. Another indicator of democracy measur ed was whether respondents had received information from their leaders or not (Table 5 2). averaged for each of the six indicators and further normalized to 0 100 scale to generate a democratic governance index. A similar procedure was used to develop a CBNRM socioeconomic index. A total of three items were used to measure CBNRM social and economic benefits i.e. cash, employment and social. Items were coded as 1 for having received the benefit in question, and 0 for not having received it. The index was built on a scale of 0 100, averaging binary responses from a set of 3 items.

PAGE 180

180 The data gathered for both democratic governance and the socioeconomic measures were analyzed using SPS S 16 and STATA software. Analysis included crosstabs and ANOVA. Results Assessment of democratic governance performance An assessment of the four indicators of democratic governance showed variation in the six communities. Political participation Two in dicators were used to measure perceptions on political participation in CBNRM. These included participation in the AGM and secondly participation in elections of board members. The average percentage score of those who participated illustrate that Khwai ha d the largest aggregate score of 86.5% followed by Sankuyo 60%. The communities with lower aggregate percentage scores on participation included OCT (29.5%) and Mulendema (29%) respectively. Political rights e measured using six indicators. These indicators included rights to be nominated for Board elections; to change the constitution; to join the CBNRM program; to make decisions on community finances; to participate in the AGM; to remove corrupt leaders and the right to vote. An aggregate percentage score was obtained for seven indicators for each community. The community with the highest score on political rights was Khwai (99%) followed by Mababe (95%), Sankuyo (95%), Mulendema (93%), Kasempa (78%) and the lowest score was OCT (69%) (Table 5 4).

PAGE 181

181 Accountability Perceptions on accountability were measured by assessing the likelihood of community members questioning their leaders on concerns of how the program was being managed. This vertical form of account ability (Schedler 1999) examines the relationship between elected leaders and their constituents. The results show that a higher percentage of respondents from Khwai (67.2%) and Sankuyo (67%) were more likely to question their leaders when compared to the othe r four communities in the sample. Only 53% of respondents from OCT were likely to question their leaders (Table 5 4). Provision of information The provision of information by leaders to their constituents is another important indicator in measuring democ ratic governance in CBNRM. Provision of information can be used a proxy indicator for measuring how accountable the leaders are to their constituents in terms of providing various types of information that is relevant for CBNRM management. Four indicators were used to measure provision of information and these included provisions of budget information, community projects, income from safari operators, and staff salaries (i.e. monies paid to staff working for the community trust or resource boards). An aggre gate percentage score of those who received this information was obtained for each community. The results show the community with the highest score on provision of information was Sankuyo (83%), followed by Khwai (81.2%), Mababe (65%), Kasempa (65%), Mulen dema (63%) and the lowest score was for OCT (43%) (Table 5 4).

PAGE 182

182 Assessing socioeconomic performance The provision of both social and economic benefits to communities is an important indicator measure of CBNRM community performance. The benefits range from i ndividual household benefits such as cash and employment to collective social benefits such as boreholes, community schools, etc. To assess the socioeconomic performance of each community, respondents were asked whether they had received any of cash, emplo yment or social benefits from the CBNRM program. Aggregate percentages of those who responded yes were obtained for each community and then assessed on a scale of 1 100. The results show that the community with the highest perception score of having rece ived CBNRM benefits was Mababe (89%) followed by Sankuyo (83%). The communities with poorest score with regards to provision of benefits were Mulendema (7%) and OCT (7%) (Table 5 5). The Zambian cases: w hy Kasempa c ommunity has had g overnance s uccess and M ulendema c ontinues to s truggle In Zambia, both Mulendema and Kasempa are multi village Community Resource Boards. Despite a multitude of donor projects linked to CBNRM, almost no progress nt farmers, accounts are weak, participation is extremely low, and there is very little evidence of capacity building efforts. By contrast, Kasempa is remote, smaller, and ethnically more homogenous. It has sound village level participatory process in pl ace, largely because it inherited a well trained ZAWA community facilitator from LIRDP and has energetic forestry officers.

PAGE 183

183 The case of OCT and why it continues to perform poorly when compared to single village communities in Botswana In Botswana, OCT is a multi village Trust that has been plagued by problems from the get go. The constitution allowed a small group of elites to capture the programme, and levels of participation and benefit sharing are low. The experience in other multi village Trusts in Bo tswana (i.e. Chobe and KALIPA) suggests that these conditions are general. Visitors to Chobe who meet the Trust are generally impressed, but this breaks down when one finds that ordinary people are largely excluded from the programme. KALIPA, consisting of three widely spaced villages along the Zimbabwe border is badly managed at all levels. The interesting lesson from Botswana is that the three single village, all of which we worked in, are performing at much higher levels than the multi village Trusts. People participate in AGMs and elections, and get cash, social and employment benefits. The performance of these villages does go up and down depending on the current nature of the committee or manager, and they may have an advantage in that they are clo se to Maun and more closely scrutinized by wildlife officials although we are not convinced this is the case. However, like the other Trusts (and unlike for example LIRDP/SLAMU and CAMPFIRE) clear, democratic procedures are not in place. In other work (i n preparation) we suggest that where small size is combined with sound procedures, some 80% of wildlife revenues are converted into cash dividends and community projects. We could only obtain financial expenditure data for Sakuyo which suggested that 59% of revenue directly benefitted communities, and we suspect that this was less so in Mababe and Khwai. Thus, we hypothesize that the absence of

PAGE 184

184 sound institutional capacity building and conformance monitoring reduces equitable benefit sharing from over 80% to 60% or less. (Murphree 1991, Hulme and Murphree 2001 p. 294) Since its inception, revenues from CBNRM in Southern Africa are earned from two sources of tourism, both consumptive (trophy hunting) and non consumptive (photographic tourism). Of the two forms, trophy hunting is particularly important for conservation and CBNRM as it creates incentives for conservation over vast areas of communal lands (Lindsey, Roulet et al. 200 7) Trophy hunters also pay higher fees per client when compared to conventional tourists (Ba ker 1997, Lewis and Alpert 1997) Communities in Botswana earn revenues from both these sources whereas the major source of revenues for communities in Zambia is from trophy hunting. The ability of ecosystems to provide economic returns through the susta inable use of natural resources is highly critical in CBNRM. Communities who are able to earn significant amounts of revenue from wildlife should in turn have more to share at both the household and communal level and therefore have higher participation in various aspects of governing their CBNRM community. Examining this relationship is therefore critical in assessing economic performance of individual communities. Methods and Data Collection Secondary data was collected from community files on trophy hun ting revenues earned by each community in the study within a period of five years. The trends on revenues earned are a good indicator for assessing the ability of the common pool resource to generate revenue. This section particularly seeks to answer the q uestion of

PAGE 185

185 whether revenues earned by individual communities are a good indicator of how well the community performs on governance aspects. The hypothesis is that communities with significant increases in wildlife revenues over time are more likely to perf orm better on democratic governance indicators such as increased participation, political rights, and accountability. Revenue trends Khwai trust (2000 2005) Revenue generated by the Khwai Development Trust (KDT) is obtained from joint venture partne rship with a private company, Soren Lindstrom Safaris. Revenue is earned from fees paid to the Khwai Trust for land rental and for animal quotas that the Department of Wildlife National Parks in Botswana provides to the communities. KDT was officially re gistered as a Trust in 2000, and it earned total revenues of US$184, 615. Within six years (2000 2005) of its inception, Khwai community earned a total of US$ 921, 842. The general trend of revenue for this community has been increasing over the years an d even though the community had a decline in revenues in ability to generate revenue (Figure 5 1). Revenue trends Mababe trust (2000 2005) ue is from both land rentals and animal quotas for hunting. Mababe has a joint venture partnership with a private safari company, African Field safaris, and was established in 2000. In their first year of establishment M ababe earned a total of US$103, 846.1 5. Within six years of its establishment (2000 2005), Mababe community earned a total o f US$ 944, 766.15. The only period the community experienced a decline in revenues was from 2006 to 2007, where there was a 40% decrease from US$281,945 to US$200, 000. The general trend for this community is

PAGE 186

186 that revenues have been increasing from the time of inception to the period when this data was obtained in 2007 (Figure 5 2). Revenue trends Sankuyo trust (2000 2005) Sankuyo Trust was one of the earliest CBNRM programs established in Botswana in 1995. This community has diversified its revenue base over the years. Initially, the revenue was from land rentals and animal quotas for safari hunting, however in recent years, Sankuyo Trust has ownership and manages a community campsite and the Santawani Lodge. Two years after its establishment, Sankuyo earned US$53, 076.92. Within a period of six years (2000 2005), Sankuyo communi ty earned a total of US$1, 090, 335. Revenue earned by Sankuyo since its inception show trends of increase over the years. The only period of decline was from 2000 2001 when there was a drastic decline of revenues from US$9 1,609.23 to US$10, 769.23 ( Figure 5 3). Revenues trends OCT (2000 2005) RM programs and was established in 1995. Its major source of revenue is photographic tourism, and currently sub leases its area to Wilderness Safaris, the largest tour operator in Southern Africa. Revenues earned by OCT from 2000 2 005 amounts to a total of US$1,369, revenue trends indicate that it has been increasing over the years. Comparisons of revenue earned in 1997, two years after the program was established, and 2005 last point of data collected, show a 77% increase (Figure 5 4). Reve nue trends Mulendema CRB ( 2003 2008 ) from Mumbwa game management area (GMA), a buffer zone located in the Kafue National Park. In Zambia, hunting revenues are obtained from bot h land rental and from

PAGE 187

187 animals on quota. Mulendema community and the Zambia Wildlife Authority (ZAWA), have an agreement, which stipulates a 50 50 share of revenues from these two sources. The 50% share the community earns from the hunting block is furth er divided among three other community groups, who get a third of the 50% each. From 2005 2010, Mumbwa GMA earned a total of US$ 857, 880 from hunting revenues (Figure 5 5). Of this amount, Mulendema CRB received a total of US$142, 979 a share of 17% of t he total revenue. During the six year period of data collected, Mulendema community was paid an average of US$ 23, 839.95 Revenue trends Kasempa CRB ( 2005 2010 ) Kasempa CRB earns revenues from the Lunga Luswishi GMA and currently, has two hunting bl ocks leased to Prohunt and Royal Zambezi hunting outfitters. Kasempa CRB is entitled to 50% share of all revenues earned from safari hunting in this GMA and 20% from land rentals. From 2005 2010 the total revenues earned for the Lunga Luswishi GMA amount ed to US$ 656, 220, of which a share of US$ 328, 110 was paid to Kasempa CRB. During the six year period of data collected, the communi ty earned an average of US$ 54, 685 (Figure 5 6). D iscussion The enabling macro environment for CBNRM. In this chapter we broadly examine the institutional changes that have taken place in CBNRM in Botswana and Zambia and how this has in turn affected both socio economic and democratic governance performance of six communities in these countries. Our study pursues the argu ment that the success of CBNRM is highly dependent on several factors, of which the macro level enabling environment is paramount. We sought to understand how CBNRM communities performed in terms of democratic and socioeconomic attributes

PAGE 188

188 within the uncer tain macro policy environments that we described in some detail. This analysis and discussion was informed with more general scholarship on decentralization and devolution. Table 5 1 summarizes our finding for the six communities; comparing our assessment of the macro environment with the data we collected to describe outcomes. In the three single villages, technical support to improve joint venture negotiations (provided through the NRMP programme and later by the Ministry) coupled with the right to retai n all revenues have resulted in a rapid increase in wildlife derived revenues, and also in widespread benefits of the community. This is not the case in multi village that we sampled in Botswana (OCT) with the increased level of benefits paid by the priva te sector getting no further than the elite who control the Trust. In Zambia, revenues were stagnant, partly because of un transparent marketing on concessions, and partly because money was retained by ZAWA and reliable records are unavailable (see Wynter Kabimba quote above). In an unsupportive macro environment, participation appears to be dictated by micro circumstances and is much higher in the three small communities and in of their political rights is higher in single villages (95 99%) than in multi villages (69 78%) though for complex reasons people in the worst governed community of all have a high CBNRM accountability is also higher in single villages (82 83% versus 43 65%) ; with Mababe being different (65%) perhaps because we surveyed it during a time of conflict.

PAGE 189

189 However, the performance of all these communities in terms of NRM is much lower tha n Zimbabwe and particularly Namibia, where NGOs placed particular emphasis on building these capacities. In terms of governance and benefit sharing, this and other (unpublished ) results showed that the level of participation and equitable benefit sharing in all multi village communities in Botswana, Zambia and Namibia is low with less than 20% of wildlife revenues benefiting ordinary people through cash or projects. There are two exceptions. In single villages, as noted above, and as the inions on benefits confirm, up to 60% of revenues get shared, even in the absence of sound procedures. If this is benchmarked against single village sites where sound procedures and external conformance monitoring are in place, such as CAMPFIRE and LIRDP, this share rises to 80% or more (Child and Dalal Clayton 2004) Multi villages, be nchmarking (where only unpublished data was available from CAMPFIRE, 60% of revenue is shared in multi villages) suggests that external conformance monitoring increases revenue sharing very substantially from less than 20% to about 60%. Thus, external con formance monitoring adds considerable value in terms of equitable benefit sharing, especially in multi village communities where governance is particularly prone to elite capture. Finally, it was noted that improvements in NRM are correlated with NRM monit oring, with Namibia having excellent community based wildlife monitoring and rapidly rising wildlife populations (NACSO 2008) and Zimbabwe having a sound quota monitoring system and increasing wildlife populations (Child, Jones et al. 2003) Wildlife monitoring in Botswana is sporadic, and Zambia almost nonexistent and related to this action is not being taken to address what appear to be declining wildlife

PAGE 190

190 populations. The correlation between monitoring a nd improvements also apply to governance and revenue sharing. Except for CAMPFIRE to some extent, and LIRDP at project level, there is almost no monitoring of CBNRM benefits at household level, and performance is universally low except in the cases mentio ned and in the three single villages. It is a bit of a jump, but this suggests that the adaptation of policy and implementation in CAMPFIRE (before 2002) and Namibia is linked to what data is monitored, and conversely that the failure of CBNRM to adapt in Botswana and Zambia, indeed the retrogression of these programmes, can be partly explained by the absence of performance monitoring.

PAGE 191

191 Table 5 1. Devolved rights and micro governance of natural r esour ces, i nc ome, benefits and political par ticip ation Khwai Mababe Sankuyo Kasempa Mulendema OCT Type of governance Single Village Single Village Single Village Multi Village Multi Village Multi Village Country Botswana Botswana Botswana Zambia Zambia Botswana DESCRIPTIO N OF ENABLING ENVIRONM ENT Rights to benefit 100% 100% 100% 30% 30% 100% Rights to sell Yes Yes Yes No No Yes Rights to manage No No No No No No NRM capacity building No No No No No No Marketing wildlife leases Yes Yes Yes No No Yes Institutional capacity building No No No No N o No NRM performance monitoring No No No No No No Governance conformance monitoring No No No No No No OUTCOMES (Table 5 4 and Figures) Revenue increase 2X 2.5X 3 4X 0.5X No increase 3X Participation 87% 55% 60% 57% 29% 29% Political rights 99% 96% 95% 78% 93% 69% Accountability 82% 65% 83% 65% 63% 43% Benefits 66% 89% 83% 19% 7% 7%

PAGE 192

192 Table 5 2. Indicators for measuring democratic governance DEMOCRATIC GOVERNANCE INDICATOR RESPONSES POLITICAL PARTICIPATION Participation in AGM Yes/ No Partici pation in Elections Yes/ No POLITICAL RIGHTS Right to be nominated for Board elections Yes/ No Right to change constitution Yes/ No Right to join CBNRM program Yes/ No Right to make decisions on CBNRM finances Yes/ No Right to participate in AGM Yes/ No Right to remove corrupt leaders Yes/ No Right to vote in elections Yes/ No ACCOUNTABILITY Demand for leadership accountability Yes/ No PROVISION OF INFORMATION Information on annual budget Yes/ No Information on community projects Y es/ No Information on hunting revenues Yes/ No Information on staff salaries Yes/ No SATISFACTION WITH DEMOCRACY democracy Yes/ No Table 5 3. Indicators for CBNRM socio economic benefits CBNRM SOCIOECONOMI C INDEX RESPONSES Cash benefit Yes/ No Employment benefit Yes/ No Social benefit Yes/ No

PAGE 193

193 Table 5 4. Democratic governance performance KHWAI MABABE SANKUYO KASEMPA MULENDEMA OCT % YES N % YES N % YES N % YES N % YES N % YES N VARIABLE S & INDICATORS Political participation Participation in AGM 84 46 64 34 61 43 37 37 19 23 15 19 Participation in elections 89 49 46 23 59 42 77 77 39 47 44 57 % Mean 86.5 55 60 57 29 29.5 Total N 55 53 70 100 120 12 9 Political rights Right to be nominated for Board elections 100 55 94 48 100 70 72 71 98 118 76 98 Right to change constitution 96 53 94 48 100 70 72 71 98 118 41 53 Right to join CBO program 100 55 96 50 64 45 68 64 64 77 76 98 Right to make decisions on CBO finances 100 55 94 48 100 70 73 72 96 116 54 70 Right to participate in AGM 100 55 96 50 100 70 78 98 118 71 91 Right to remove corrupt leaders 100 55 96 50 100 70 90 89 98 118 70 90 Right to vote in elections 100 55 100 50 100 70 95 94 98 118 94 121 % Mean 99 96 95 78 93 69 Total N 55 51 70 99 120 129 Accountability Likelihood of communities questioning Chairperson 67.2 55 54 52 59 70 67 98 61 113 53 122 Provision of information Annual Bud get 80 44 75 38 86 55 68 64 59 61 53 65 Community Projects 76 42 61 31 83 53 78 73 68 70 42 51 Income from safari operator 89 49 65 33 83 53 59 55 64 66 48 59 Staff salaries 80 44 58 30 79 51 56 53 60 62 29 35 % Mean 81.2 65 83 65 63 43 Total N 55 51 64 94 103 122

PAGE 194

194 Table 5 5. Socioeconomic performance CBNRM Benefits Index Khwai Mababe Sankuyo Kasempa Mulendema OCT % Yes N % Yes N % Yes N % Yes N % Yes N % Yes N Cash 70 38 96 51 69 48 12 12 03 4 02 3 Employment 62 34 100 53 100 1 00 13 13 05 6 15 19 Social 65 36 71 38 80 56 33 33 13 16 04 5 SOE Score 65.7 55 89 53 83 70 19.33 100 7.00 120 7.00 129

PAGE 195

195 Figure 5 1. Khwai revenue Figure 5 2 Mababe revenue

PAGE 196

196 Figure 5 3. Sankuyo revenue Figure 5 4. OCT revenue

PAGE 197

197 Figure 5 5. Mulendema revenue Figure 5 6. Kasempa revenue

PAGE 198

198 CHAPTER 6 CONCLUSIONS The Importance of Micro and Macro Governance Institutions in Shaping CBNRM Performance The concept and practice of CBNRM has over the past two dec ades attained a central role in how communal lands are managed in southern Africa. Initially started as experiments to address the challenges of poor management by the state of protected areas, it has gained recognition through development and conservation sectors, as the model for managing common pool resources such as wildlife, forests and fisheries. Institutions of collective action such as CBNRM are critical in shaping how communal lands in southern Africa can be better managed to ensure provision of be nefits to rural communities, empowering them through transparent and participatory forms of governance and ultimately improving management of ecosystems. Despite these good intentions, very few cases of success are evident in the southern African context. Critics have been quick to view CBNRM as a failed model for conserving communal lands. These criticisms include the heavy reliance of most CBNRM programs on donor funding that is short term and sustainable, the continued depletion of the wildlife resources the increasing practices of unsustainable use of natural resources and finally how these programs have continued to disempower local communities. Regardless of these criticisms, CBNRM still continues to have champions both in the practi ti oner and academ ic world that believe in its success. Though few and far between, there has been a reexamination, on what factors actually make CBNRM work? After two decades of its implementation in southern Africa, a critical mass of scholars in the region, begun to ask key questions on how these programs could be

PAGE 199

199 better evaluated. SASSUG, a group of scholar practitioners in southern Africa, identified the important role that governance played in shaping performance of CBNRM. This was a response to concerns that the model advocated for in pioneer CBNRM programs had a narrow focus on governance. The implications of this were that the effects of poor macro and micro governance began to emerge. Of particular concern was what actually governance in CBNRM looked like, and how i t impacted performance of CBNRM programs. Efforts towards understanding this have emerged over the past few years. As part of contributing to this debate, this doctoral research examined governance systems in CBNRM and their role in providing public goods. The study was undertaken by addressing three objectives: Examining the importance of Annual General Meetings as platforms for improved democratic governance Examining the factors that influence community participation in CBNRM elections and its role in improved CBNRM governance. Examin in g the macro enabling conditions that shape micro CBNRM governance and the provision of public goods. Role of Micro Governance Institutions and Recommendations for I mproved CBNRM governance With respect to AGMs, the stu platforms for increasing participatory forms of governance in CBNRM. These institutions play a critical role in ensuring that ordinary members of the communities participate in the governance and distribution of b enefits from the CBNRM programs. The study shows that single village communities perform better in organizing these meetings when compared to multi village communities that have more representational forms of governance. In addition, that study shows that participation at these meetings is highly dependent on the incentives individuals receive from the CBNRM programs. Among the

PAGE 200

200 incentives, cash benefits largely influenced whether members participated in AGMs or not. The study further shows that elections i n CBNRM are an important institution that increases accountability of leaders to their constituents. The findings show that among the six communities in the study, the single village communities have held more elections when compared to the multi village c ommunities. It shows that there were higher numbers of community members from single villages who had participated in elections of their leaders when compared to their peers from multi villages. One of the contributing factors to low participation in multi villages is the high cost of organizing larger groups of people from different villages. Another factor was the lack of incentives for members to participate in elections. Community members from single villages were more likely to participate in elections as they received both cash and social benefits at the same meetings where elections were held. Both AGMs and elections are important institutions for improving micro governance in CBNRM. These two institutions are proxy measures of participatory democrac y, the main model of democracy advocated for in CBNRM. They are therefore important indicators for measuring governance both in programs and research that examine CBNRM performance. This study identifies how these institutions are better placed in single v illage communities that are homogeneous, but are poorly aligned to multi village, heterogeneous communities. This has huge implications for CBNRM going forward, as the majority of communities in southern Africa, are predominantly large village communities. It is crucial that more investment be made to strengthen conformance monitoring in future CBNRM programs. Conformance monitoring has

PAGE 201

201 contributed to the improved performance of CBNRM programs in Namibia. This role has mainly been undertaken by civil societ y, and has shown to be critical in ensuring that the micro level is externally monitored. The question remains however, what happens in communities with very little civil society presence. How can governance be strengthened through conformance monitoring b y external agents? An area that might need to be explored in future programs is the role of district councils. Despite poor records of this model in the CAMPFIRE program in Zimbabwe, they are still an important institution in providing checks and balances to micro level governance in CBNRM. Over the past few years, the move to decentralize development to local authorities has grown in importance and recognition. This debate has been taking place particularly in Zambia, and this could be an important institu tional change for future governance at the micro level in CBNRM could be the future of how these programs move forward. Macro Enabling C onditions for CBNRM Governance The final objective of the study was to examine the macro enabling conditions that shape CBNRM micro governance and the provision of CBNRM public goods. The recently moved towards m acro policies that have recentralized wildlife management from the local communities. The study further shows that there is some variation in how these outcomes have been shaped in both countries, with Botswana having a milder form of recentralization as c ompared to Zambia. Our findings further show that despite these current macro factors shaping CBNRM, the three single villages in our study have been able to improve their natural resource base, increase incomes to livelihoods and

PAGE 202

202 have shown increased demo cratic governance when compared to their peers from multi village communities. Lessons from Namibia and Zimbabwe and Future I mplications for CBNRM in Botswana and Zambia In both Botswana and Zambia devolution was the intended form of decentralization, a nd this was advocated in the earlier stages of these CBNRM programs but for different reasons was not sustained. Devolution was promoted by a network of national and regional conservationists (SASUSG 1996) who gained the support of donors to support these programs, and in turn got buy in from state actors. As Murphree says, the Khaki shorts brigade [i.e. conservationists] did an end run around the political process to initiate devolution, and maybe in some cases did not even know the political consequences of their actions to empower local communities In Zimbabwe, the wildlife agency did recognize the politica l nature of CAMPFIRE, and quickly developed a grassroots political organization (the CAMPFIRE Association) to which it passed primary responsibility for the programme. The integration of CAMPFIRE into national political processes is, to a large extent, th e reason why CAMPFIRE maintained its effectiveness well after donor support was withdrawn (Child, Jones et al. 2003) and, though not much written about today, is still operational. Legislated devolution in Zimbabwe and Namibia are another strength of these programmes that is missing in Botswana and Zambia. Similar, attempts to develop such organizations in Botswana and Zambia were not successful, so with the withdrawal of donor projects or key personnel, no mechanism was in place to defend devolution even where most civil servants and many community members were in favor of it. Thus, we see devolution being undermined by state actors post the donor projects

PAGE 203

203 a legislation, lack of subsidiary legislation to guide policy implementation, and above all the n eed of ZAWA to fund itself from community revenues, are factors that have enabled moves towards recentralization. In the case of Botswana, high level political actors have been able to reverse the devolutionary process against the will of the majority bec ause the majority is not politically organized, because administratively devolved rights were coming up for their 15 year renewal, and because the poor performance of some communities was used to argue (in the absence of effective monitoring systems) that CBNRM was wasting money. Despite these constraining factors, Botswana would rate higher in the move to devolve natural resource rights when compared to Zambia. This is evident in the manner Botswana pursued its CBNRM policy implementation. In Botswana, CB NRM was implemented legally and programmatically, and after several years of consultation (1995 2007), a policy was agreed by parliament to guide further implementation (even if the policy contained the highly unfavorable clause that communities lose 65% of their revenues). CBNRM is supported by many civil servants in Botswana who feel greatly In comparison, Zambia is generally a much more personalized state than the more rule following Botswana, and CB NRM similarly has been greatly affected by the influence of a few key actors, both positively and negatively. In other words, CBNRM in Zambia is not really managed through any formal programmatic process. It moved forward when certain conservationist and donors linked up with key government

PAGE 204

204 officials, but rapidly went backwards if these individuals were relocated and especially after ZAWA began to perceive (however inequitably) that it could finance its activities from hunting in communal areas even thoug h it is obvious that this strategy is unsustainable. Thus in Botswana, CBNRM was facilitated and implemented through Government with initial support from donors such as USAID and SNV. This support was later continued post donor funding and the Department o f Wildlife and National Parks was the key implementer until 2009 when this role was realigned to the Botswana initially, the support and commitment has over the years waned, the support for and capacity to implement CBNRM in government agencies is low, and there has been increased competition for resources between the Zambia Wildlife Authority (quasi government institution) and the communities. In Zambia the incentives for sta te actors to devolve power to local communities continue to diminish, because wildlife officials do not get salaries or allowances unless these are self funded 1 By contrast, in Botswana the level of direct competition between state actors and communities is relatively low as the wildlife department does not compete for funding with communities: it is relatively well funded from state coffers (diamonds) and community monies that are centralized are ring fenced in Trust. The future is hard to predict. Wild life populations in Zambia are highly likely to continue to decline outside protected areas (and even inside them) unless the current policies are reversed, but individuals within ZAWA have strong short term incentives to 1 Nelson and Agrawal, 2008 identify th is as the reason why there is lack of a deliberate move towards decentralization in Zambia.

PAGE 205

205 perpetuate the current inequitable situation. In Botswana, recent reports show a large decline in wildlife, especially in the Okavango delta ( Chase 2011) There is also widespread support for CBNRM and democratization, but civil society is weak relative to government, indeed in this small country much civic activity appears to occur through government. Recentralization is driven strongly by o ne or two high level individuals against the will of the majority, and the outcome of CBNRM is likely to be determined by whether these individuals can be won over to the cause of devolution, or the outcome of the next election cycle.

PAGE 206

206 APPENDIX A DEMOCRA TIC GOVERNANCE QUESTIONNAIRE BOTSWANA Respondent Number Research Assistant Name QUESTIONNAIRE ON UNDERSTANDING THE NATURE OF DEMOCRATIC INSTITUTIONS IN CBNRM Household Selection Procedure : Interviewer: From the membership list provided, we selec t respondents using a simple random sample procedure. Interview includes members that are 18 years and above. INFORMED CONSENT Good day. My name is: ___________________________________. I am a Researcher with the University of Florida and this is an indepe ndent research institution. I do not represent the Government or any Non Governmental Organization. S TATEMENT ABOUT INFOR MED CONSENT : 1. The purpose of this study is to understand what you think about your how things are governed in your Community. [ Maikael elo a patlisiso ee ke go batla go tlhaloganya ka fa dilo di dirwang ka teng mo morafeng wa lona ]. 2. We would like to learn more on how you vote ; chose your leaders and participate in various political activities in the Trust. [ Re eletsa go itse ka fa lo tlho phang moemedi wa lona wa lona wa tsamaiso ya mokgatlhong wa lona ] 3. This survey should take less than one hour [ Re solofela fa patlisiso e tla tsaya nako e sa feteng oura ] 4. You do not have to answer any question you do not want to [ Ga o patelesege go araba di potso tse fa o sa battle ] 5. All information is confidential [ Sepe se se o tla se buang e tla nna sephiri ]. 6. You can stop the interview process at any time [ O letlelesega go ka emisa nako nngwe lenngwe ] 7. You can ask for clarification on any question at any time [ O ka kopa go tlholosediwa fa o sa tlhologanyeng teng nako nngwe le nngwe fa o batla ] 8. There is no right or wrong answers, and most importantly candid and honest answers are most useful. [ Dikarabo tsotlhe di a amolesega, se se botlhokwa segolo bogolo ke d ikarabo tse di boammaaruri ebile dina le mo mosola ]. 9. There are no direct benefits, risks, or compensation to you for participating in the study [ Ga re solofetse dikatso dipe mo go tseeng karolo mo patlisiso e ] For questions about your rights as a resea rch participant contact the IRB at Tel: 352 392 043

PAGE 207

207 1. BACKGROUND CBO CBO VILLAGE CONTROL HUNTING AREA (CHA) Khwai Development Trust Khwai NG 18 Mababe Zokotsama Community Development Trust Mababe NG 41 Sankuyo Tshwaragano Management Trust San kuyo NG 33 & NG 34 1.1 Household Check: Infrastructure [ Tshekatsheko ya malwapa: Dikago] Please Tick [ Tshwaa ] Please Tick [ Tshwaa ] 1.1.1 House: [ Ntlo] Traditional (Mud): [ Ya ngwao ee agilweng ka mmu] Cement Block: Samente Yes [ ee ]: .................. Yes [ ee ]: ................... No [ nnyaa ].................. No [ nnyaa ].................. 1.1.2 Roof: [ Marulelo ] Thatch [ Bojang ] Tin [ Senke ] Yes [ ee ]...................... Yes [ ee ]..................... No [ nnyaa ].................. No [ nnyaa ]................. 1.2 Are the following services present in the Village [ Interviewer determines ] [ A ditlamelo tse di latelang di teng mo motseng] Yes [ ee ] No [ nnyaa ] Determine [Gakeitse Electricity grid that most houses could access [ Kgokelo ya motlakase ee gaufi ] Piped water system that most houses can access [ Metsi a a phepa a a ka fitlhlelwang ke malwapa ka bontsi ] Sewage system that most houses could access [ Kgopo ya metsi a a leswe e e ka fitlhlelwang malapa a mantis ] Cell phone service [ Kgokelo ya me gala ya letheka ] 1.3 Are the following facilities in the Village, within easy walking distance? [ A ditlamelo tse di latelang di gaufi? ] [ Interviewer determines this ] Yes No

PAGE 208

208 [ ee ] [ nnyaa ] Determine [ Gakeitse ] School [ Sekolo ] Police Station [ Diofisi tsa sepodise ] Health Clinic [ Kokelwana ] Market Stalls (selling groceries and/ or clothing) [ Marekisetso a dijo le diaparo ] 2. DEMOGRAPHICS 2.1 [ Are simolole ka go itse k a gag ago ] [ age ] [ Mmotsolotsi, kopa digwaga tsa ba o ba botsang, kopa dikarata tsa omang mo go ba ba sa itseng dingwaga tsa bone go bona dingwaga tsa bone ] 2.2 Ho w old are you? [ O dingwaga di kae ?]............................................. 2.3 How long have you lived in the village? [O nnile mo motseng o lebaka le le kae?]........................................................................................... .................... 2.5 Which is your home language? [ Ke teme/ puo e fe e le e dirisang mo lwapeng? ] [ Interviewer: Prompt if necessary: That is the language of your group ] Please tick [ Ka tsweetswee tshwaa ] Code [ Letshwao ] English 101 Setswana 201 Seyei 202 Sembukushu 203 Sesarwa 204 Setawana 205 Seherero 2 06 No Response [Ga gona karabo] 207 [ Please tick ] [ Interviewer: Tick correct response number ] Yes [1] [ ee ] No [2] [ nnyaa ] Know [9] [ Gakeitse ] 2.4 Are you the Head of the Household ? [ A ke wena tlhogo ya lelwapa? ]

PAGE 209

209 Other [ indicate ][Supa] 208 2.6 Sex [ Bong ] Male [ Rre/ monna ]: Female [ mme/mosadi ]: 2.7 Education Grade [ Tsa thuto ] No Education [ Ba basa tsenang sekolo ]: Some Primary School [ Sekolo se se botlana ] Completed Primary School [ Sekolo se se botlana ]: Junior Secondary (F1 F2 F3) [ Sekolo se segolwane ]: Senior Secondary (F4 F5) [ Sekolo se segolwane ] College [ Sekolo se segolo/sa ithutel o ditiro ] 2.8 Gender of Head of Household [ Bong jwa tlhogo ya lelwapa ] Male [ Rre/mo nna]: Female [ Mme/ mosadi ]: Female (male away for 6 + months/ year: [ mme (Rre a seo lebaka la kgwedi tse thataro kgotsa Ngwaga )] 2.9 No. of peopl e in Household: [ Palo ya ba ba nnang mo lwapeng ] Adults: Male Female Children: Male Female 2.10 No. of people in Household employed in: [ Palo ya ba ba thapilweng mo lwapeng ] CBO: [M okgatlho]..................... Tourism [ Tsa Bojanala ]: ..................... Hunting [ Tsa letsomo ]: .................... Other [ Tse dingwe ]: ............................................................. [ Specify, prompt to get respondents to recall names in household in other employment ][Rotloetsa baarabi go fa maina] 2.11 Position in CBO [ Maemo mo Mokgatlho ] : Manager/ Management Employee [ Moeteledipele ]...............................................................

PAGE 210

210 [Tick/ Tshwaa ] Employee (Game Guard, Community Scout).............. [ motlhokomedi wa diphologolo ] CBO Committee.......................................................... [ Komiti ya Mokgatlho ] Ordinary Member........................................................ [ leloko fela ] Other: [ s pecify ]........................................................... [ Tse dingwe [tlhalosa )] LIVELIHOOD STATUS AND PROVISION OF CBNRM GOODS [ Seemo sa itshetso/ tse di abiwang ke ba CBNRM ] 3.1 In general, how do you rate the way you live compared to those of ot her members in this community? [ O kala jang ka fa o tshelang ka teng fa o itshwantsha le ba bangwe mo motseng ] [ Read out response options, please tick ] Code [ Letshwao ] Tick [Tshwaa] Much worse [ Se kwa tlase thata ] 5 Worse [ Se kwa tlase ] 4 Same [ Di a tshwana ] 3 Better [ Se botoka ] 2 Much Better [ Se botoka thata ] 1 Ga ke itse ] 9 3.2 Over the past year, how often, if ever, have you or anyone in your family gone without : [ Mo dingwageng tse di fetileng a go na le mongwe mo lapeng yo o neng a nna a sena tse di latelang :] [Read out options] Never [ Nnyaa gotlhelele ] Just once or twice [ Gangwe kgotsa gabedi ] Several times [ Makgetho a mantsinyana ] Many times [ Makgetho a mantsi ] Always [ Nako tsotlhe ] Know [ Ga keitse ] COD E [ Letshwao ] 5 4 3 2 1 9 3.2.1 Food to eat? [ Dijo? ]

PAGE 211

211 3.2.2 Clean water for home use? [ Metsi a a phepa go dirisiwa mo lwapeng? ] 3.2.3 Medicin es or medical treatment? [ Melemo kgotsa tsa kalafi ] 3.2.4 Fuel to cook your food? [ Tse di dirisiwang go apaya? ] 3.2.5 A cash income? [ Madi a dituelo ] 3.3 Looking back 10years ago [ 1999 say year ], is your community better off now, or before CBNRM started? [ Fa o leba, mo dingwageng tse some tse di fetileng, a o bona tokafalo mo bathong ba motse wa gago gompieno, kgotsa pele ga CBNRM e simolola? ] [ Read out response options & tick ] CODE [Letshwao] Tick [Tshwaa] Much worse [ Seemo se maswe thata ] 5 Worse [ Seemo se maswe ] 4 Same [ Go tshwana fela ] 3 Better [ Go botoka ] 2 Much Better [ Go botoka thata ] 1 Ga ke itse ] 9 3.4 What would you say are the major changes that have been brought about by CBNRM? [ Ke diphetogo dife tse di tsisitsweng ke CBNRM?(tse disiameng) ] [ Interviewer: prompt good and bad changes ] .................................... ........................................................................................................................ ............................................................................................................................. .......... .....................

PAGE 212

212 ............................................................................................................................. ............................... ............................................................................. ............................................................................... ............................................................................................................................. ............................... ................... ............................................................................................................................. ............ ...................................................................................................................... ...................................... 3.5 past 5 years, has anyone in your household received any of the following benefits? [ Bomosola jwa CBNRM. A gona le mongwe mo lwapeng la lona yo o a mogetseng tse di latelang, mo ngwageng tse tlhano tse di fetileng? ] 3.5.1 Employment [ Tiro ] Safari Operator No. Employed in HH.................. Tourism/ Lodges No. Employed in HH....... ........... CBO Employment No. Employed in HH.................. 3.5.2 Cash Dividends [ Dituelo ] Yes [ ee ] No [ nnyaa ] Amount [ Selekanyo ]: ....................... ............................................. 3.5.3 Orphanage Fund [ Dithuso tsa masiela ] Yes [ ee ] No [ nnyaa ] Amount [ Selekanyo ]: ................................................................ ..................... 2. [ Dithuso tsa bagodi ] Yes [ ee ] No [ nnyaa ] Amount [ Selekanyo ]: .................................................................................. ......... 4. POLITICAL PARTICIPATION IN CBNRM [ Go tsenya letsogo ga sepolotiki mo CBNRM ]

PAGE 213

213 4.1 Membership: [ Boloko ] 4.1.1 As a member of .................................. village do you have a r ight to join the CBNRM program? [ lenaneo la CBNRM ] [Interviewer mention the name of village] Yes [ ee ]: No[ nnyaa ]: Gake itse ]: 4.1.2 Here is a list of positions in the CBNRM community. For each one, could you tell me whether you are: [ Mo maemong a CBNRM a a latelang, ke afe a o mo go one? ] [Tick box] Tick [ Tshwaa ] CODE [ Letshwao ] a. Trust Leader [ Moete ledipele wa lekgotla ] b. Trust Employee [ Mothapiwa/ mohiriwa wa lekgotla ] c. Active Member [ Leloko le le matlhagatlhaga ] d. Not an active Member [ Leloko le le seng matlhagatlhaga ] [If respondent ticks (d) ask question 4.2, if not proceed to question 4. 3] [ Fa karabo e le (d) botsa potsa 4.2, e seng jalo fetela kwa go 4.3 ] 4.2 Please list the reasons why you are not an active member of the Trust? [ Tsweetswe, bolela mabaka a a go tlhokisang matlhagatlhaga mo bolokong jwa gago? ] ............................... ............................................................................................................................. ............................................................................................................................. ...... ......................... ............................................................................................................................. ............................... ........................................................................... ................................................................................. ............................................................................................................................. ............................... [If respondent, ti cks (a c) ask the following set of questions] [ Fa karabo e le (a c) botsa dipotso tse di latelang ] 4.3 Participation in Annual General Meetings: [ Tseo karolo mo kokoanong ya ngwaga le ngwaga ] 4.3.1 In the last year, have you attended any Annual General Meeting (AGM)? [ A o kile wa tsenelela kokoano ya ngwaga le ngwaga mo ngwageng o o fetileng ]

PAGE 214

214 Yes [ ee ]: No [ nnyaa ]: 4.3.2 Here is a list of the decisions that are made at the CBNRM AGM. At the last AGM, please tell us your level of participation when the following was discussed? [ Tse di latelang ke megopolo e e tsewang kwa bokopanong jwa ngwaga le ngwaga. Bolela boleng jwa karolo ya gago mo bokopanong jwa ngwaga o o fetilang ] Often [ Gangwe le gape ] Several times [ Kgapetsa kgapetsa ] Once or twice [ Gangwe kgotsa ga bedi ] Would if I had the chance [ Ke ka tsaya karolo fa ke filwe sebaka ] Not interested [ Ga ke na kgatlhego ] Know [ Ga ke itse ] CO DE [ Letshwao ] 1 2 3 4 5 9 a. Reading and adoption of minutes for last AGM [ Go bala le go amagela metsotso ya bokopano jo bofetileng ] [ Ga ke itse ] b. Annual Report on status of Natural Resources [ Pego ya ngwaga ka disa tlhalego ] c. Annual Report on Trust Projects [ Pego ya ngwaga ka ditiro tsa lekgotla ] d. Financial Report [ Pego ya tsa madi ]

PAGE 215

215 e Audit Report [ Pego ya tshekatsheko ya tsa madi ] f. Elections of members of the Board [ Tlhopho ya maloko a botsamaisi ] g. Review of membership register. [ Tshekatshelo ya dikwalo tsa loloko ] h. Amendment of Constitution [ Paakanyo ya melao ] 4.4 [ Ke mabaka afe a a ka tswang a dir ile gore o seka wa tsaya karolo mo bokopanong jwa ngwaga le ngwaga? ] Please Tick [ Tshweetswee tshwaa ] CODE Not interested [ Gake na kgatlhego ] 1 I am not consulted [ Ga ke a rerisiwa ] 2 My participation makes no difference [ Boleng jwame gab o dire ph arologanyo ] 3 Only Trust Leaders can participate [ Baeteledipele ba lekgotla ke bone fela ba tsayang karolo ] 4 4.5 Please tell us any other reasons why you have not participated in AGM decisions. [ Ka tsweetswee re bolelele mabaka a mangwe a a dirileng gor e o seka wa tsaya karolo mo bokopanong jwa ngwaga le ngwaga ]

PAGE 216

216 5. RIGHTS [ Ditshwanelo ] 5.1 Name of Community ]. Do you have the following rights: [ Are bue ka ditshwanelo tsa gago jaaka leloka ditshwanelo tse di latelang: ] [Rights here are restricted to rights provided for in the CBNRM constitution] RIGHTS [ Ditshwanelo ] YES [ ee ] NO [ nnyaa ] WHO GIVES RIGHTS [ Kemang yo o rebolang ditshwanelo ] TICK [ Tshwaa ] 5.1.1 Right to sta nd in an election? [ Tshwanelo ya go emela ditlhopho? ] Government [ Goromente ] Trust Leaders [ Baeteledi ba lekgotla ] Traditional Authority [ Babusi ba ngwao/ dikgosi ] A few individuals in the Community [ Palo ee seng ya sepe ya morafe ] Th e Community [ Morafe ka kakaretso ] 5.1.2 Right to be nominated for Government [ Gotomente ] Trust Leaders [ Baeteledi ba

PAGE 217

217 election onto the Board? [ Tshwanelo ya go tlhopiwa go tsena mo lekgotleng la boeteledipele ] lekgotla ] Traditional Authority [ Babusi ba ngwao/ dikgosi ] A few individuals in the Community [ Palo ee seng ya sepe ya morafe ] The Community [ Morafe ka kakaretso ] 5.1.3 Right to vote? [ Tshwanelo ya go tlhopha ] Government [ Goromente ] Trust Leaders [ Baeteledi ba lekgotla ] Tradi tional Authority [ Babusi ba ngwao/ dikgosi ] A few individuals in the Community [ Palo ee seng ya sepe ya morafe ] The Community [ Morafe ka kakaretso ] 5.1.4 Right to remove corrupt CBNRM Leaders? [ Tshwanelo ya go ntsha boeteledipele bo bo sokameng m o lekgotleng (CBNRM) ] Government [ Goromente ] Trust Leaders [ Baeteledi ba lekgotla ] Traditional Authority [ Babusi ba ngwao/ dikgosi ] A few individuals in the Community [ Palo ee seng ya sepe ya morafe ] The Community [ Morafe ka kakaretso ] 5.1.5 Right to choose a Leader of your choice as Trust Representative? [ Tshwanelo ya go tlhopa moeteledipele yo o ratang jaaka moemedi wo gago kwa lekgotlaneng ] Government [ Goromente ] Trust Leaders [ Baeteledi ba lekgotla ] Traditional Authority [ Babusi ba ngwao/ dikgosi ] A few individuals in the Community [ Palo ee seng ya sepe ya morafe ] The Community [ Morafe ka kakaretso ]

PAGE 218

218 5.1.6 Right to participate at AGM? [ Tshwanelo ya go tsaya karolo mo bokopanong jwa ngwagwa le ngwagwa ] Government [ G oromente ] Trust Leaders [ Baeteledi ba lekgotla ] Traditional Authority [ Babusi ba ngwao/ dikgosi ] A few individuals in the Community [ Palo ee seng ya sepe ya morafe ] The Community [ Morafe ka kakaretso ] 5.1.7 Right to make decision on CB NRM finances? [ Tshwanelo ya go tsaya ditshwetso mo go tsa madi a lekgotlana (CBNRM) ] Government [ Goromente ] Trust Leaders [ Baeteledi ba lekgotla ] Traditional Authority [ Babusi ba ngwao/ dikgosi ] A few individuals in the Community [ Palo ee seng ya sepe ya morafe ] The Community [ Morafe ka kakaretso ] 5.1.8 Right to change the constitution? [ Tshwanelo ya go fetola molao motheo ] Government [ Goromente ] Trust Leaders [ Baeteledi ba lekgotla ] Traditional Authority [ Babusi ba ngwao/ dikg osi ] A few individuals in the Community [ Palo ee seng ya sepe ya morafe ] The Community [ Morafe ka kakaretso ] 6. RIGHTS OF POLITICAL LEADERS TO COMPETE FOR SUPPORT & COMPETE FOR VOTES [ Ditshwanelo tsa baeteledipele ba sepolotiki go gaisanela kemo nokeng le ditlhopho ] 6.1 Leadership: [ Boeteledipele ] 6.1.1 In your opinion, how likely is it that you could get together with others and make: [ Read out options ] [ O bona le ka dira eng le ba bangwe go dira tse di latehangi ]

PAGE 219

219 [ Balela megopolo kwa godimo ] Very like ly [ Go ka kgonega thata ] Somewhat likely [ Go ka kgonega ] Neutral [ Go fa gare ] Not very likely [ Go ka se kgonege ] Not at all likely [ Go ka se kgonege gotlholele ] CODE 5 4 3 2 1 a. The elected Trust Chairperson listen to your concerns about a mat ter of importance to the community? [ Modulasetilo yo o tlhophilweng o reetsa dikeletso tsa gago tse di amang morafe? ] b. The Trust Manager listen to your concerns about how money is spent? [ Mogogi wa lekgotla o reetsa dikeletso/ megopolo tsa gago ka tsa tiri so ya modi ] 6.1.2 in your CBNRM community [ name the CBO e.g Khwai Trust ]. Which of the following statement is closest to your view? Choose Statement 1 or Statement 2 [How the community perceives Leadership] [ Tla re tseye sebaka re bue ka komiti ye o batlang go nna le yone mo lekgotleng la CBNRM. Ke mulaetsa ofe o o gaufi le maikutho a gago? Tlhopha molaetsa wa ntlha kana wa bobedi (Morafe o amogela jang boeteledipele ] [ State ment 1 : [ Molaetsa wa ntlha: ] Members of ............[name of CBO] should be more active in questioning the actions of [ nna matlhagatlhaga ka go botsa ka ditiro tsa baeteledi ] Statement 2 : [ Malaetsa wa bo bedi ] In this community, members should show more respect for authority [ Mo motseng o, maloko a tshwanetse go itsaya itshupa maitseo mo bodireding jwa bone. Agree very Agree wit h Agree very Agree with Agree

PAGE 220

220 strongly with Statement 1 [ Ke dumalana thata le mulaetsa wa ntlha ] Statement 1 [ Ke dumalana le molaetsa wa ntlha ] strongly with Statement [ Ke dumalana thata le molaetsa wa bobedi ] 2 Statement [ Ke dumalana le molaetsa ] 2 with Neither 1 or 2 [ Gke dumalana le molaetsa wa ntl ha kgotsa wa bobedi ] Know [ Ga ke itse ] CODE [ Letswao ] 1 2 3 4 5 9 Tick [Tshwaa] 6.1.3 your CBNRM community [ name the CBO e.g Khwai Trust ]. Which of the following state ment is closest to your view? Choose Statement 1 or Statement 2 [ A re tseye sebaka re bue ka morafe o o batlang go nna le one mo lekgotleng la Khwai/ Mababe/ Sankuyo ke ofe molaetsa o o latelang o o gaufi le maikutlo a gago? Tlhopha molaetsa wa ntlha kgot sa wa bobedi. ] Statement 1: [ Molaetsa wa ntlha: ] favour their own family or group Ka baeteledipele ba emetse mongwe le mongwe, gab a a tshwanela go tsaya letlhakore la ba malwapa a ] Statement 2: [ Malaetsa wa bobedi ] [ Baeteledi ba solofelwa go thusa ba malwapa a bone fa ba setse ba tsere marapho ] Agree very strongly with Statement [ Ke dumalana thata le mulaetsa wa ntlha ] 1 Agree with Statement 1 [ Ke dumalana le molaetsa wa ntlha ] Agree very strongly with Statement 2 [ Ke dumalana thata le molaetsa wa bobedi ] Agree with Statement 2 [ Ke dumalana le molaetsa ] Agree with Neither 1 or 2 [ Gke dumalana le molaetsa wa ntlha kgotsa wa bobedi ] Know [ Gakeitse ]

PAGE 221

221 CODE [ Letswao ] 1 2 3 4 5 9 Tick [Tshwaa] 7. ELECTIONS [ Ditlhopho ] 7.1 methods, how was the Tr ust Leader chosen at the last AGM elections? [ A re bue kafa le tlhophang baeteledipele ba lekgotlana ka teng. Tse di latelang ke mekgwa e go tlhophilweng moeteledipele mo kokoanong ya ngwaga le ngwaga mo ngwageng o o fetileng: ] Yes [ ee ] No [ nnyaa ] CODE [ Letshwao] a. Appointed by Headman/ Chief [ A tlhophiwa ke kgosi/ kgosana ] 1 b. Appointed by Wildlife Department [ A tlhophiwa ke ba lephata la diphologolo ] 2 c. Vote by Hands [ Ba tlhopha ka go tsholetsa diatla ] 3 d. Vote by Secret Ballot [ Ba tlhopha ka tlhopho ya sephiri (secret ballot) ] 4 e. Other [Specify] [ Tse dingwe ]_________________________________________ 5 7.2 Please rank the method of voting that you prefer, starting with your highest preference first [1 is highest and 4 lowest] [ Tlhopha mefuta e e di risiwang go tlhopha e o bonang e le botoka, o simolole ka e e kwa godimo pele ]. Rank [1 4] CODE a. Appointed by Headman/ Chief [ A tlhophiwa ke kgosana/ kgosi ] 1 b. Appointed by Wildlife Department [ A tlhophiwa ke ba lephata la diphologolo ] 2 c. Vote by Han ds [ Ba tlhopha ka go tsholetsa seafla ] 3 d. Vote by Secret Ballot [ Ba tlhopha ka tlhopho ya sephiri ] 4 7.3 How many times have you voted in the Trust elections? [ O tlhophile ga kae mo ditlhophong tsa lekgotla? ] Circle CODE Once [ Gangwe ] 1 1

PAGE 222

222 Twice [ Ga be di ] 2 2 Three times [ Gararo ] 3 3 Four times [ Gane ] 4 4 Five times [ Gatlhano ] 5 5 More than 5 times [ Go feta botlhano ] 5 6 7.4 Did you vote for the current CBO Leaders in the last AGM elections? [ A o tsere karolo mo tlhophong ya baeteledipele ba lekgot la (Trust) mo bokopanong jwa ngwage le ngwaga mo ngwageng o o fetileng? ] Yes [ ee ] No [ nnyaa ] Refused to answer [ Ga araba ] Remember [ Ga ke itse/ ga ke gakologelwe ] CODE 1 2 0 9 Chairperson [ Modulasetilo ] Treasurer [ Motsholamadi ] Secretary [ Mokwaledi ] 7.5 Do you think the last Trust Elections were free and fair? [ A o akanya gore ditlhopho tsa bofelo di re di gololesegile ] Very Fair [ Was able to chose my candidate with freedom ] [ Gololsegile thata: ke ne ka tlhopa moemedi ka tshosologo ] Somewhat Fair [ Free to choose my candidate ok ] [ Golo lesegile nayana: ka tshosologo ka tlhopa moemedi sentle ] Not Fair [ Not able to choose my candidate well ] [ Ga ke a kgona go tlhopha moemedi sentle ] Bad [ Only a few people participated ] [ Batho ba se kae batsere karolo ] Not interested in voting [ participate ] [ Ga ke a tsaya karolo] CODE 1 2 3 4 5 Tick [ Tshwaa ] 7.6 As a member of [ Name of CBO ] would you be eligible to stand for a position in the Trust Committee? [ Jaaka le loko la lekgotlana la Trust, a o kgatlhegela go nna le maemo mo komiting? ]

PAGE 223

223 Yes [ ee ] [1] No [ nnyaa ] [2] 7.7 If no, what are the reasons why you would not stand for office? [ Trying to them [ Fa karabo e le nnyaa, fa lebaka la gore ke eng o sa eme mo komiting (O leke go tlhalosa gore a ke bone bas a batleng go tsaya maemo kgotsa ke melao e e sa ba letleng) ] 7.8 In your opinion, WHO should for stand for a positio n in the Trust Committee? [ Ka kakanyo ya gago o bona e le mangy o o tshwanetseng go emela maemo mo komiting ] Please Tick [ Tsweetswee tshwaa ] CODE [ Letshwao ] All Adults above 18 years [ Bagolo botlhe ba ba dingwaga tse di kwa go dimo ga lesome le boferab obedi ] 1 Only CBNRM Members [ Maloko a lekgotlana la CBNRM fela ] 2 All members of [ Khwai ] Community above 18 years [ Maloko a Khwai / Mababe / Sankuyo a a dingwaga tse di kwa godimo g lesome le boferabobedi ] 3 Members who can Read and Write [ Maloko a a k gonang go bala le go kwala ] 4 7.9 If you voted, please provide three reasons why you voted for the Trust Chairperson [ Fa o tlhophile, tsweetswee fa mabaka a le mararo gore ke eng o tlhophile modulasetilo wa komiti ] Instructions: Do not read out response s, circle the code that best fits the responses, ask respondent to rank them in order of importance and capture the rankings in the blank column [ Ditaelo: O seka wa bala dikarabo, tshwaa letshwao le le siametseng karabo, kopa mmotsolotsi go di tlhomaganya ka maemo a tsone le botlhokwa jwa tsone, dikwale ka maemo a tsone fa go sa

PAGE 224

224 kwalwang sepe ] Code [ Letshwao ] Ranking [ Maemo ] a. Not for any particular reason [ E seka mabaka ape ] 1 b. He is a good person [ Ke motho yo o siameng ] 2 c. He offered myself/my family personal assistance [ O nthusitse le ba lelwapalome ] 3 d. He is a member of my ethnic group [ Ke wa letso lame ] 4 e. My family always votes for him [ Balelwapa lame ba motlhopha nako tsotlhe ] 5 f. My friends always vote for him [ Ditsala tsame di motlhopha nako t sotlhe ] 6 g. He improved my community [ O tlhabolotse motse wa rona ] 7 h. He performed well at the CBO meetings [ O dira sentle mo bokopanong jwa lekgotla (Trust) ] 8 i. He listens to our community concerns [ O reetsa dingongorego tsa motse ] 11 j. Other [ Tse dingw e ] 12 k. Not Applicable [ Ga go a tshwanela ] 99 l. Refused to answer [ Ba ganne go araba ] 999 m. Ga ba itse ] 9 7.10 When you voted at the last AGM, what were the three most important things you hope the incoming Committee would do? [ Fa o tlhopha ko bo kopanong jwa ngwaga jo bofetileng, ke dife dilo tse tharo tse di botlhokwa tse o neng o batla komiti e ntsha e di dira ] a. _______________________________________________________________________ b. ______________________________________________________________ _________ c. ___________________________________________________________________________

PAGE 225

225 7.11 there been a change in the members of the Board? [ Are bue ka komiti kgolo ya lekgotla la (CB NRM). Mo dingwageng tse tlhano tse di fetileng, a gona le pharologanyo mo malokong a komiti kgolo ] Yes [ ee ] [1] No [ nnyaa ] [2] [If no, continue to 7.13] [ Fa karabo e le nnyaa tswelela ko go 7.13 ] 7.12 If no, would you want a change of CBO Committee member s? [ Fa karabo e le nnyaa, a o botla pharologanyo mo malokong a Trust ] Yes [ ee ] [1] No [ nnyaa ] [2] 8. CONSTITUTION AND VOTING [ Molao motheo le tlhoppho ] 8.1 Please rank the number of Terms of Office you would want your CBO Committee to serve. [1 5; 1 being highest] [ Tsweetswee rulaganya di nomoro tsa boloko jwa Trust gore e tshwanetse go thusa leng ] [ 1 5; 1 e le e e kwa godimo ] Rank [1 5] [ Maemo ] 5 years [5] [ Ngwaga tse tlhano ] 4 years [4] [ Ngwaga tse nne ] 3 years [3] [ Ngwaga tse tharo ] 2 years [2] [ Ngwaga tse pedi ] 1 year [1] [ Ngwaga e nngwe ] 8.2 Has any member of the Board of Trustees been removed from office by a majority vote at a Special General Meeting? [ A gona le mongwe wa komiti kgolo ya mokgatlho yoo kileng a ntshiwa ka tlhop ho mo phuthegong kgolo ya mokgatlho? ] Yes [ ee ] [1] No [ nnyaa ] [2]

PAGE 226

226 8.3 Does your community follow the Constitution when removing a corrupt CBO Leader from Office? [ A komiti ya lona e s ala molao motheo morago, fa e ntsha leloko e le sa tshepegeng/ senyang mo komiting ] All of the time [ Nako tsotlhe ] Sometimes [ Fa gongwe ] Know [ Gake itse ] Not all the time [ e seng nako tsotlhe ] Never [ Gaba nke ] Tick [Tshwaa] 8.4 Pl ease list some of the qualities you would want in the CBO Committee Leader: [ Tsweetswee kwala ditshetla tsa boleng jo o botlang go bo bona mo mseteledipeleng wa komiti ya lekgotla la Trust? ] 9. ALTERNATIVE SOURCES OF INFORMATION [ Metswedi ya kitso e mengwe ] 9.1 How often do you get the following information? [ O bona kitso ya tse di latelang ga kae? ] Very Often [ Gangwe le ] Sometimes [ Fa gongwe ] Never [ Ganke ] Not Interested [ Ga ke kgatlhege ] 9.1.1 Annual Budget [ Lenaneo la ngwaga le ngwaga la m adi ] 9.1.2 Information on Projects [ Kitso ya ditiro ]

PAGE 227

227 9.1.3 Income from Safari Operator [ Dipoelo ga tswa kwa batsamaisi ba safari ] 9.1.4 Money spent on Staff Salaries [ Madi a a dirisitsweng mo dituelong tsa babereki ] 9.1.5 Information on Trust Leade rship activities [ Kitso go tswa mo ditirong tsa bo eteledipele jwa lekgotla 9.1.6 Campaign material for Trust Elections [ Dilo tsa ipapatso mo ditlhophong tsa lekgotla ] 9.1.7 Who voted in Elections [ Ke mang yo o tlhophileng mo ditlhophong ] 9.2 During the past year, how often have you contacted any of the following persons about concerns on how the Trust is being managed? [ Mo dingwageng tse difitileng, a o kopane le batho ba ba latelang ka dingongorego tsa ka fa lekgotla le tsamaisiwang ka teng? ] Often [ G antsi ] A few times [ Gase gantsi ] Only once [ Gangwe fela ] Never [ Ga ke ise ] [ Ga ke itse ] The Kgosi [ Dikgosi ] Trust Chairperson [ Moeteledipele wa lekgotla ] Trust Manager [ Mookamedi wa lekgotla/ motsamarse ] 9.3 Do you think employi ng a Trust Manager has helped improve things in your community? [ A o akanya gore go thapiwa go mookamedi wo lekgotla go fetotse dilo tse dintsi mo motseng? ] Improved A lot [ Go fetogile Improved things slightly N o change [ Go gona phetogol Things are bad [ Dilo/ ditiro Things are worse [ Dilo di ile

PAGE 228

228 thata ] [ Go fetotse dilo go le gonnye ] pharologanyo ] di ile masweng ] masweng thata ] Tick [ Tshwaa ] 9.4 Overall how satisfied are you with democracy [ people having power to make decisions ] in your C BNRM Community? [ Tshoboko ya tsamaiso (batho bana thata ya go tsaya tshwetso) mo lekgotleng la CBNRM ] Very satisfied [ Ke kgotsofetse thata ] Fairly satisfied [ kgotsofetse sentle ] Not very satisfied [ Ga ke a kgotsofala thata ] Not at all satisfied [ Ga ke a kgotsofala gotlhelele ] Do not know [ Ga ke itse ] Tick [ Tshwaa What tribe do you belong to ? O wa lotso lofe? [1] Bayei [3] Basarwa [5] Other (specify) [8] Don't know [2] Humbukushu [4] Batawana [6]None [9] No response To what chur ch or religious group do you belong? O tsena kereke efe? [1] United Congregatio n Church of South Africa (UCCSA) [2] Zions Christia n Church (ZCC) [3] Morongw a (Twelve Apostle Church) [4] Methodi st [5] Seventh Day Adventi st [6] Delta Cross Ministrie s [7] New Jerusale m [8] No Churc h [98] Don' t kno w [9] Othe r THE END THANK YOU FOR YOU TIME [ Bokhutlo ke lebogela nako ya gago ]

PAGE 229

229 APPENDIX B DEMOCRATIC GOVERNANCE QUESTIONNAIRE ZAMBIA Respondent Number Research Assistant Name QUESTIONNAIRE ON UNDERSTANDING THE NATURE OF DEMOCRATIC INSTITUTIONS IN CBNRM Household Selection Procedure : Interviewer: We select respondents using a simple random sample procedure. Interview includes members that are 18 years and above. INFORMED CONSENT Good day. My name is: ___________________________________. I am a Researcher with the University of Florida and this is an independent research institution. I do not represent the Government or any Non Governmental Organization. S TATEMENT ABOUT INFOR MED CONSENT : 10. The purpose of this study is to understand what you think about your how things are governed in your Community. 11. We would like to learn more on how you vote ; chose your leaders and participate in various political activities in the Trust. 12. This survey should take less than one hour. 13. You do not have to answer any question you do not want to 14. All information is confidential. 15. You can stop the interview process at any time. 16. You can ask for clarification on any question at any time 17. There is no right or wrong an swers, and most importantly candid and honest answers are most useful. 18. There are no direct benefits, risks, or compensation to you for participating in the study. For questions about your rights as a research participant contact the IRB at Tel: 352 392 043

PAGE 230

230 2. BACKGROUND CBO/ CRB TICK CBO VILLAGE/ VAG TICK Kasempa Kamfuwe Mbusunte Shimuka Kamakechi Lunga Mulendema Mapoko Isalama Musapa Katele Kapepe 2.1 Are the following services present in the Village [ Interviewer determines ] Yes No Determine Electricity grid that most houses could access Piped water system that most houses can access Sewage system that most houses could access Cell phone service 2.3 Are the following facilities in the Villag e, within easy walking distance? [ Interviewer determines this ] Yes No Determine School Police Station Health Clinic Market Stalls (selling groceries and/ or clothing) 1.3 Household Check: Infrastructure Traditional mud house Yes No Cement block house Yes No Thatched roof (traditional) Yes No Tin roof Yes No 1.4 Source of Energy Yes No Electricity Gas & Paraffin Wood Charcoal 1.5 Household Assets Yes No

PAGE 231

231 Radio Television Fridge Telephone Bicycle Motor Vehicle Motor cycle Plough Boat/ Canoe Scotch cart Donkeys Chickens Cows 2. DEMOGRAPHICS 2.1 How o ld are you? ............................................. 2.2 How long have you lived in the village? ............................................................................................................................. ..... 2.3 Are you the head of t he household? Yes No 3.5 Which is your home language [ kituduka ]? [ Interviewer: Prompt if necessary: That is the language of your group ] TICK Code English 101 Bemba 209 Lala 210 Bisa 211 Lamba 212 Tonga 213 Lenje 214 Ila 215 Luvale 216 Lunda (N/ West) 217 Kaonde 218 Lozi 219 Chewa 220 Nsenga 221 Ngoni 222 Nyanja 223 Lungu 224 Mambwe 225

PAGE 232

232 Namwanga 226 Tumbuka 227 Senga 228 2.5 What is your tribal group? [ Wi mushoboka ] [ Interviewer: Prompt if necessary: That is the lan guage of your group ] Please tick Code Bemba 229 Lunda (Luapula) 230 Lala Bisa 231 Ushi 232 Chishinga 233 234 Lamba 235 Tonga 236 Lenje 237 Ila 238 Luvale 239 Lunda (N/ West) 240 Mbunda 241 Kaonde 242 Lozi 243 Che wa 244 Nsenga 245 Ngoni 246 Lungu 247 Mambwe 248 Namwanga 249 Tumbuka 250 Senga 251 3.6 Sex Male: Female: 3.7 Education Grade No Education: Some P rimary School [ Lower basic: Grades 1 4 ] Upper Primary School [ Middle basic: Grades 5 7 ]: Junior Secondary [ Grades 8 9 ]: Senior Secondary [ Grades 10 12 ] College [ Diploma holder ]

PAGE 233

233 University [ Degree holder ] 3.8 Gender of Head o f Household Male: Female: Female (male away for 6 + months/ year: 3.9 No. of people in Household: Adults : Male Female Children : Male Female 3.10 No. of people in Household employed in: CRB: .................................... Village Scouts: Hunting [No of people]: .......... Type of Jobs: Other employment [ job ]... .....................................No. of pple................... Other employment [ job ]........................................No. of pple................... Position in CRB CRB Board Member: ........................................................ .................... Employee [Village Scout]...................................................................... VAG Committee Member..................................................................... Ordinary Village Member..................... ................................................. Other: [ specify ]...................................................................................... 3. LIVELIHOOD STATUS AND PROVISION OF CBNRM GOODS 4.1 In general, how do you rate the way you live compar ed to those of other members in this community? Code Tick Much worse 5

PAGE 234

234 Worse 4 Same 3 Better 2 Much Better 1 9 4.2 Over the past year, how often, have you or anyone in your family gone without: [Read out option s] Never Just once or twice Several times Many times Always Know 4.2.1 Food to eat? 4.2.2 Clean water for home use? 4.2.3 Medicines or medical treatment? 4.2.4 Fuel to cook your food? 4.2.5 A cash income? 4.3 Looking back 10years ago [ 1999 say year ], is your community better off now, or before CBNRM started? [ Read out response options & tick ] CODE Tick Much worse 5 Worse 4 Same 3 Better 2 Much Better 1 9 4.4 What would you say are the major changes that have been brought about by CBNRM? [ Interviewer: prompt good and bad changes ] ............................................................................................................................. ............... .............. ........... ............................................................................................................................. .... .............. ................................................................................................................ ............................ .............. ............................................................................................................................. ............... ..............

PAGE 235

235 .......................................................... .................................................................................. .............. ............................................................................................................................. ............... .............. 3.5 Now has anyone in your household received any of the following benefits? Please list the benefits you and your household got from wildlife in the last 12 months: Type of Benefit D o you and your household receive benefit? Amount/Describe Cash Yes / No Meat Yes / No Employment Yes / No Other [ Specify ] ........................................................................... ................................................ ........................... 3.6 PROJECTS: Please name the projects being developed by your CRB and VAG: Name of Project Year Started Is the Project Implemented Well? Very Well Well Neutral Badly Very Badly Write name Year 1. 2. 3. 4. 5. 6. 7. 5. POLITICAL PARTICIPATION IN CBNRM 5.1 Membership:

PAGE 236

236 5.1.1 As a member of this community do you have a right to participate in CRB activities? [Interviewer mention the name of village] Yes: 5.1.2 Here is a list of positions in the CRB community. For each one, could you tell me whether you are: TICK e. CRB Board Member f. VAG Committee Member g. Active CRB Member h. Not an acti ve CRB Member [If respondent ticks (d) ask question 4.2, if not proceed to question 4.3] 5.2 Please list the reasons why you are not an active member of the CRB? .............................................................................................. .............................................. ............................................................................................................................. ............... ...................................................................... ...................................................................... ............................................................................................................................. ............... .............................................. .............................................................................................. ...................................................................... [If respondent, ticks (a c) ask the following set of questions] 5.3 Participation in Annual G eneral Meetings: 5.3.1 In the last year, have you attended any Annual General Meeting (AGM)? Yes: No: 4.3.2 In the last year, have you attended any General Me eting (GM) in your VAG? Yes: No: 4.3.3 Here is a list of the decisions that are made at the CRB AGM. At the last AGM, please tell us your level of particip ation when the following was discussed? Often Several times Once or twice Would if I had the Not interested Know

PAGE 237

2 37 chance a. Reading and adoption of minutes for last AGM b. Annual Report on status of Natu ral Resources c. Annual Report on Community Projects d. Financial Report Often Several times Once or twice Would if I had the chance Not interested Know e Audit Report f. Elections of members of the Board g. Review of membership register. h. Amendment of Constitution 5.4 Please Tick Not interested I am not consulted My participation makes no di fference Only CRB and VAG Leaders can participate 5.5 Please list OTHER reasons for not participating in AGM decisions.

PAGE 238

238 6. RIGHTS 6.1 [Rights here are restricted to rights provided for in the CBNRM constitution] RIGHTS YES NO WHO GIVES RIGHTS TICK 6.1.1 Right to stand in an election? Government CRB Leaders Traditional Authority/ Chief A few individuals in the Community The Co mmunity 6.1.2 Right to be nominated for election onto the Board? Government CRB Leaders Traditional Authority/ Chief A few individuals in the Community The Community 6.1.3 Right to vote? Government CRB Leaders Traditional Authority/ Chief A few individuals in the Community The Community 6.1.4 Right to remove corrupt CRB Leaders? Government CRB Leaders Traditional Authority/ Chief A few individuals in the Community

PAGE 239

239 The Community 6.1.5 Right t o choose a Leader of your choice as CRB Representative? Government CRB Leaders Traditional Authority/ Chief A few individuals in the Community The Community 6.1.6 Right to participate at AGM? Government CRB Leaders T raditional Authority/ Chief A few individuals in the Community The Community 6.1.7 Right to make decision on CRB finances? Government CRB Leaders Traditional Authority/ Chief A few individuals in the Community The Commu nity 6.1.8 Right to change the constitution? Government CRB Leaders Traditional Authority/ Chief A few individuals in the Community The Community 7. RIGHTS OF POLITICAL LEADERS TO COMPETE FOR SUPPORT & COMPETE FOR VOTES 7.1 Leader ship: 7.1.1 In your opinion, how likely is it that you could get together with others and make: [ Read out options ] Very likely Somewhat likely Neutral Not very likely Not at all likely c. The elected CRB Chairperson l isten to your concerns about a matter of importance to the community?

PAGE 240

240 d. The VAG leaders listen to your concerns about how money is spent? 7.1.2 your CRB comm unity. Which of the following statement is closest to your view? Choose EITHER Statement 1 or Statement 2 Statement 1: Members of CRB community should be more active in questioning the Statement 2: In this community, me mbers should show more respect for CRB leaders Agree very strongly with Statement 1 Agree with Statement 1 Agree very strongly with Statement 2 Agree with Statement 2 Agree with Neither 1 or 2 Know Circle 7.1.3 Le CRB community. Which of the following statement is closest to your view? Choose EITHER Statement 1 or Statement 2 Statement 1: n ot favor their own family or group Statement 2: expected to help their family and Agree very strongly with Statement 1 Agree with Statement 1 Agree very strongly with Statement 2 Agree with Statement 2 Agree with Neither 1 or 2 Know CIRCLE

PAGE 241

241 8. ELECTIONS 7.1 was the CRB Leader chosen at the last elections? Yes No f. Appointed by Headman/ Chief g. Appointed by ZAWA h. Vote by Hands i. Vote by Secret Ballot j. Other [Specify] ________________________________________ 8.2 Please rank the method of voting that you prefer, starting with your highest preference first [1 is highest and 4 lowest] Rank [1 4] e. Appointed by Headman/ Chief f. Appointed by ZAWA g. Vote by Hands h. Vote by Secret Ballot 8.3 How many times have you voted in the CRB/VAG elections? Circle CODE Once 1 1 Twice 2 2 Three times 3 3 Four times 4 4 Five times 5 5 More than 5 times 5 6 8.4 Did you vote in the last CRB/VAG elections? TICK Yes No Refused to answer 8.5 Do you think the last elections were free and fair? Very Fair Somewhat Fair No t Fair Very Unfair Not interested in voting

PAGE 242

242 CIRCLE 8.6 As a member of this VAG would you be eligible to stand for a position in the CRB Committee? TICK Yes No Not interested 8.7 If no, what are the reasons why you wou ld not stand for office? 8.8 In your opinion, WHO should for stand for a position in the Committee? TICK All adults above 18 years in this village. Only members that have been born in this village. Only members who can Read and Write. 8.9 If yo u voted, please provide three reasons why you voted for the CRB Committee at the VAG elections? TICK RANK n. Not for any particular reason. o. They are good people. p. They offered myself/my family personal assistance q. They are members of my tribal group r. My family always votes for them. s. My friends always vote for them. t. They improved my community.

PAGE 243

243 8.11 members of the Boar d? Yes No [If no, continue to 7.13] 8.12 If no, would you want a change of CBO Committee members? Yes No 9. CONSTITUTION AND VOTING Has your constitution been explained to you in the last 12 months? YES NO Were you consulted during the constitution building process? YES NO Do you think your constitution organizes the community well? The constitution works very well u. They performed well at the CRB meetings v. They listen to our community concerns. w. They are related to the Chief. x. They are related t o the Village Headman. y. Refused to answer z. 8.10 When you voted at the last AGM, what were the three most important things you hope the incoming Committee would do? d. _______________________________________________________________________ e. __ _____________________________________________________________________ f. ___________________________________________________________________________

PAGE 244

244 The constitution works reasonably wel l (just ok) Neutral The constitution is bad The constitution is very bad. ??? 9.1 Please rank the number of Terms of Office you would want your CRB Committee to serve. [1 5; 1 being highest] Rank [1 5] 5 years [5] 4 years [4] 3 years [3] 2 years [2] 1 year [1] 9.2 Has any member of the CRB been removed from office by a majority vote at a Special General Meeting? Yes No 9.3 Does your community follow the Constitution when removing a corrupt CRB Leader from Office? All of the time Sometimes know Not all the time Never CIRCLE 9.4 Please list some of the qualities you would want in the CBO Committee Leader: 10.4 ATTITUDES ABOUT WILDLIFE 9.1 Overall (taking into account positives and negatives) my attitude towards wildlife is:

PAGE 245

245 CIRCLE Strongly Positive Positive Neutral I do not support Wildlife I strongly dislike Wildlife ?? Not sure 9.2 What do you think of the Kafue National Park? CIRCLE Strongly Support the Park Support the Park Neutral / Not sure I do not support the Park I strongly d i slike the Park 9.3 What do you think of the ZAWA Staff? CIRCLE I like them a lot I like them Neutral / Not sure I do not like them I strongly dislike them 9. Why do you like wildlife? For each reason, indicate how important this is for you. Very Important Important Neutral Not Ver y important Not at All Important REASON FOR LIKING WILDLIFE YES NO Conservation for non financial reasons Household benefits Jobs Development projects / Community income Brings development (i.e. economic growt h) Hunting / Meat Helps us get better organized/empowered Others reasons

PAGE 246

246 10. ALTERNATIVE SOURCES OF INFORMATION 10.1 How often do you get the following information? Very Often Sometimes Never Not I nterested 10.1.1 Annual Budget 10.1.2 Information on Projects 10.1.3 Income from Safari Operator 10.1.4 Money spent on Staff Salaries 10.1.5 Information on CRB Leadership activities 10.1.6 Campaign material for CRB Elections 10.1.7 Who voted in Elections 10.2 During the past year, how often have you contacted any of the following persons about concerns on how the CRB is being managed? Often A few times Only once Never The Chief The Village Headman CRB Leaders 10.3 Do you t hink employing a Unit Leader has helped improve things in your CRB? Improved A lot Improved things slightly No change Things are bad Things are worse

PAGE 247

247 CIRCLE 10.4 Overall how satisfied are you with democracy [ people having pow er to make decisions ] in your CRB Community? Very satisfied Fairly satisfied Not very satisfied Not at all satisfied Do not know Tick THE END THANK YOU FOR YOU TIME

PAGE 248

248 LIST OF REFERENCES Acheson, J. M. (2006). Institutional Failure in Resource Management. Annual Review of Anthropology 35 : 117 134. Adams, W. and D. Hulme (2001). Community conservation: changing narratives, policies & practices in african conservation. African wildlife and livelihoods: the prom ise and performance of community conservation D. Hulme and M. W. Murphree. Capetown, James Currey. Adcock, R. and D. Collier (2001). Measurement validity: A shared standard for qualitative and quantitative research. American Political Science Review 95 (3 ): 529 546. Adhikari, B. (2003). Property rights and natural resources: Socioecnomic heterogeneity and distributional implications of common property resource management. South Asian Network for Development and Environmental Economics Adhikari, B. (20 05). Poverty, property rights and collective action: understanding the distributive aspects of common property resource management. Environment and Development Economics 10 : 7 31. Agarwal, B. (2001). Participatory exclusions, community forestry and gend er: An analysis for South Asia and a conceptual framework. World Developmen t 29 (10): 1623 1648. Agrawal, A. and C. C. Gibson (1999). Enchantment and disenchament: The Role of community in natural resource conservation. World Development 27 (4): 629 649 Agrawal, A. and J. C. Ribot (1999). Accountability in Decentralization: A Framework with South Asia and African cases. Journal of Developing Areas 33 : 473 502. Agrawal, A. (2001). Common Property Institutions & Sustainable Governance of Resources. Wo rld Development 29 (10): 1649 1672. Agrawal, A. and E. Ostrom (2001). Collective Action, Property Rights, and Decentralization in Resource Use in India and Nepal. Politics & Society 29 (4). Agrawal, A. and S. Goyal (2001). Group Size and Collective Actio n: Third party Monitoring in Common pool Resources. Comparative Political Studies 34 (63). Agrawal, A. (2003). Sustainable Governance of Common Pool Resources: Context, Methods and Politics. Annual Review Anthropology 32 : 243 262.

PAGE 249

249 Ake, C. (1993). The unique case of African democracy. International Affairs 69 (239 244). Almond, G. A. and S. Verba (1963). The civic culture: Political attitudes and democracy in five nations. Princeton, NJ, Princeton University Press. Altman, D. and A. Perez Linan (2002 ). Assessing the quality of democracy: Freedom, competitiveness and participation in eighteen Latin American countries. Democratization 9 (2): 85 100. Anstey, S. (2009). Beacon and barometer: CBNRM & evolution in local democracy in Southern Africa. Beyon d propriertorship: Murphree's laws on community based natural resource management B. B. Mukamuri, J.M. Manjengwa & S. Anstey. Harare, Weaver Press : 41 57. Antinori, C. and D. B. Bray (2005). Community forest enterprises as enterpreneurial firms: econom ic and institutional perspectives from Mexico. World Development 33 (9): 1529 1543. Armitage, D. (2005). Adaptive capacity and community based natural resource management. Environmental Management 35 (6): 703 715. Arnold, J. E. M. (2001). 25 years of C ommunity Forestry Rome, Food and Agricultural Organization of the United Nations. Arnstein, S. R. (1969). A Ladder of Citizen Participation. American Institute of Planners Journal 35 (4): 216 224. Arntzen, J. (2003). An economic view on wildlife manageme nt areas in Botswana. CBNRM Support Programme Occassional Paper No. 10. IUCN/ SNV CBNRM Support Programme, Gaborone, Botswana. Ascher, W. (2007). Issues and best practices in the decentralization of natural resource control in developing countries Decent ralizing governance: Emerging concepts and practices. G. Shabir Cheema and D. A. Rondinelli. Washington DC, Brookings Institution Press. Ashley, C. (1998). Intangibles matter: non financial dividends of community based natural resource management Windho ek, Namibia, WW LIFE Programme. Baker, J. E. (1997). Trophy hunting as a sustainable use of wildlife resources in southern and eastern Africa. Journal of Sustainable Tourism 5 : 306 321. Baland, J. and J. Platteau (1996). Halting degradation of natural resources: Is there a role for rural communities? Rome Italy, Food and Agriculture Organization, United Nations.

PAGE 250

250 Bardham, P. and D. Mookherjee (2007). Decentralization and local governance in developing countries: A comparative perspective. Decentralizing governance: Emerging concepts and practices. G. Shabir Cheema and D. A. Rondinelli. Washington DC, Brookings Institution Press. Barnes, J. I. (1994). Suggested criteria for shadow pricing in cost benefit analysis of projects in Namibia Windhoek, Namibia Directorate of Environment and Tourism. Barnes, J. I. (1995c). Economic analysis of community based wildlife utilization initiatives in Botswana. Development Southern Africa 12 (6): 783 803. Barnes, J. I. (2001). Economic returns and allocation of res ources in the wildlife sector of Botswana. Southern African Journal of Wildlife Research 31 (3 & 4): 141 153. Barnes, J. I., J. MacGregor and L. C. Weaver (2002). Economic efficiency and incentives for change within Namibia's community wildlife use initi atives. World Development 30 (4): 667 681. Barrett, C. B. and P. Arcese (1995). Are integrated conservation and development projects (ICDPs) sustainable? On the conservation of large mammals in sub Saharan Africa. World Development 23 : 1073 1084. Barr ett, C. B., K. Brandon, C. Gibson and H. Gjertsen (2001). Conserving tropical biodiversity amid weak institutions. Bioscience 51 (6): 497 502. Barrow, E. and M. W. Murphree (2001). Community conservation: changing narratives, policies & practices. Africa n wildlife and livelihoods: the promise and performance of community conservation D. Hulme and M. W. Murphree. Capetown, James Currey. Bernard, R. H. (2002). Research methods in antrhopology: Qualitative and quantitative approaches Walnut Creek, CA, Atl aMira Press. Blaike, P. (2006). Is small really beautiful? Community based natural resource management in Malawi and Botswana. World Development 34 (11). Blair, H. (2000). Participation and accountability at the periphery: Democratic local governance in s ix countries. World Development 28 (1): 21 39. Blaise, A. (2000). To vote or not to vote? The merits and limits of rational choice theory Pittsburgh, University of Pittsburgh Press. Bohman, J. (1998). The coming age of deliberative democracy. Journal o f Political Philosophy 6 : 399 423.

PAGE 251

251 Bond, I. (2001). CAMPFIRE and the Incentives for Institutional Change. African Wildlife & Livelihoods. The Promise and Performance of Community Conservation D. Hulme and M. Murphree. Oxford, James Currey : 227 243. B ond, I., L. Goredema, R. Taylor and S. Vermeulen (2007). Empowering rural communities to manage wildlife: lessons learned from WWF SARPO's Support to CAMPFIRE Project 1993 2002 Harare : 33. Botchway, K. (2001). Paradox of empowerment: Reflections on a cas e study from Northern Ghana. World Development 29 (1): 135 153. Botswana CSO (2011). 2011 Population and housing census Gaberone, Botswana, Ministry of Finance and Development Planning, Government of Botswana. Bratton, M. and N. Van de Walle (1997). De mocratic experiments in Africa: Regimes transitions in comparative perspectives Cambridge, UK, Cambridge University Press. Bryan, F. M. (2004). Real democracy: The New England town meeting and how it works. Chicago, IL, University of Chicago Press. Cahy at, A., C. Gonner and M. Haug (2007). Assessing household poverty and wellbeing: A manualwith examples from Kutai Barat, Indonesia Bogor, Indonesia, Center for International Forestry Research. CAMPFIRE Association (2007). CAMPFIRE Association 2006 2007 A nnual report : 18. Cassidy, L. (2000). CBNRM and legal rights to resources in Botswana. CBNRM Support Programme Occassional Paper No. 4. Gaborone, Botswana, IUCN/ SNV CBNRM Botswana. Charnley, S. and M. R. Poe (2007). Community Forestry in Theory and Pra ctice: Where Are We Now? Annual Review of Anthropology 36 (1): 301 336. Charnley, S. and M. R. Poe (2007). Community Forestry in Theory and Practice: Where AreWe Now? Annual Review of Anthropology 36 (301 336). Chase, M. (2011). Dry Season Fixed Wing Aeria l Survey of Elephants and Wildlife in Northern Botswana. Septermber November 2010, Elephants without Borders : 152. Cheema, G. S. (2007). Devolution with accountability: Learning from good practices. Decentralizing governance: Emerging concepts and practi ces. G. S. Cheema and D. A. Rondinelli. Washington DC, Brookings Institute Press. Cheema, G. S. and D. A. Rondinelli (2007). Decentralizing governance: Emerging concepts and practices Washington DC, Havard University and Brookington Institute Press.

PAGE 252

252 Chil d, B. and J. Peterson (1991). CAMPFIRE in Rural development. The Beitbridge Experience, Branch of Terrestrial Ecology Department of National Parks and Wildlife Management & Centre for Applied Social Studies, University of Zimbabwe : 89. Child, B. (1996). The practice and principles of community based wildlife management in Zimbabwe: the CAMPFIRE programme. Biodiversity and Conservation 5 : 369 398. Child, B., B. Jones, D. Mazambani, A. Mlalazi and H. Moinuddin (2003). CAMPFIRE. Communal Areas Management P rogramme for Indigenous Resources. Final Evaluation Report: Zimbabwe Natural Resources Management Program USAID/Zimbabwe Strategic Objective No. 1. Harare, USAID. Child, B. (2004). Parks in transition: Biodiversity, rural development and the bottom line UK, Earthscan. Child, B. and B. Dalal Clayton (2004). Transforming approaches to CBNRM: learning from the Luangwa experience in Zambia. Getting biodiversity projects to work. Towards more effective conservation and development T. O. McShane and M. Well s. New York, NY, Columbia University Press. Child, B. (2005). Principles, practices and results of CBNRM in Southern Africa SASUGG Child, B. (2006). Revenue Distribution for empowerment and democratization. Participatory Learning and Action 55 : 20 29. Child, B. and C. Weaver (2006). Marketing hunting and tourism joint ventures in community areas. Participatory Learning and Action 55 : 37 44. Child, B. and G. Barnes (2010). The conceptual evolution and practie of community based natural resource management in southern Africa: past, present and future. Environmental Conservation 37 (3): 283 295. Child, B., P. Mupeta, S. Muyengwa and R. Lubilo (In review). Community Based Natural Resource Management: Micro Governance and Face to Face Participatory Democracy Chopra, K., G. K. Kadekodi and M. N. Murty (1990). Participatory development: People and common property resource. New Delhi, Sage Publishers. Collomb, J. G. E., P. Mupeta, G. Barnes and B. Child (2010). Integrating governance and socioeconomic indic ators to assess the performance of community based natural resources managment in Caprivi (Namibia). Environmental Conservation 37 (3): 1 7. Coppedge, M. (2005). Thickening Thin Concepts: Issues in Large N Data Generation." In ed., Oxford, Oxford Uni versity Press.

PAGE 253

253 Crook, R. C. and J. Manor (1998). Democracy and decentralization in South East Asia and West Africa: Participation, Accountability and Performance. Cambridge, Cambridge University Press. Crook, R. C. and A. C. Sverrisson (2001). Decentraliz ation and poverty alleviation in developing countries: A comparative analysis, or is West Bengal unique? Working Paper No. 130. Brighton, UK, Institute of Development Studies. Dahl, R. (1961). Who Governs? New Haven, CT, Yale University Press. Dahl, R. ( 1971). Democracy and its Critics New Haven, Connecticut, Yale University Press. Dahl, R. (1989). Democracy and It's Critics New Haven, Yale University Press. Dalal Clayton, B. and B. Child (2003). Lessons from Luangwa. The story of the Luangwa Integra ted Resource Development Project, Zambia London, International Institute for Environment and Development. DEA (2008). Okavango Delta Management Plan Department of Environmental Affairs, Government of Botswana. Deitz, T., N. Dolsak, E. Ostrom and P. C. Stern (2002). The drama of the commons Washington DC, National Academy Press. Deitz, T., E. Ostrom and P. C. Stern (2003). The struggle to govern the commons. Science 302 (5652): 1907 1912. DeMotts, R. and P. Hoon (2012). Whos elephants? Conserving, co mpensating and competing in Northern Botswana. Society and Natural Resources 0 : 1 15. Demsetz, H. (1967). Towards a theory of property rights. American Economic Review 57 (2): 347 359. DeVellis, R. (2003). Scale development: theory and applications. T housand Oaks, CA, Sage Publications. Dressler, W., B. Buscher, M. Schoon, D. Brookington, T. Hayes, C. A. Kull, J. McCarthy and K. Shrestha (2010). From hope to crisis and back again? A critical history of the global CBNRM narrative. Environmental Conserv ation 37 (1): 5 15. Drez, J. and A. Sen (1995). India: Economic development and social opportunity Delhi, Oxford University Press.

PAGE 254

254 Emerton, L. (2001). Nature of benefits and the benefits of nature: why wildlife conservation has not economically benefi tted communities in Africa. African wildlife and livelihoods: the promise and performance of community conservation D. Hulme and M. W. Murphree. Capetown, James Currey : 208 226. Feeny, D., F. Berkes, B. J. McCay and J. M. Acheson (1990). The tragedy of the commons: twenty two years later. Human Ecology 18 (1): 1 19. Filer, J. E., L. W. Kenny and R. B. Morton (1993). Redistribution, income and voting. American Journal of Political Science 37 : 63 87. Fritzen, S. (2007). Can the Design of Community Dr iven Development Reduce the Risk of Elite Capture? Evidence from Indonesia. World Development 35 (8): 1359 1375. Furubotn, E. and S. Pejovich (1974). The economics of property rights. Cambridge, MA, Ballinger. George, A. L. and A. Bennett (2005). Case stu dies and theories development in the social sciences Cambridge, MA, MIT Press. Ghimire, K. B. and M. P. Pimbert (1997). Social change and conservation: environmental politics and impacts of national parks and protected areas. WWF UK, Earthscan. Gibson, C. and S. A. Marks (1995). Transforming rural hunters into conservationists: an assessment of community based wildlife management programs in Africa. World Development 23 (6): 941 957. Gibson, C. (1999). Politicians and poachers: The political economy of wildlife policy. Cambridge, MIT Press. Goredema, L., I. Bond and R. Taylor (2006). Building capacity for local level management through participatory technology development. Participatory Learning and Action 55 : 30 36. Grindle, M. (2007). Going Local: D ecentralization and the Promise of Good Governance Princeton N.J, Princeton University Press. Grooves, R. M., F. J. Fowler Jnr, M. P. Couper, J. M. Lepkowski, E. Singer and R. Tourangeau (2004). Survey Methodology Hoboken, New Jersey, John Wiley & Sons, Inc. GRZ (1998). Wildlife Act 12 Z. W. Authority. Lusaka, Zambia.

PAGE 255

255 Hicks, U. K., & and L. U. K. W. Hicks (1961). Development from below: local government and finance in developing countries of the Commonwealth Oxford University Press. Hulme, D. and M. W. Murphree (2001). African wildlife and livelihoods: the promise and performance of community conservation Cape town, James Currey. Hutton, J., W. M. Adams and J. C. Murombedzi (2005). Back the barriers? Changing narratives in biodiversity conservation Forum for Development Studies 32 (2): 341 370. Hyden, G. (2005). African politics in comparative perspective. Cambridge, UK, Cambridge University Press. Hyden, G., J. Court and K. Mease (2005). Making sense of governance: Empirical evidence from sixte en developing countries. Boulder, CO, Lynne Rienner. Infield, M. and A. Namara (2001). Community attitudes and behaviour towards conservation: An assessment of a community conservation programme around Lake Mburo National Park, Uganda. Oryx 35 : 48 60. Jansen, D. J. (1990). Sustainable wildlife utilization in the Zambezi valley of Zimbabwe: economic, ecological and political tradeoff. Project Paper No. 10 Harare Zimbabwe, WWF Multispecies Animal Production Systems Project. Jenkins, C. N. and L. Joppa ( 2009). Expansion of the global terrestrial protected area system. Biological Conservation 142 Johnson, C. (2001). Local democracy, democratic decentralization and rural development: Theories, challenges and options for policy. Development Policy Review 1 9 (4): 521 532. Jones, B. and M. W. Murphree (2004). Community based natural resource management as conservation mechanism: lessons and directions. Parks in transition: Biodiversity, rural development and the bottom line. B. Child. London, UK, Earthscan : 63 104. Jones, B. (2007). Synthesis of the CBNRM policy and legislation in Botswana, Malawi, Mozambique, Namibia, Zambia and Zimbabwe. WWF SARPO Occassional Paper Series Harare, Zimbabwe, WWF Norway. Kaminski, M. (2002). Do parties benefit from elect oral manipulation? Electoral laws and heresthetics in Poland, 1989 1993. Journal of Theoretical Politics 14 (3): 325 358. Kearl, B. (1976). Field datacollection in the social sciences: Experiences in Africa and the Middle East New York, Agricultural D evelopment Council.

PAGE 256

256 Khwai Trust (2007). Khwai Deed of Trus t Khwai. Larson, A. M. (2004). Formal decentralization and the imperative of decentralization 'from below': A case study of natural resource management in Nicaragua. Decentralization of natural r esources: Experiences in Africa, Asia and Latin America J. C. Ribot and A. M. Larson. London, Frank Cass. Larson, A. M. and F. Soto (2008). Decentralization of Natural Resource Governance Regimes. Annual Review of Environment and Resources 33 (1): 213 239 Lewis, D. and P. Alpert (1997). Trophy hunting and wildlife conservation in Zambia. Conservation Biology 11 : 59 68. Lijphart, A. (1997). Unequal participation: democracy's unresolved dilemma: Presidential address. American Political Science Associati on 91 1 14. Lindberg, S. I. (2006). Democracy and elections in Africa Baltimore, Maryland, The John Hopkins University Press. Lindsey, P. A., P. A. Roulet and S. S. Romanach (2007). Economic and conservation significance of the trophy hunting industry in Sub Saharan Africa. Biological Conservation 134 : 455 469. Lubilo, R. and B. Child (2010). The Rise and Fall of Community Based natural Resource Management in Zambia's Luangwa Valley: An Illustration of Micro and Macro Governance Issues. Community R ights, Conservation & Contested Land F. Nelson. London, Earthscan : 202 226. MacKenzie, A. F. D. (1988). The empire of nature: hunting, conservation and British imperialism. Manchester UK, Manchester University Press. MacKinnon, J., K. MacKinnon, G. Chil d and J. Thorsell (1986). Managing Protected Areas in the Tropics Gland, Switzerland, International Union for the Conservation of Nature and Natural Resources. Maddick, H. (1963). Democracy, Decentralization and Development Bombay, Asia Publishing House Madison, J. (1787). Papers of James Madison IX : 370 384. Manor, J. (1999). The political economy of democratic decentralization Washington DC, World Bank. Manor, J. (2004). User committees: A potentially damaging second wave of decentralization? D ecentralization of natural resources: Experiences in Africa, Asia and Latin America. J. C. Ribot and A. M. Larson. London, Frank Cass.

PAGE 257

257 Martin, R. B. (1986). Communal Areas Management Programme for indigenous Resoruces (CAMPFIRE) Harare, Branch of Terrestr ial Ecology, Department of National Parks and Wildlife Management, Zimbabwe : 135. Mawhood, P. (1983). Local government in the third world: the experience of Tropical Africa Chichester, John Wiley & Sons Ltd. McCay and J. M. Acheson (1987). The question of the commons: the cuture and ecology of communal resources Tucson, Arizona, University of Arizona Press. McKean, M. A. (1992). Success on the commons: A comparative examination of institutions for common property management. Journal of Theoretical Poli tics 4 (3): 247 281. McShane, t. O. and M. Wells, P. (2004). Getting Biodiversity Projects to Work. Towards more Effective Conservation and Development New York, Columbia University Press. Miller, T., K. Holmes, A. Kim, D. Markheim, J. Roberts and C. Walsh (2010). 2010 Index of Economic Freedom. Washington DC, The Heritage Foundation & The Wall Street Journal. Moe, T. M. (2005). Power and political institutions. Perspectives on Politics 3 : 215 233. Mohan, G. K., Stokke (2000). Participatory develop ment and empowerment: the dangers of localism. Third World Quarterly 21 (2): 247 268. Molinas, J. (1998). The impact of inequality, gender, external assistance and social capital on local level cooperation. World Development 26 (3): 413 431. Moore, B. and J. Putzel (1999). Politics and poverty: A background paper for the World Development Report 2000/ 2001. Institute of Development Studies at the Univesity of Sussex. U. Brighton. Mueller, D. C. and T. Stratmann (2003). The economic effects of democrati c participation. Journal of Public Economics 87 2129 2255. Murombedzi, J. C. (2001). Committees, rights, costs and benefits: Natural resources stewardship & community benefits in Zimbabwe's CAMPFIRE programme. African wildlife and livelihoods: the promi se and performance of community conservation. D. Hulme and M. W. Murphree. Cape town, James Currey : 244 255. Murphree, M. W. (1991). Communities as resource management institutions. Gatekeepers series. United Kingdom, International Institute for Environ ment and Development.

PAGE 258

258 Murphree, M. W. (1995). Optimal principles and pragmatic strategies: creating an enabling politico legal environment for community based natural resource management. National Resources Management Programme, SADCC Technical Coordinati on Unit Chobe, Botswana, USAID NRMP Regional. Murphree, M. W. (1997). Community conservation research in Africa: Principles and comparative practice Working Paper No. 2. Manchester United Kingdom, Institute for Development Policy and Management. Murphre e, M. W. (2000). Boundaries and borders. International Association for the study of common property Bloomington USA. Murphree, M. W. (2004). Communal approaches to natural resource management in Africa: From Whence to Where? Journal of International Wild life Law and Policy 7 : 203 216. Murphree, M. W. (2005). Congruent objectives, competing interests, and strategic compromise: concept and processes in the evolution of Zimbabwe's CAMPFIRE, 1984 1986. Communities and conservation: Histories and politics of community based natural resource management J. P. Brosius, A. L. Tsing and C. Zerner. Walnut Creek, CA, AltaMira Press : 105 147. NACSO (2008). Namibia's commmunal conservancies. A review of progress and challenges in 2007 Windhoek, Nambian Associ ation of CBNRM Support Organizations : 120. Narayan, D. (1996). The contribution of people's participation: Evidence from 21 rural water supply projects. Occassional Paper Series No. 1 Environmentally Sustainable Development Washington DC, World Bank. Ne lson, F. (2007). Emergent or illusory? Community wildlife management in Tanzania. IIED & Pastoral Civil Society in East Africa Issue Paper No. 146 : 26p. Nelson, F. and A. Agrawal (2008). Patronage or participation? Community based natural resource manag ement reform in Sub Saharan Africa. Development & Change 39 (4): 557 585. Neumann, R. P. (1998). Imposing wilderness: Struggles over livelihood and nature preservation in Africa. Berkeley, University of California Press. North, D. (1990). Institutions, institutional change and economic performance. Cambridge; New York, Cambridge University Press. Noss, R. F. (1991). Sustainability and wilderness. Conservation Biology 5 : 120 122.

PAGE 259

259 O'Donnell, G. (1998). Horizontal Accountability in New Democracies. Jour nal of Democracy 9 (3): 112 126. Oates, J. F. (1995). The dangers of conservation by rural development: a case study from the forests of Nigeria. Oryx 29 : 115 122. Olson, M. (1971). The logic of collective action: Public goods and the theory of groups. Cambridge, MA, Havard University Press. Olson, M. (1996). The collective choice approach to economic policy and institutions. Institutions and economic development. C. Clague. Ann Arbor, University of Michigan Press. Olson, M. (1998). Group size and gro up behavior. Managing the commons J. Baden and D. S. Noonan. Bloomington, Indiana, Indiana University Press. Ostrom, E. (1990). Governing the commons : the evolution of institutions for collective action Cambridge ; New York, Cambridge University Press. Ostrom, E. and E. Schlager (1992). Property rights regimes and natural resources: a conceptual analysis. Land Economics 68 (3): 249 262. Ostrom, E., R. Gardner and J. Walker (1994). Rules, games and common pool resources Ann Arbor, University of Michi gan Press. Ostrom, E. (1999) a Coping with Tragedies of the Commons. Annual Review Political Scientific 2 : 493 535. Ostrom, E. (1999) b Revisting the Commons: Local Lessons, Global Challenges. Science 284: (278). Ostrom, E. (2007). A diagnostic approac h for going beyond panaceas. The National Academy of Science 104 (39): 15181 15187. Ostrom, E. (2009). A general framework for analyzing sustainability of social ecological systems. Science 325 : 419 422. Overdevest, C. (2000). Participatory Democracy, Representative Democracy, and the Nature of Diffuse and Concentrated Interests: A Case Study of Public Involvement on a National Forest District. Society and Natural Resources 13 : 685 696. Pallant, J. (2007). SPSS survival manual: A step by step guide to data analysis using SPSS for Windows (Version 10) Buckingham, PA, Open University Press.

PAGE 260

260 Pimbert, M. P. and J. N. Pretty (1995). Parks, People and Professionals: Putting `Participation' into Protected Area Management. Discussion Paper No 57, February 1 995. Geneva, U nited N ations R esearch I nstitute F or S ocial D evelopment I nternational I nstitute F or E nvironment A nd D evelopment W orld W ide F und F or N ature Poffenberger, M. (2006). People in the forest: community forestry experiences from Southeast Asia. International Journal of Environment and Sustainable Development 5 (1): 57 69. Pomeroy, R. S. (1995). Community based and co management institutions for sustainable coastal fisheries management in Southeast Asia. Ocean & Coastal Management 27 (3): 143 162. Poteete, A. (2007). How national political competition affects natural resource policy: The case of Community based natural resource manaegment in Botswana. Annual Conference of the Canadian Political Science Association Saskatoon, Canada. Poteete, A. R. (2009). Defining political community and rights to natural resources in Botswana. Development & Change 40 (2): 281 305. Powell, B. G. J. (2000). Elections as Instruments of Democracy New Haven, Yale University Press. Pretty, J. N. (1995). Participato ry learning for sustainable agriculture. World Development 23 (8): 1247 1263. Przeworski, A., M. Alvarez, J. A. Cheibub and F. Limongi (1996). What makes democracies endure? Journal of Democracy 7 (1): 39 55. Putnam, R. (1993). Making Democracy Work: C ivic Traditions in Modern Italy. Princeton, Princeton University Press. Ribot, J. C. (2002). Democratic decentralization of natural resources: Institutionalizing popular participation Washington DC, World Resource Institute. Ribot, J. C. (2003). Democra tic decentralisation of natural resources: institutional choice and discretionary power transfers in Sub Saharan Africa. Public Administration and Development 23 (1): 53 65. Ribot, J. C. (2004). Waiting for democracy: The politics of choice in natural reso urces decentralization Washington DC, World Resources Institute : 7 45. Ribot, J. C. and A. M. Larson (2004). Decentralization of natural resources: Experiences in Africa, Asia and Latin America. London, Frank Cass.

PAGE 261

261 Ribot, J. C. and A. Agrawal (2006). Recentralizing while decentralizing: how national governments reappropriate forest resources. World Development 34 (11): 1864 1886. Ribot, J. C. (2008). Building Local Democracy through Natural Resource Interventions. An Environmentalist's Responsibili ty World Resources Institute. Rihoy, E. C. and B. Maguranyanga (2007). Devolution and democratization of natural resource management in Southern Africa: A comparative analysis of CBNRM policy processes in Botswana and Zimbabwe. CASS PLAAS CBNRM Occassion al Paper No. 18. Cape Town, University of Western Cape. Riker, W. and P. Ordeshook (1968). A theory of the calculus of voting. American Political Science Review 62 : 25 42. Rondinelli, D. A. (1981). Government decentralization in a comparative perspecti ve: Theory and practice in developing countries. International Review of Administrative Sciences 47 : 136 145. Rondinelli, D. A., J. S. McCullough and R. W. Johnson (1989). Analysing decentralization politics in developing countries: A political economy framework. Development & Change 20 : 57 87. Rozemeijer, N. (2003). CBNRM in Botswana 1989 2002: A select and annotated bibliography and other stories. Occassional Paper No. 11. Gaborone, Botswana, IUCN/ SNV CBNRM Support Programme. Sankuyo Trust (1995 ). Sankuyo Deed of Trus t Sankuyo. Sartori, G. (1962). Democratic Theory. Based On The Author's Translation Of Democratic E Definizion 2nd Ed Detroit, Wayne State University Press. SASUSG (1996). Sustainable use issues and principles Southern Africa Su stainable Use Specialist Group, IUCN Species Survival Commission : 23. Schedler, A. (1999). Conceptualizing Accountability. In The Self Restraining State: Power Accountability in New Democracies London, Lynne Rienner Publishers. Schlager, E. and E. Ostr om (1992). Property rights regimes and natural resources: a conceptual analysis. Land Economics 68 (3): 249 162. Schmitter, P. C. and T. L. Karl (1991). What democracy is....and Is not. Journal of Democracy 2 : 75 88. Schuster, B. (2007). Proceedings of the 4th National CBNRM conference in Botswana 20th 23rd of November 2006 and the CBNRM status report 2006 B. Schuster. Gaborone, Botswana, IUCN CBNRM Support Programme.

PAGE 262

262 Shackleton, S. B., E. Campbell, E. Wollenberg and D. Edmunds (2002). Devolution and community based natural resource management: Creating for local people to participate and benefit? Natural Resource Perspectives No. 76. London, UK, Overseas Development Institute. Simasiku, P., H. Simwanza, G. Tembo, S. Bandyopadhyay and J. M. Pavy (2008 ). The impact of wildlife management policies on communities and conservation in Game Management areas in Zambia: Message to policy makers. Natural Resources Consultative Forum Lusaka, Zambia, World Bank. Smoke, P. (2003). Decentralization in Africa: Goa ls, dimensions, myths and challenges. Public Administration and Development 23 : 7 16. Spinage, C. (1998). Social change and conservation misrepresentation in Africa. Oryx 32 (4): 265 276. Stuart Hill, G., R. Diggle, B. Munali, J. Tagg and D. Ward (200 7). The event book system: community based monitoring in Namibia. Participatory Learning and Action 55 (December 2006): 70 78. Taylor, M. (2002). The shaping of san livelihood strategies: Government policy and popular values. Development & Change 33 (3): 4 67 488. Taylor, R. (2009). Community based natural resource management in Zimbabwe: the experience of CAMPFIRE. Biodiversity and Conservation 18 : 2563 2583. Teklu, T. and S. Asefa (1999). Who participates in labor intensive public works in Sub Saharan Africa: Evidence from rural Botswana and Kenya. World Development 27 (2): 431 438. Turner, M. and D. Hulme (1997). Governance, Adminstration and Development: Making the State Work Houndmills, Inglaterra: MacMillan. Verba, S., K. L. Schlozman and H. E. Brady (1995). Voice and equality: Civic volunteerism in American Politics Cambridge, MA, Havard University Press. Wade, R. (1987). The Management of Common Property Resources: Finding a Cooperative Solution. The World Bank Research Observer. 2 (2): 219 234. Wells, M. and K. Brandon (1992). People and parks: linking protected area management witrh local communities. Washington DC, World Bank. Westergren, A., S. Karlsson, P. Andersson, O. Ohlsson and I. R. Hallberg (2001). Eating difficulties, need for a ssisted eating, nutritional status and pressure ulcers in patients for storke rehabilitation. Journal of clinical nursing 10 : 257 269.

PAGE 263

263 White, S. (1996). Depoliticising development: The uses and abuses of participation. Development in Practice 6 (1): 6 1 5. Wuensch, K. L. (2009). Binary logistic regression with PASW/SPSS Retrieved October 4th, 2010, from core.ecu.edu/psyc/wuenschk/MV/Multreg/Logistic SPSS.pdf. Zambia CSO (2010). Zambia 2010 census of population and housing Lusaka Zambia, Republic of Zambia, Central Statistics Office. ZAWA (2003). Village Action Group Constitution Z. W. Authority. ZAWA (2011). Kafue National Park General Management Plan Lusaka, Zambia, Zambia Wildlife Authority. Zimmerman, J. F. (1999). The New England town meetin g: Democracy in action. Westport, CT, Praeger Publishers.

PAGE 264

264 BIOGRAPHICAL SKETCH Patricia Mupeta Muyamwa is a natural resource governance specialist with 13 rural dev elopment and conservation. She has worked in several protected area systems in Southern Africa, utilizing the CBNRM approach as a means of reducing threats to large ecosystems and addressing rural poverty. Patricia grew up in the copper mining town Kitwe in north west in b iological s ciences at the University of Zambia. Upon completing her first degree, she worked for the Environmental Council of Zambia and the Wildlife Conservation Society of New York from 1999 2004. It was during this period she found her career calling, has conducted research and worked in a total of fifteen rural communities in Botswana, Namibia and Zambia Prior to her PhD, she attained her Master of Science degree in conservation and tourism at the Durell Institute of Conservation and Ecology (DICE) at the University of Kent in the United Kingdom. For her masters program, she conducted r esearch in sustainable tourism in northern Zambia. Her doctoral research examines the decentralization process in the wildlife sectors of Botswana and Zambia The study examines whether this has created an enabling environment for rural communities to ben efit from wildlife revenues She began her studies at the University of Florida in August 2006 and intends to receive her PhD in the fall of 2012.