IDRC - Lib.

Institutional Analysis

Readings and Resources for Researchers

Volume 5

Compiled By Steve Langill

for the Community-Based Natural Resource Management Program Initiative, IDRC

April, 1999

For further information about this document, please contact IDRC at the following address:

Claire Thompson, Programs Branch, IDRC, P.O. Box 8500, Ottawa, Ontario, Canada K1G 3H9

/jRc14,v A/ 6:4L

. /('/ 51 o Acknowledgments

Special thanks to Sam Landon of the Canadian International Development Agency (CIDA), Karen McAllister of IDRC, and Lisa Armstrong (Development Consultant) for their help and input in compiling this resource book. CBNRM Social Science Resource Kit Institutional Analysis Resource Book

Table of Contents

A. The Community-Based Natural Resource Management (CBNRM) Social Science Resource Kit

B. Readings on Institutional Analysis

C. Bibliography

D. Obtaining Documents Listed in the Bibliography

E. Websites and Electronic Information CBNRM Social Science Resource Kit Institutional Analysis Resource Book A. The CBNRM Social Science Resource Kit

What is the CBNRM Social Science Resource Kit? This kit is a reference tool to assist researchers funded through IDRC's Community Based Natural Resource Management (CBNRM) program in Asia to apply concepts, analytical approaches and research methods from the social sciences in their research.

What is the Format of the Kit? The kit is being delivered as a set of resource books, each dealing with a different key issue area related to CBNRM research. The topics/issue areas covered include: Gender; Community-Based Natural Resource Management; Participatory Research; Indigenous Knowledge; Institutional Analysis; Conflict Management and Multi-Stakeholder Analysis; and Common Property Resources, Tenure and Property Rights. Depending on feedback received from these materials, other topics or issues may be considered for coverage in future. In addition to the resource books, limited funds are being provided for IDRC project researchers to purchase books from an assembled list covering the above CBNRM-related topics. Further information on this is being sent separately to each project.

What is in the Resource Books? The resource books contain photocopies of selected readings excerpted from books, academic journals, field reports and training manuals. Depending on the subject, the readings include conceptual and methodological issues, research tools, and illustrative case studies. Each source book also includes an annotated bibliography, a list of references, and information on electronic (internet) resources. Instructions on how to use the Centre's literature search and document delivery services (free to IDRC-funded institutions) are also provided.

Readers will find that some of the material in each resource book is contradictory. The intent of the Kit is to expose researchers to a range of academic perspectives, rather than to choose only one view. This means that readers of this material will have to think about the different arguments presented and choose for themselves an interpretation of these concepts and methods which is sensible for their own research project. Readers should also note that the views expressed in the readings are those of the author(s) and do not necessarily represent those of IDRC.

Why Has the Resource Kit Been Prepared? The impetus for developing the kit stems from specific requests from IDRC research recipients for tools and resources to assist them in doing research for community-based natural resource management. For many of these researchers CBNRM is a new concept requiring analytical tools and research methods that are quite different to those they had received through formal or

A-1 CBNRM Social Science Resource Kit Institutional Analysis Resource Book other training. Researchers wanting to learn these new concepts and methods have been constrained by a lack of access to well-stocked libraries, relevant databases and internet sites.

The kit is also part of an effort by the CBNRM Program at IDRC to promote approaches to research that are participatory, action-oriented, multidisciplinary and grounded in local experience and local knowledge.

Who Should Use the Kit? If your research deals with Community-Based Natural Resource Management and is sponsored by IDRC, you should refer to the information in each volume to help you to undertake your research. IDRC-supported researchers will find that the concepts, tools and methods covered in these reference books will be used repeatedly in research reports, workshops, meetings, correspondence, and in evaluation of your work. You will also find it helpful to understand and apply these concepts if you submit future research proposals. The Kit will also be of wider interest and we hope that it can serve as a useful reference collection for researchers who otherwise would have difficulty getting access to this material.

How Were Readings Selected for the Resource Kit? The reading were selected from existing publications based on literature searches and consultations with academics and practitioners in the respective fields. From these sources the materials have been further selected for:

readability/clarity of the writing suitability for an audience with limited English language skills suitability to the CBNRM project contexts emphasis on definition of terms and detailed explanation of concepts

IDRC-supported CBNRM researchers are working in over 11 countries in Asia representing a wide range of cultural and educational backgrounds. Many researchers do not read English as a first language and a majority have not had formal training in the Social Sciences. For these reasons an effort has been made to include materials that will be instructive and accessible both for newcomers to the topic and for those with a background in the subject area.

How Might the Resource Kit be Used? These resource books are only a starting point for researchers looking for information on a specific topic. The readings are meant to stimulate research questions and further inquiry. The research tools provided are intended as catalysts for adaptation and innovation of new site-specific tools, methods and analytical frameworks. The bibliographies will assist each project and researcher to pursue more targeted information beyond what is provided here.

A-2 CBNRM Social Science Resource Kit Institutional Analysis Resource Book

Some specific actions you might take within your research team and/or institution to make more effective use of this material:

identify specific topics which are most relevant to your research and assign responsibility to specific members of the team to review these materials. Take turns briefing other team members on what you have learned from each Kit volume.

questions? Ask external project advisors or IDRC program staff if you have questions arising from your review of this material.

organize training sessions using these reference materials together with local resource persons, designated team members, or other experts.

translate the best articles for broader circulation.

request reference materials or literature searches from the IDRC library.

read some of the books in the bibliography to deepen your knowledge and learn other cases and examples. Books and articles which you have read and which are relevant to your own research can be cited, if appropriate, in your research proposals or reports.

order books selected from the list (provided separately).

inform IDRC of any changes to your projects that have come about as a result of this material.

discuss the contents of the readings within your research team and identify what adaptions you could make for the conditions of your project.

CBNRM Social Science Resource Kit Institutional Analysis Resource Book B. Readings on Institutional Analysis

This section includes photocopied readings on institutional analysis as it relates to community-based natural resource management (CBNRM). A brief introduction to the topic and an overview of the readings is provided below, followed by the reference information for each selection. The readings themselves are numbered and marked with corresponding tabs for convenience.

1. Introduction

Institutional analysis has become a useful tool in the field of community-based natural resource management (CBNRM) for understanding how local communities manage resources, and how improvements in management can be initiated. Institutions are generally defined as "complexes of norms and behaviors that persist over time by serving collectively valued purposes" (Uphoff, 1986). They are the arrangements or 'rules of the game' which shape the behavior of local community members and include common understandings about how issues and problems are to be addressed and solved. Institutions are dynamic. They respond to changes in local actors and their understanding, as well as to external power or environmental conditions, but the process of change can be difficult. Institutional analysis in the CBNRM context involves examining these rules and the ways they affect the management strategies of resource users.

When undertaking institutional analysis, it is important to distinguish between the terms institution and organization. While institutions can be described as the long-standing rules and rights governing social and productive behavior, organizations are the 'players' and structures, or "groups of individuals bound together by some common purpose to achieve objectives" (North, 1990). Institutions thus form the 'framework' upon which organizations are based. While CBNRM research is primarily interested in uncovering the underlying rules of resource management, organizations can be an important focus for research as well in that they often serve to operationalize institutional arrangements (Pomeroy, 1996).

Of particular relevance to CBNRM research are local-level institutions-those institutions which operate within a single community or between several communities. They can range from formal, with documented and transparent rules for transactions and decision-making (e.g. regulations, membership, executive structure and powers, fines, taxes, duties and obligations, etc) to informal, where rules and roles are usually not documented. Informal institutions can be more flexible or dynamic over time, and

B-1 CBNRM Social Science Resource Kit Institutional Analysis Resource Book may not be recognized or understood especially by outsiders; but they can also be very rigid, very strong and very clear to internal participants. Local institutions can also range from indigenous, with their origins clearly going back many generations, to contemporary or more "modern" forms.

Conventional approaches to community development have tended to either ignore altogether the importance of local institutional arrangements for natural resource management, or focus on only a single local institution. It is important for CBNRM researchers to realize that institutions form a fundamental link between local communities and their environments, and that it is through these rules that the collective action associated with controlling access to local resources is organized. Furthermore, researchers must recognize that community resource management is often shaped by a number of overlapping institutions from the social, political, economic and religious spheres. Communities, rather than being homogeneous entities governed by a single set of rules, are made up of diverse institutions and organizations with conflicting values and priorities based on differences in gender, age, wealth and other factors. These community institutions, in turn, are interconnected with and influenced by broader institutions operating at the district, regional, national and international levels. This internal and external diversity of institutions must be taken into consideration when investigating how rules affect behaviors related to resource management.

Institutional analysis is carried out for two basic reasons. The first is to understand, from an insider's perspective, how local resources are currently managed: what institutions are operating at the local level to manage resources? How do these institutions affect behavior related to resource management? How do they interact with each other and with external institutions? The second is to determine the best way to manage local resources based on a sound knowledge of current local management regimes: How effective are existing institutions for resource management? How should these capabilities be utilized for the design of more effective resource management schemes e.g. should local or indigenous institutions be enlisted in their current forms, or should new forms be created that are patterned on local or indigenous ones? At what other institutional level(s) should solutions be focused?

Several analytical frameworks have been put forth to organize information about the interaction between local institutions and the environment. For example, the Institutional Analysis and Development Framework (IAD), developed by Elinor Ostrom and others focuses on how rules (institutions) and attributes of the physical and cultural world combine to shape outcomes. Another framework, the Environmental Entitlements

' See, for example, E. Ostrom, Gardner and Walker, 1994 (ch.2); Oakerson, 1992; Schlager and Ostrom, 1992; Tang, 1992.

B-2 CBNRM Social Science Resource Kit Institutional Analysis Resource Book approach developed at the Institute of Development Studies, focuses on the behaviors which result from rules and is based upon three key elements-endowments, environmental entitlements, and capabilities-and how the relationships between them are mediated by various institutions.

Researchers need to exercise caution in certain areas when carrying out institutional analysis and utilizing results. For example, researchers should not assume that local or indigenous institutions for resource management always exist, or that they are always effective or efficient. They should also not assume that a particular scale of arrangement-international, national, regional, district or local-is the correct scale for all resource management improvements. Situations will differ relative to context, and solutions will have to be based on the particular circumstance of a given locale. As well, it is important to realize that while adjustments to existing local institutions are possible, these adjustments must be based on the perceptions and suggestions of local people themselves. Outsiders cannot simply 'mix and match' local institutions according to their purposes and without consultation with community members. Local institutions are interdependent and integrated with other aspects of community life, so that changes made to one will likely result in changes to others.

The overall strength of institutional analysis lies in its potential to contribute to local capacity-building. By understanding local institutions and drawing upon the repertoire of management arrangements that they represent, development practitioners and local people can work together to develop the most effective, sustainable institutional arrangements for the community-based management of resources.

H. An Overview of the Readings

The photocopied readings that appear here offer conceptual and methodological insights into the role of institutions and institutional analysis in community-based natural resource management. The readings have been classified under four sub-headings for the purpose of clarity.

1. Concepts and Analytical Frameworks

The first three readings are meant to provide an introduction to the topic of institutional analysis, defining key concepts and describing analytical frameworks that have been developed to organize information about institutions and the role they play in the management of local resources.

In the first reading, "Environmental Entitlements: A Framework for Understanding the Institutional Dynamics of Environmental Change", Leach, Mearns and Scoones first

B-3 CBNRM Social Science Resource Kit Institutional Analysis Resource Book provide a clear review of the literature on the role of institutional dynamics in environmental change, and then describe a conceptual framework which highlights the central role of institutions in mediating the relationships between environment and society 2. They use this framework to explore how different social actors command environmental goods and services, illustrating their approach with recent research done in India, South Africa and Ghana.

In the second reading, "The Institutional Analysis and Development Approach", Ostrom describes the IAD framework for identifying and analyzing the links between the physical environment and the sociocultural and institutional arrangements of a community. The paper describes the main components of the IAD framework, defines the concept of `rules', and discusses criteria by which rules and their outcomes can be evaluated.

The next reading, "Analysis of Fisheries Co-Management Arrangements: A Research Framework", provides a good illustration of applying the IAD framework in a project setting. ICLARM staff describe the institutional analysis framework which they are using in their research to analyze the institutional and organizational aspects of fisheries co-management arrangements. The paper also includes a background discussion of important concepts such as common property, fisheries co-management, institutional analysis and rights and rules.

2. Participatory Methods for Institutional Analysis

The following readings offer descriptions of participatory research techniques that can be used specifically for field-level analysis of local institutions. Readers should also check the resource books we have compiled on Participatory Research and Indigenous Knowledge for descriptions of other relevant field methods and techniques.

In the reading "Institutional Analysis", taken from IIRR's toolkit Participatory Methods in Community-Based Coastal Resource Management, Pomeroy provides a suggested approach for carrying out a participatory, field-level analysis of local institutions. The paper includes some guide questions and a brief discussion of the strengths and limitations of institutional analysis.

2 A shorter discussion of this framework was also included as a reading in the resource book we compiled on Community-Based Natural Resource Management (Leach et al, 1997. "Challenges To Community-Based Sustainable Development: Dynamics, Entitlements, Institutions". See Section C of this manual for bibliographical information and abstract).

B-4 CBNRM Social Science Resource Kit Institutional Analysis Resource Book

The fifth reading, from FAO and SEAGA's Field Handbook: Socioeconomic and Gender Analysis Programme, provides a description of two participatory field techniques that have been used in field-level research of local institutions: Venn diagrams and institutional profiles. Venn diagrams can be used to learn about local institutions and their linkages with outside organizations and agencies, while institutional profiles can be used to learn about the goals, achievements and needs of local groups and institutions.

3. Case-Studies

The final readings provide the reader with case-studies of research with a focus on institutions and institutional analysis.

In the sixth reading, "Transforming Customary Law and Coastal Management Practices in the Maluku Islands, Indonesia, 1870-1992", Zerner discusses the development of sasi, a family of traditional institutions and practices used to regulate access to agricultural and marine resources throughout the Maluku Islands. The study underscores the dynamic nature of sasi over time, and shows how wider political, economic and religious forces have greatly influenced its historical development. The focus of any study of sasi, according to Zerner, should be on developing an understanding of why and how this institution has changed over the years, and how this knowledge can be used to design effective changes for the future.

In the next reading, "Institutions and Externalities in the Inland Fisheries of Bangladesh", Toufique, using an IAD approach, analyzes the institutional arrangements for regulating the fisheries of Sirajganj, Bangladesh. He explores the relationship between the establishment of private property rights and resource conservation, analyzing the physical characteristics of the resource system and the nature of existing property rights. He concludes that, for the most part, these institutional arrangements have been successful in containing resource losses, and that the rules traditionally followed by resource users are being maintained by the agents who now hold property rights over the resource system.

In the next reading, "Organizational Structure, Performance and Participation: Forest User Groups in the Nepal Hills", Joshi et al analyze the organizational structure and effectiveness of forest user groups in Nepal based on three elements: mode of formalness; decision-making structure; and group size. The authors provide a useful discussion of their methodological approach, including site selection, sampling procedure, research method, and interpretation of results using basic statistical analysis. They conclude by making some recommendations about how to make forest user groups more effective for forest development.

B-5 CBNRM Social Science Resource Kit Institutional Analysis Resource Book

In the final reading, "The Symbolic Making of a Common Property Resource: History, Ecology and Locality in a Tank-irrigated Landscape in South India", Mosse criticizes current approaches to the analysis of common property regimes for producing inaccurate representations of local communities. He uses his study of indigenous institutional arrangements for tank irrigation in Tamil Nadu, India to show the importance of considering how community resource institutions are affected by both their specific political and historical context, an approach which he argues is lacking in other models of collective action. CBNRM Social Science Resource Kit Institutional Analysis Resource Book References

A copy of the full-text of each of the following articles is included in this document. To find a reading, flip to the corresponding tab number. These materials have been reproduced with permission from the publishers.

Concepts and Analytical Frameworks

1. Leach, Melissa, Robin Mearns and Ian Scoones. 1997. Environmental Entitlements: A Framework for Understanding the Institutional Dynamics of Environmental Change. IDS Discussion Paper 359 (39 pages). Reprinted with permission from the Institute of Development Studies (IDS).

2. Ostrom, Elinor. 1998. The Institutional Analysis and Development Approach. In Designing Institutions for Environmental and Resource Management, eds. Edna Tusak Loehman and D. Marc Kilgour. Cheltenham, UK: Edward Elgar Publishing, pp. 68-90. Reprinted with permission from Edward Elgar Publishing.

3. Pomeroy, R. S. (senior author). 1996. Analysis of Fisheries Co-Management Arrangements: A Research Framework. Prepared by Fisheries Co- Management Project Core Staff at the International Center for Living Aquatic Resources Management (ICLARM) and the North Sea Center (NSC). Reprinted with permission from senior author.

Participatory Methods for Institutional Analysis

4. Pomeroy, R.S. 1998. Institutional Analysis. In Participatory Methods in Community- Based Resource Management (3 Vols.). Vol. 2, pp. 118-130. International Institute for Rural Reconstruction (IIRR). Silang, Cavite, Philippines. Reprinted with permission from IIRR.

5. FAO, SEAGA. 1998. Venn diagrams; Institutional profiles. In Field Handbook: Socioeconomic and Gender Analysis Programme. Tools A5 and A6. Rome: FAO. Available at http://www.fao.org/sd/seaga/SEfh000l.htm Reprinted with permission from FAO. CBNRM Social Science Resource Kit Institutional Analysis Resource Book

Case Studies

6. Zerner, C. 1994. Transforming Customary Law and Coastal Management Practices in the Maluku Islands, Indonesia, 1870-1992. Chapter 4 in Natural Connections: Perspectives in Community-Based Conservation. Western, D. et al. (eds.). Pp. 80-112. © Island Press, Washington, DC and Covelo, CA. Reprinted with permission from Island Press. For book information or ordering, 1-800-828-1302.

7. Toufique, Kazi Ali. 1998. Institutions and Externalities in the Inland Fisheries of Bangladesh. Land Economics, vol. 74, no.3 © Pp. 409-421. Reprinted with permission from the University of Wisconsin Press.

8. Joshi, N.N., Jali, N.M., and Hamid, A.H. 1997. Organizational Structure, Performance and Participation: Forest User Groups in the Nepal Hills. In People and Participation in Sustainable Development. G. Shivakoti et al (eds.). Proceedings of an International Conference held at the Institute of Agriculture and Animal Science, Rampur, Chitwan, Nepal, 17-21 March 1996. Bloomington, Indiana and Rampur, Chitwan. Reprinted with permission from the Workshop in Political Theory and Policy Analysis, Indiana University.

9. Mosse, D. 1997. The Symbolic Making of a Common Property Resource: History, Ecology and Locality in a Tank-irrigated Landscape in South India. Development and Change, Vol. 28, Pp. 467-504. Reprinted with permission from Blackwell Publishers. ENVIRONMENTAL ENTITLEMENTS A Framework for Understanding the Institutional Dynamics of Environmental Change'

IDS Discussion Paper 359

Melissa Leach, Robin Mearns and Ian Scoones2

Summary

While there is now widespread consensus within international development circles that 'sustainable development' should be based on local-level solutions derived from community initiatives, there is also a growing perception that the practical implementation of what is termed 'community-based sustainable development' often falls short of expectations. This overview paper seeks to complement and add to this emerging critique, with a particular focus on the implications of infra-community dynamics and ecological heterogeneity. It offers a conceptual framework that highlights the central role of institutions in mediating the relationships between environment and society, where institutions are understood as regularised patterns of behaviour between individuals and groups in society. This framework is grounded in an extended form of entitlements analysis used to explore the ways differently positioned social actors command environmental goods and services that are instrumental to their wellbeing. Further insights are drawn from areas of scholarship including analyses of social difference and social agency; 'new', dynamic ecology; new institutional economics; structuration theory; and landscape history. The theoretical argument is illustrated with. reference to recent empirical research in India, South Africa and Ghana.

' We thank Richard Black, Susanna Davies, Stephen Devereux, Martin Greeley, Ann Hudock, Karim Hussein, Hamish Jenkins, Simon Maxwell, Mark Robinson and Jeremy Swift for their valuable comments on a previous draft. The paper has also benefited from comments by participants in seminars at the Institute of Development Studies, University of Sussex; and CSERGE, University College London. The research is supported by the ESRC Global Environmental Change Programme (Award No. L320253213).

2 Environment Group, Institute of Development Studies, University of Sussex Copyright © Institute of Development Studies, 1997

ISBN 1 85864 115 2

2 Contents

1 Introduction 5

2 Narratives on Community and Environment 7

3 Community, Social Difference and Dynamics 10

4 Dynamic Ecologies 12

5 Environmental Entitlements 14

6 Institutions 24

7 Structuration of the Environment 27

8 Conclusions 29

References 32

1 Introduction

The consensus in the wake of the United Nations Conference on Environment and Development (UNCED) suggests that the implementation of what has come to be known as 'sustainable development' should be based on local-level solutions derived from community initiatives (Ghai and Vivian 1992; Ghai 1994). This reasoning comes with a long pedigree, dating at least from The Ecologist's (1972) 'Blueprint for Survival', Schumacher's (1973) Small is Beautiful and, more recently, the Brundtland Commission (WCED 1987; Conroy and Litvinoff 1988). Statements of intent on global environmental problems issued following the 1992 Earth Summit, including Agenda 21 and the Desertification Convention, strongly advocate as solutions a combination of government decentralisation, devolution to local communities of responsibility for natural resources held as commons, and community participation (Holmberg et al. 1993). Such approaches - evident in the policies and programmes of national governments, donor agencies and nongovernmental organisations (NGOs) - argue for 'co-management', or an appropriate sharing of responsibilities for natural resource management between national and local governments, civic organisations, and local communities (Adams and McShane 1992; Berkes 1995; Baland and Platteau 1996; Borrini-Feyerabend 1996). The practical implementation of community-based sustainable development initiatives, however, has often fallen short of expectations. A number of reasons have been identified, including a tendency (despite rhetoric to the contrary) for the 'intended beneficiaries' to be treated as passive recipients of project activities (Pimbert and Pretty 1995; cf Arnstein 1969); a tendency for projects to be too short-term in nature and over-reliant on expatriate expertise; and a lack of clear criteria by which to judge sustainability or success in meeting conservation or development goals (Western et al. 1994). Others suggest that the interests of certain social groups have been consistently marginalised (Hobley 1992; Sarin 1995). In this paper, we seek to complement and add to this emerging set of critiques and attempts to improve the practice of community-based sustainable development, through a particular focus on institutions as mediators of people-environment relations. We see institutions as 'regularised patterns of behaviour between individuals and groups in society' (Mearns 1995a: 103), rather than as community-level organisations. Yet the latter view prevails in current approaches to community- based sustainable development, when they are concerned with institutions at all. In the emerging consensus, community-level organisations are commonly assumed to regulate the use of relatively homogeneous environments in the community's interests. Environmental degradation is assumed to reflect a growing lack of synchrony between the community and its natural environment, and the implied solution is to reconstitute community-based natural-resource management institutions so as to restore harmony to environment-society relations. Recent advances in ecological theory suggest, however, that many more environments than was previously thought are characterised by high variability in time and space. This has important implications for managing natural resources and

5 environmental risk, and suggests that understanding environmental change involves looking beyond natural-resource depletion or degradation in the aggregate. Similarly, local communities may be shown to be dynamic and internally differentiated, and the environmental priorities and natural- resource claims of social actors positioned differently in power relations may be highly contested. These factors point to the importance of diverse institutions operating at multiple scale levels from macro to micro, which influence who has access to and control over what resources, and arbitrate contested resource claims. To date, a poor understanding of such dynamic institutional arrangements has impeded practical efforts in community-based sustainable development. In this paper, we offer a review of relevant theoretical literature and build a conceptual framework, extending our earlier work on the notion of 'environmental entitlements' (Leach and Mearns 1991; Mearns 1995b, 1996a), to assist in understanding the role of institutional dynamics in environmental change. This framework seeks to elucidate how ecological and social dynamics influence the natural-resource management activities of diverse groups of people, and how these activities in turn help to produce and to shape particular kinds of environment. Rather than framing environmental problems simply in terms of aggregate population pressure on a limited natural-resource base, a more disaggregated entitlements approach considers the role of diverse institutions in mediating the relationships between different social actors, and different components of local ecologies. The insights derived from such analysis are intended to help target external interventions more effectively, whether the objectives are to protect and promote the environmental entitlements of particular social groups, or to foster particular environmental outcomes. Such forms of intervention, we argue, may involve a much more diverse range of institutions, whether locally or at the level of national legal and policy frameworks, than have been addressed in community-based sustainable development efforts to date. The paper is organised as follows. Section 2 considers the prevailing assumptions underlying approaches to community-based sustainable development, and finds that they are flawed in their characterisations of communities and their local environments. Sections 3 and 4 review the theoretical foundations of these characterisations in the social and ecological sciences respectively, and offer alternative perspectives from contemporary debates on social and ecological dynamics and heterogeneity. In section 5, we outline the basis of our environmental entitlements framework, tracing its origins in the broader entitlements literature, defining terms, and specifying the distinguishing features of our own approach. Section 6 discusses in more detail the way we conceive of institutions, since these are the factors operating at different scale levels that undergird entitlement relations. In other words, a variety of formal and informal institutions influence which sets of social actors are able to command various environmental goods and services that are instrumental to their wellbeing. Our understanding of institutions is enriched by the alternative perspectives on community and ecology reviewed in sections 3 and 4, and leads us to a view of institutional change that takes seriously both history and power relations. Section 7 shows how the institutional dynamics that are at the core of the environmental entitlements framework also powerfully influence

6 environmental change itself. Section 8 concludes the paper with some preliminary remarks concerning the implications of the analysis for improving policy and practical approaches to community-based sustainable development. Throughout the paper, we refer to examples drawn from recent empirical research on community-based or co-management initiatives in the fields of forestry (Ghana), protected areas (South Africa), and watershed development (India) .3

2 Narratives on Community and Environment

Our environmental entitlements framework is in part a critical response to key aspects of the community-based sustainable development consensus, so we begin by clarifying what is implied by this consensus. At least superficially, recent approaches to community-based sustainable development appear as diverse as their varied implementing agencies and natural-resource settings, yet they rest on a set of common assumptions about community, environment, and the relationship between them. One fundamental assumption is that a distinct community exists. While definitions vary, approaches commonly focus on 'the people of a local administrative unit ... of a cultural or ethnic group ... or of a local urban or rural area, such as the people of a neighbourhood or valley' (IUCN/WWF/UNEP 1991: 57). Such communities are seen as relatively homogeneous, with members' shared characteristics distinguishing them from 'outsiders'. Sometimes social difference within communities is acknowledged, and explicit efforts are made, using participatory rural appraisal methods, for example, to specify the implications for project interventions. But all too often it is implied that the public airing of conflict is sufficient, and that social consensus and solidarity will necessarily result (cf. Mosse 1994). Equally fundamental is the assumption of a distinct, and relatively stable, local environment (usually not defined) which may have succumbed to degradation or deterioration, but has the

potential to be restored and managed sustainably. The community is seen as the appropriate body to

carry out such restoration and care, and is envisaged as being capable of acting collectively towards common environmental interests. For instance 'primary environmental care', a term coined to encapsulate a range of operational experiences in the field of community-based sustainable development, has been defined as ;a process by which local groups or communities organise themselves with varying degrees of outside support so as to apply their skills and knowledge to the care of natural resources and environment while satisfying livelihood needs', (Pretty and Guijt 1992: 22). The tendency to subscribe to obviously oversimplified assumptions is not confined to community-based sustainable development. It is now well recognised that most development programmes and projects that succeed in mobilising funds and resources rely on sets of simplifying

3 These case studies arc being researched and authored by Seth Afikorah-Danquah (Ghana, with supervision from Edwin Gyasi, Department of Geography, University of Ghana at Legon); Thembela Kepe (South Africa, with supervision from Ben Cousins, Land and Agrarian Studies Programme, University of the Western Cape), and Meenakshi Ahluwalia (India, with supervision from M.S. Rathore, Institute of Development Studies, Jaipur).

7 assumptions about the problem to be addressed and the approach to be taken which provide what Hoben (1995) calls a 'cultural script for action' (cf Hirschmann 1968). Such assumptions are frequently linked together within what Roe (1991) has termed 'development narratives': stories about the world that frame problems in particular ways and in turn justify particular solutions. It is partly through narratives - or what others have theorised as development discourses (e.g. Ferguson 1990; Escobar 1995; Apthorpe and Gasper 1996) or received wisdom (Leach and Mearns 1996) - that assumptions are stabilised and rendered unquestionable, so that they gain power and persistence in project and policy arenas despite the frequent absence of empirical data to support them. In the case of community-based sustainable development, a powerful, generalised narrative links the above, related assumptions. At some (usually unspecified) time in the past, so it goes, the community lived in balanced harmony with nature. Either population levels were so low that the environment was little disturbed, or community institutions, including ritual and traditional authority, served to regulate resource use so that society and environment remained in equilibrium. But the narrative goes on to suggest that this harmony has been disrupted, leading to a disequilibrium between people and environment, and hence to environmental degradation. A range of factors may be called to account, including population growth, technological change, the breakdown of traditional authority, commercialisation, modernity, social change and new urban aspirations, the immigration of stranger populations, or the intrusion of inappropriate state policies. What is required, the narrative suggests, is to bring community and environment back into harmony: 'policies that bring human numbers and life-styles into balance with nature's capacity' (IUCN/WWF/UNEP 1991: 10). This requires either the rediscovery and rebuilding of traditional, collective resource management institutions, or their replacement by new ones, such as the community management plans and village environmental committees so often associated with community-based sustainable development strategies. Versions of this narrative are to be found in all three of our research sites, where they help to frame the ways that policymakers conceive of localised problems, and to justify the particular approaches taken in project interventions. In Ghana, the landscape of the study area in Wenchi district of the northern forest-savanna transition zone is characterised by forest patches in savanna. In pre-colonial times, many policymakers suppose, the area was entirely covered by the semi-deciduous forest thought to be its natural climax vegetation. This vegetation cover was maintained by village communities that collectively practised forest farming styles (long rotational bush fallow and minimum tillage which preserved a root mat of forest species); that had effective traditional organisations which regulated forest product collection and fire setting; and whose traditional religious and ritual practices involved

the preservation of certain areas of undisturbed natural forest as sacred groves. But changes during the twentieth century, accelerating during the last few decades, are seen to have disrupted this harmony between community and environment, encouraging progressive loss of forest to savanna. Migrants have increasingly settled from the north, bringing with them, it is argued, forest-destructive

8 savanna farming practices. With the weakening of traditional controls, the use of bush fire is perceived to have increased for the worse. State laws have reduced security of tenure for local people and have weakened incentives to preserve trees and forest patches, while new commercial activities such as charcoal burning have hastened deforestation. Many initiatives by government, donors and NGOs are attempting to re-establish and enhance local control over forest resources in the interests of reforestation and more sustainable management. Activities include encouraging communities to prepare management plans for remaining forest patches and sacred groves within their territory, which are to be recognised in state forest policy. A strong narrative of environmental degradation also surrounds the Indian case study site. The Aravalli Hills of Rajasthan are portrayed as having been clothed, until the mid-twentieth century, in an unbroken green forest canopy, replete with wildlife. Rainfall was abundant, and pressure on forest resources was minimal from low populations of (mainly tribal) gatherers. Farming was largely confined to fertile valley bottoms irrigated under well-organised community systems. A series of changes are held to have denuded and 'desertifred' the hills, with accompanying desiccation and soil erosion. In response to new market opportunities and changes in Rajasthani state policies associated with post-independence land reforms, growing populations deforested large areas of hillside for commercial wood fuel and timber sale. People subsequently grazed livestock there, under unregulated open-access systems which led quickly to overgrazing, while farming spread onto increasingly marginal hill land with accompanying soil erosion. This growing pressure from

unregulated resource use is held to have gone hand- in- hand with a secular decline in rainfall. This narrative of decline justifies the perceived need to re-green the hills through watershed development programmes intended to bring the communities living within watershed boundaries back into harmony with their local environments through technical and organisational innovations to facilitate improved resource use. The Nayakheda watershed project, supported by the Udaipur-based NGO Seva Mandir, has promoted soil and water conservation and a system of enclosures to regulate the use of commons through a community-level committee. Working together for the common environmental good, it is argued, in turn promotes social cohesion and enhanced community capacity to engage in collective social and political action. In the South African case, the coastal areas of the former Transkei in the Eastern Cape are regarded as areas of great conservation importance. Some argue that forest patches in ravines and along the coast are relics of once-large expanses of forest land and that the existence of rare, endemic species highlights the biodiversity significance of forest patches and intervening areas of coastal grassland. The established conservation narrative argues that such areas should be protected by controlling human use through the creation of nature reserves, strictly managing the fire regime and

introducing certain game species. The level of human use should, it is argued, be limited to the selective harvesting of certain grass and tree species. According to some commentators, particularly within the new majority-rule government of Eastern Cape Province, community involvement is

9 essential in order to encourage conservation awareness and compliance with regulations, and so community-based conservation initiatives are being encouraged. We do not dispute all elements of these community-based sustainable development narratives, either in general or in relation to the specific cases. However, we do argue that in important respects they have provided misleading guidelines for policy. This is less because their assumptions about community and environment are oversimplified than because they are basically flawed. We now develop this argument by addressing some of their foundations respectively in social and ecological theory, and by raising alternative perspectives put forward in recent theoretical debates.

3 Community, Social Difference and Dynamics

The assumption that communities can be treated as static, relatively homogeneous entities so often implicit in today's community-based sustainable development literature has identifiable roots in much earlier social theory. Early sociology and anthropology conceived of society as a bounded object or closed social system: a 'thing' more than the sum of its individual parts, analogous to an organism with its own goals and capacities (Durkheim 1895; cf Frisby and Sayer 1986). Individual members were seen as united by culture into 'moral communities' sharing common interests and mutual dependence. The related ideas of structural-functionalism, which came to dominate colonial anthropology (e.g. Radcliffe-Brown 1952; Fortes and Evans-Pritchard 1940), also find strong echoes in recent approaches to community-based sustainable development. Social structure was seen to drive rules which unproblematically governed people's behaviour and maintained social order. As a coherent system, social structure comprised parts that interlocked functionally to fulfil society's needs and maintain an equilibrium. Change to this equilibrial state - or traditional zero point - was conceived of mainly as being a result of external factors that precipitated breakdown and dysfunction, much as in community-based sustainable development narratives. A dominant approach in ecological anthropology and cultural ecology, especially in the 1950s

and 1960s, saw such functionally integrated societies as also functionally integrated with their environment; in much the same way that community-based sustainable development portrays community and environment as (once or potentially) linked in harmonious equilibrium. Society was seen as adapted to the environment through homeostatic feedback loops which ensured that resource use did not disrupt the ecological balance (e.g. Rappaport 1968). Social structure and institutions

such as traditional authority and ritual were seen as maintaining this functional adaptation, often over and above the consciousness of community members. Social science debates and empirical work have, however, fundamentally questioned many of these suppositions. First and most obviously, a very large body of work concerned with social difference, including much feminist scholarship, has highlighted the ways that gender, caste, wealth, age, origins, and other aspects of social identity divide and crosscut so-called 'community' boundaries. This work emphasises how diverse and often conflicting values and resource priorities - rather than shared beliefs and interests - pervade social life and may be struggled and 'bargained'

10 over (e.g. Carney and Watts 1991; Leach 1994; Moore 1993). Equally, it draws attention to power - however theorised - as a pervasive feature of social relations (e.g. Lukes 1974; Gaventa 1995). Feminist work especially has shown clearly how institutions can shape and reproduce relations of unequal power and authority (e.g. Kabeer and Subrahmanian 1996; Goetz 1996). While an emphasis on social difference is now uncontroversial within development studies more generally, arising as it did out of a critique of earlier, failed approaches to so-called 'community development', it is striking the degree to which simplistic notions of community are being reinvented in the context of practical efforts towards community-based sustainable development. Second, and closely related, has been a sustained critique of structural-functionalist approaches - and others that privilege structure such as Marxism - from a perspective that emphasises actors, action and agency. As Long and others have forcefully argued, communities cannot be treated as static, undifferentiated wholes, since they are composed of active people and groups. The behaviour of social actors is not driven automatically and unconsciously by structures; rather they actively monitor, interpret and shape the world around them (e.g. Long and van der Ploeg 1989; Long and Long 1992). Some of this actor-oriented work has been rightly criticised for being excessively voluntaristic, grounded in a methodological individualism rooted in the very different sociological traditions of Weber (cf. Lukes 1973). Interactionism and transactionalism in sociology and anthropology, for instance, tended to portray social life in terms of active decision-making and strategising by people with free choice (e.g. Barth 1992, Bailey 1969, Vincent 1978). Others have grappled with the links between agency and structure, emphasising how structures, rules and norms can emerge (and only exist) as products of people's practices and actions, both intended and unintended. These structural forms subsequently shape people's action, not by strict determination but by providing flexible orientation points which may either constrain or enable what is possible. Whilst some, routinised, action serves to reproduce structures, rules and institutions, other action has agency, serving to change the system and perhaps, in time, remake new rules (Giddens 1984; Bryant and Jary 1991; Long and van der Ploeg 1994; Bebbington 1994). Such a perspective conceives of social change very differently from the community-based sustainable development narrative of external disruption to a static society or community. Rather, history is seen as a manifold process of interaction between external and internal actions and events, in which contingencies and path-dependency play a significant part. As Abrams (1982) and others have emphasised, history is central and necessary to sociological enquiry, as the lens through which the relationships between agency, structure and power - or social actors and institutions - become apparent. These more critical perspectives do not reject the notion of 'community' altogether, but rather contextualise it by describing a more or less temporary unity of situation, interest or purpose among particular groups of social actors. For instance, the people of the Indian case study area are divided among at least seven multi-caste and multi-tribe hamlets, drawn from three different revenue

11 villages, and associated with different occupations and resource priorities. Whether or not they are regarded as comprising a single community depends on one's perspective and scale of analysis. They may sometimes appear as a larger, united community, for instance when village committee members represent Nayakheda to the NGO project staff, or in apparently uniting to support a tribal member's election to the local Panchayat council (Franzen 1996). Such representations of wider community are arguably best seen as actively created and manipulated by powerful actors for particular (and not necessarily shared) purposes, for instance in maintaining the project's support in brokering local politics. In other words, 'community' may be the contingent and temporary outcome of dynamic interaction between differentiated social actors, more than an enduring whole (cf Mosse, forthcoming).

4 Dynamic Ecologies

The science of ecology over much of the twentieth century - including the science that has influenced community-based sustainable development - has been built upon notions of equilibrium, balance, harmony and functional order, showing notable parallels with structural-functionalist theories in the

social sphere. But, just as there are problems with structural-functional explanations of community and social relations, so interpretations of nature and environment based on assumptions of balance and system regulation are subject to dispute. Key concepts in ecological science include gradual, linear change, homeostatic regulation of systems and stable equilibrium points or cycles (Cherrett 1987; McIntosh 1985; Pimm 1991). These concepts derive from much longer traditions of thought about the natural world, and from religious ideas about a divine, natural order (Worster 1985, 1993). The idea of the 'balance of nature' has been central to this thinking, and the quest for such balance or equilibrium has dominated natural science

enquiry. Evidence, however, has proved extremely limited; as Connell and Sousa (1983: 808) comment: If a balance of nature exists it has proved exceedingly difficult to demonstrate.' Despite this, the conceptual foundations of mainstream ecological science have been informed by equilibrium thinking. Theories of vegetation succession, population modelling, ecosystem functioning, or species- area relationships all have equilibrium assumptions at the core of their models and, not surprisingly, their findings and applied management recommendations (cf Botkin 1990; Zimmerer 1994). Thus succession theory has emphasised linear vegetation change and the idea of a stable and natural climax. Since Clements's early work in the United States (Clements 1916), this has become the guide for managing rangelands and forests, the benchmark against which environmental change is assessed. In the Ghana case study, for instance, semi-deciduous forest has been seen as the natural climax vegetation, and its restoration has been identified as a key management objective. Equally, models of population dynamics identified carrying capacities and maximum sustained yield levels for managed animal populations, using calculations based on assumptions of predictable and linear growth patterns in stable environments (Caughley 1977). Management regimes for domestic livestock,

12 wildlife and fisheries populations have been devised according to such assumptions; they have, for instance, informed the enclosure strategies for grazing and fodder management promoted in the India case study. At the same time, following the classic work by Odum (1953), ecosystem theory has focused on the system regulation of flows and so on how environmental impacts are assessed. The assumption is that natural systems are homeostatically regulated and, if disturbed, such regulation may collapse with detrimental consequences. Finally, conservation biology has claimed a stable relationship between species diversity and area (MacArthur and Wilson 1967) and so a basis on which biodiversity policy could be created and protected areas could be designed. For example, in South Africa a control-oriented protected area policy has dominated conservation thinking for much of the twentieth century, resulting in an exclusionary approach to national park and nature reserve management (Carruthers 1995). These ideas in ecological science have made some major contributions, of course, but as with any theory or model, it is unwise to assume they apply in all places, all of the time. Whilst there were disputes within each of these areas of theory as new empirical material emerged, new ideas made little headway and rarely challenged the core assumptions of the equilibrium framework, until the explosion of interest in mathematical ecology and the stability properties of models and - later - real, natural or managed systems during the 1970s (e.g. May 1976, 1977, 1986). This period saw the emergence of key concepts making up non-equilibrium theory and, more broadly, what has been termed the new ecology' (e.g. Botkin 1990). What does the new ecology look like? Three themes stand out (cf Scoones in preparation). First, an understanding of variability in space and time, which has led to work on time-series population analysis, stochastic and dynamic modelling (Hastings et al 1993; Elner and Turchin 1995) and an interest in the relationships between disturbance regimes and spatial patterning from patches to landscapes (e.g. Kolasa and Pickett 1991; Turner 1989). Second, nonequilibrium perspectives suggested a need to explore the implications of scaling on dynamic processes, which has lead to work on hierarchies and scale relationships in ecosystems analysis (Allen and Hoekstra 1991; O'Neill et al 1986; Dunning and Stewart 1995). Third, a recognition of the importance of history on current dynamics has led to work on environmental change at a variety of timescales (Worster 1990; Williams 1994). The latter parallels the historical emphasis of much recent social theory, and invites analytical attention to the historical relationships between environmental and social change. These ecological themes have prompted increased interest in understanding dynamics and its implications for management. For example, recent thinking in ecology informs our understanding of the key relationship between savanna grassland and forest areas. For both the Ghana and South

Africa case study sites, this is an important issue, as different products and different environmental values are associated with forests and grasslands. The conventional, equilibrial interpretation of

succession theory sees forest as a later successional form, closer to natural 'climax' vegetation, and

the presence of grasslands as evidence of degradation from a once-forested state. This linear

interpretation of vegetation dynamics has a major influence on the way such landscapes are viewed

13 by policy makers and others (Fairhead and Leach 1996). But in some areas forest and savanna may be better seen as alternative vegetation states influenced by multiple factors. Despite powerful environmental narratives to the contrary, there is strong evidence, in both the forest-savanna transition zone of Ghana and the coastal grasslands of the former Transkei in South Africa, that forest or woodland areas have been increasing over the century timescale as a result of a combination of disturbance events (Fairhead and Leach forthcoming for Ghana; Feeley 1987 for South Africa). Changes in soils, shifts in fallowing systems, manipulation of fire regimes, alterations in grazing patterns, and climatic rehumidification have combined to change the relationship between forests and grasslands. This dynamic interaction is thus less the outcome of a predictable pattern of linear succession, but more a result of combinations of contingent factors, conditioned by human intervention, sometimes the active outcome of management, and often the result of unintended consequences. With a view of ecology that stresses spatial and temporal variability, dynamic, nonequilibrial processes and histories of disturbance events, a different view of the landscape emerges: a landscape that is transforming, not simply degrading, and one which is emerging as a product of both social and ecological history, not simply the result of deterministic patterns of environmental change.

5 Environmental Entitlements

We have argued, then, that communities cannot be treated as static or undifferentiated, made up as they are of active individuals and groups. The environment, equally, needs to be disaggregated into its constituent parts, and viewed dynamically. We now go on to consider the implications of these dynamic, differentiated views for analysing the links between people and environment. Many analyses of people-environment relations, including those informing many community-based sustainable development narratives, conceive of a simple, linear relationship between population and resource availability, affected only by such factors as level of technology (cf. Ehrlich and Ehrlich 1991). There is little space left in this formulation for social actors and institutions. The concern with community organisations in many approaches to community-based sustainable development serves merely to modify this Malthusian model; a functional community is seen as playing an overall, regulatory role, controlling resource use and technology in such a way as to limit environmental impact in the interests of a collective good. But with environmental issues framed in terms of 'people in places' and as part of history, rather than simple, static conceptions of natural resource availability, a very different set of questions about people-environment interactions arise. We need to ask, for instance, which social actors see what components of variable and dynamic ecologies as resources at different times? How do different social actors gain access to and control over such resources? And how does natural-resource use by different social actors transform different components of the environment? To address such questions we draw on entitlements analysis, an approach first developed by Amartya Sen to explain how it is that people can starve in the midst of food plenty as a result of a

14 collapse in their means of command over food (Sen 1981). Undue emphasis on aggregate food availability, Sen argues, diverts attention from the more fundamental issue of how particular individuals and groups of people gain access to and control over food. Thus 'scarcity is the characteristic of people not having enough ... , it is not the characteristic of there not being enough. While the latter can be the cause of the former, it is one of many causes' (Sen 1981: 1). By analogy with this approach for explaining famine, we have argued that entitlements analysis is useful in explaining how the consequences of environmental change in general, and access to and control over natural resources in particular, are also socially differentiated (Leach and Mearns 1991; Mearns 1995b, 1996a). Just as with the food and famine debate, the environmental debate has been dominated by a supply-side focus, often giving rise to Malthusian interpretations of resource issues. But, as Sen observed, absolute lack of resources may be only one of a number of reasons for people not gaining access to the resources they need for sustaining livelihoods. It is important not to polarise this distinction excessively, however, as others have pointed out in the context of famine analysis: resource availability and access are often interconnected. Conflicts over access often intensify when the resources in question become scarce in absolute terms (Devereux 1988; Nolan 1993). Before exploring our own 'extended entitlements' approach, applied to environmental questions, it is worth returning to the origins of the approach in Sen's work, in order to clarify concepts and establish any key distinctions between the approach developed here and Sen's original work. In Sen's concept of entitlements, a number of central elements are stressed. First, Sen emphasised entitlements in the descriptive sense. The term 'entitlements' therefore does not refer to people's rights in a normative sense - what people should have - but the range of possibilities that people can have. In Sen's words, entitlements are 'the set of alternative commodity bundles that a person can command in a society using the totality of rights and opportunities that he or she faces' (Sen 1984: 497). While Sen was careful to specify that entitlements refer to 'alternative commodity bundles', in practice his principal concern was with command over food. Second, entitlements arise through a process of 'mapping': endowments, defined as a person's 'initial ownership' for instance of land or labour power, are transformed into a set of entitlements. According to Sen, entitlement mapping is 'the relation that specifies the set of exchange entitlements for each ownership bundle' (Sen 1981: 3). In Sen's work, these entitlement relations may be based on processes such as production, own-labour, trade, inheritance or transfer (Sen 1981: 2). Sen's concern was therefore to examine how different people derive entitlements from their endowments and so improve their wellbeing or capabilities, a descriptive approach to understanding how, in a given legal setting, people do or do not survive. Third, within this descriptive framework, Sen had a broader agenda within moral philosophy, which was to highlight the injustice of a legal system that can legally permit people to starve (Sen 1981). In order to emphasise this moral point, Sen has at times referred to 'entitlements' in a normative sense, and initially restricted the notion of entitlements to command over resources through formal legal arrangements, thus downplaying other, extralegal or informal means of

15 commanding resources which may become especially important during times of famine (Gore 1993). This is one of the sources of confusion within the literature on entitlements, on which Gasper (1993) has offered helpful clarifications. As a framework for analysing how different people gain access to and control over resources, we find Sen's descriptive notion of entitlements useful. However, along with others (Gasper 1993; Gore 1993; Devereux 1993, 1996; Jenkins 1997), we find some elements of Sen's framework too restrictive. First, at least in his early work, he focuses almost exclusively on entitlement mapping - how endowments are transformed into entitlements - and pays limited attention to endowment mapping - how people come to have their 'initial' endowments. Instead of assuming that endowments are simply given, our extended entitlements framework focuses on how both endowments and entitlements arise, a possibility Sen recognised in his later work with Jean Dreze (Dreze and Sen 1989: 23). In their use of an entitlements approach to analyse access to environmental justice in urban areas, Jenkins and colleagues emphasise another, nonmaterial dimension to endowments: namely, those institutional skills that enable people to access official decision-making structures and other public goods (Jenkins 1997; cf. Rein 1983). We return to this dimension of endowments below. Second, Sen is principally concerned with command over resources through market channels, backed up by formal-legal property rights. Although in later work (e.g. Sen 1984, 1985; Dreze and Sen 1989: 11), the idea of 'extended entitlements' is introduced, it is unclear to what extent the concept is restricted to mechanisms governing the infra-household distribution of resources, and to what extent it also includes other nonmarket channels of entitlement mapping such as public action. Either way, in our view, Sen's version of 'extended entitlements' does not go far enough. Since there are many ways of gaining access to and control over resources beyond the market, such as through kin networks, and many ways of legitimating such access and control outside the statutory legal system, including customary law, social conventions and norms, it seems appropriate to extend the entitlements framework to the whole range of socially sanctioned as well as formal-legal institutional mechanisms for gaining resource access and control (cf Gore 1993). Given these concerns, we adopt the following definitions of key terms4. First, endowments refer to the rights and resources that social actors have.' For example, land, labour, skills and so on.

4 These differ in certain respects from our earlier work on environmental entitlements (Leach and Meams 1991; Meams 1995b, 1996a), which did not effectively establish the distinction between endowments and entitlements (Gasper 1993).

' Devereux (1996) identifies units of analysis as one of two important sources of'fuzziness' in Sen's entitlements approach. For Sen the unit of analysis is held to be the individual, but the analysis may also apply at the collective level (e.g. household, group, class) by assuming a notional 'representative individual' (Osmani 1995: 254). Devereux is concerned that this device'sidesteps the reality that any group of people is composed of diverse individuals who are neither homogeneous nor "representative(Devereux 1996: 2). We adopt the 'social actor' as our unit of analysis, which both meets Sen's (and Osmani's) requirements and overcomes Devereux's concern. While 'social actor will often refer to an individual person, it could also refer to a group who share a certain set of characteristics (e.g. age, class, gender, caste) held to be important for the particular entitlement mapping in question. The empirical contexts with which we are concerned, like Devereux's, are characterised by property rights that overlap and are contested. (Devereux identifies this as the second source of fuzziness in the entitlements approach.) The use of 'social actor' as the unit of analysis allows for multiple, complex resource claims to be separately specified to the level of actors who share similar bundles of claims without necessarily moving to the level of each individual person. It is not necessary for analytical purposes to disaggregate to all possible degrees of difference among social actors in order to demonstrate the importance of particular forms of entitlement mapping for different social actors.

16 Second, entitlements, following Gasper (1993), refer to legitimate effective command over alternative commodity bundles. More specifically, environmental entitlements refer to alternative sets of utilities derived from environmental goods and services over which social actors have legitimate effective command and which are instrumental in achieving wellbeing.6 The alternative set of utilities that comprise environmental entitlements may include any or all of the following: direct uses in the form of commodities, such as food, water, or fuel; the market value of such resources, or of rights to them; and the utilities derived from environmental services, such as pollution sinks or properties of the hydrological cycle. Entitlements, in turn, enhance people's capabilities, which are what people can do or be with their entitlements. For example, command over fuel resources derived from rights over trees gives warmth or the ability to cook, and so contributes to wellbeing. Consider the example of a cow, owned by a particular social actor7. The cow is that actor's endowment. Alternative sets of utilities that could potentially derive from that cow include various combinations of the following: milk, blood or meat for food to feed the social actor and her family members; cash income from the sale of the live animal, or from its meat, hide, tail and other products; future breeding potential, representing a form of interest-earning savings; or a contribution to the brideprice of a family member. The cash income, food (stock of meat or flow of milk or blood), 'interest' (in the form of future calves) and social value constitute the environmental entitlements accruing to that social actor as a function of her initial endowment of a cow. Those entitlements in turn provide the actor with various capabilities, depending on the actual set of utilities derived, including nutrition for herself and her family, social status, or the ability to satisfy other basic needs through the market. However, since endowments and entitlements are analytical constructs, the distinction between them depends on the empirical context and on time. Environmental entitlements analysis thus reflects a cyclical rather than a linear process, at different time scales. For a given social actor at a particular time period, entitlements represent the set of potential outcomes given the initial endowments set, depending on the actual constraints that that actor faces and the opportunities open to her. Those entitlements may in turn represent endowments at another time period, from which a new set of entitlements may be derived. There is nothing inherent in a particular environmental good or service that makes it a priori either an endowment or an entitlement. In the above example, the cow was the endowment and the alternative set of utilities derived from the cow were the entitlements. At an earlier time period, the cow might form part of the environmental entitlement set of the social actor,

6 For the purposes of our analysis, we adopt the following working definitions, based on teach and Meams (1991). Environmental goods refer to the specific source (material and energy natural-resource) inputs that are essential to sustaining the livelihoods of present and future generations of people. Environmental services refer to sink (pollution-absorbing) and other service functions of the environment (e.g. the hydrological cycle) that are also essential to sustaining the livelihoods of present and future generations of people. Environmental goods are resources in the sense that they are 'materials available "in nature" that are capable of being transformed into things of utility to man [sic]' (Harvey 1979: 178). As Harvey makes clear, resources can only be defined in relational terms, and are a function of a knowledge and technology within a given cultural context.

7 We are grateful to Stephen Devereux for suggesting this generic example.

17 where access to a village commons is one of the particular endowments of interest. The set of utilities from that commons might include fuelwood, medicinal plants and wild foods as well as grazing. In practice, the social actor may or may not effectively be able to exercise her rights of access to the commons. She may have those rights in principle, by virtue of her membership in a particular residence-based group, but fail to exercise them in practice because owing to the prevailing gender division of labour, or the ideology of seclusion among women of her caste, she is unable to move about publicly to take the cow to pasture. Alternatively, she may find her rights of access to common grazing are not translated into nourishment for the cow, because other, more powerful actors (e.g. a politically well-connected farmer) have encroached on and enclosed a portion of the commons for crop production, thus reducing the availability of forage. It will be seen from this example that the phrase 'legitimate, effective command' refers to many dimensions of entitlement mapping which are far from simple. The central difference between endowments and entitlements, following Gasper (1993), is that endowments are the rights and resources social actors have in principle, while the entitlements derived from them are what social actors actually get in practice. An emphasis on the 'effectiveness' or otherwise of command over resources highlights two issues. First, resource claims are often contested, and within existing power relations some actors' claims are likely to prevail over those of others. Second, certain social actors may not be able to mobilise some endowments (e.g. capital, labour) that are necessary in order to make effective use of others (e.g. land). For instance, kinship-based institutions that regulate command over labour may embody power relations structured around gender and age, that leave young men, and especially young women, strongly disadvantaged in their ability to control their own labour and to call on that of others. By 'legitimate' we refer not only to command sanctioned by a statutory system but also to command sanctioned by customary rights of access, use or control, and other social norms. In some cases, these sources of legitimacy might conflict, and different actors may espouse different views of the legitimacy or otherwise of a given activity. In the example given above, the well-connected farmer and the female herder may call on different but, in their eyes, equally valid sources of legitimacy to justify their respective claims to the same resource, but the female herder is prevented by her weak position within prevailing power relations from realising her claim. In our South Africa case study, communities surrounding Mkambati Nature Reserve, a government-owned protected area on the Eastern Cape's Wild Coast, are legally prevented from hunting game within the reserve. Neverthless, groups of young men, with active encouragement from local civic organisations and/or the tacit blessing of their local chief, depending on his current political stance vis-a-vis the local authorities, regularly hunt within the reserve. They justify their actions by calling on customary rights locally referred to as ukujola, which are based on historical claims predating the gazetting of the protected area, and which amount to legitimised poaching.8

8 Jenkins (1997) includes a very useful discussion on questions of illegality and enforceability in the context of entitlement mapping processes, where forms of legitimacy for contested resource claims come into conflict with one another. For our purposes, we maintain the

18 Figure 1 links together in diagrammatic form those elements of our environmental entitlements framework discussed so far. An undifferentiated 'environment' has been replaced by one that is disaggregated into particular environmental goods and services. Their distribution, quality and quantity are influenced by ecological dynamics (see section 4 above) which are in part shaped by human action (discussed in section 7 below). The relationship between a given 'community', made up of differentiated social actors (see section 3 above), and the changing ecological landscape, can be analysed in terms of the ways different social actors gain capabilities, or a sense of wellbeing, by acquiring legitimate, effective command over resources through processes of endowment and entitlement mapping. Endowments for particular social actors are distinct from environmental goods and services (which are given 'in nature'), and so lie outside the ellipse in Figure 1 that represents environmental goods and services. By contrast, capabilities are attributes of particular social actors, and so are included within rather than lying outside the ellipse representing differentiated social actors. Our analysis has led us away from a focus on the particular endowments, entitlements and capabilities of a given social actor at a given moment, since these represent only a snapshot in time. Instead, our analysis focuses principally on the dynamic mapping processes that underlie each of these static sets, which are mediated by various forms of institutions (appearing to the right of Figure 1) operating at a range of scale levels from the macro to the micro. Section 6 examines in more detail what we mean by institutions, since they are central to our framework. The relationships among these institutions and between scale levels is of central importance in influencing which social actors - both those within the community and those at some considerable remove from it - gain access to and control over local resources. In turn they influence the uses to which resources are put and the ways they are managed, and thus progressively help to modify and shape the landscape over time (see section 7). The environmental entitlements framework therefore links both the macro and the micro levels of concern. It situates 'a disaggregated (or "micro") analysis of the distinctive positions and vulnerabilities of particular [social actors] in relation to the "macro" structural conditions of the prevalent political economy' (Jenkins 1997: 2). In various ways our three case studies illustrate the interactions among institutions at different scale levels, and the ways they circumscribe the resource claims and management practices of different social actors. At the international level, for example, the policies of donor agencies play an important role not only in directly shaping local approaches to community-based natural resource management, but also in influencing domestic macroeconomic policy or governance in ways that cascade down to affect local natural resource management. At national or state level, government policies and legislation are of primary interest, including land tenure reform policies, or approaches to forestry and wildlife conservation and tourism. At progressively more local levels these intersect with rural livelihood systems, intra-household dynamics, and so on. But the interrelationships practical distinction that while endowments may include various formal and informal rights over resources that social actors have in principle, entitlements refer to the sets of utilities they are able to derive in practice.

19 Figure 1: The environmental entitlements framework

ENVIRONMENTAL GOODS & SERVICES

Macro INSTITUTIONS IMeso Micro

ENDOWMENTS

Macro INSTITUTIONS Meso Micro

ENTITLEMENTS

Macro INSTITUTIONS IMeso Micro

CAPABILITIES

DIFFERENTIATED SOCIAL ACTORS Figure 2: Endowments- and entitlements-mapping in the case of Marantaceae leaf collection, southern Ghana

Marantaceae sp., occurring in disturbed forest sites, abandoned cocoa farms, etc.

Macro Government Forest Department policies (e.g. permit system for NTFPs) Meso - international cocoa prices Micro

customary property rights (village membership, land ownership) rights over Marantaceae leaves

gender division of labour 'queen mother of leaf gatherers' (customary conflict resolution mechanism) - cooperative work groups

Marantaceae leaves for wrapping food, or cash income from their sale

Micro hh- intra- household bargaining V"I arrangements

contributions to well-being

Women leaf gatherers, other family members between scale levels are far from deterministic. Land claims at local level may spill over into national, state or provincial-level politics, for example, and influence the direction of future policy and the scope of legally enforceable rights. To illustrate some of the types of endowment and entitlement mapping processes that are of interest in our case studies, consider the following, simplified example from Ghana. The particular sets of endowments, entitlements and capabilities discussed, and the relevant mediating institutions, are summarised in Figure 2. In southern Ghana, the leaves of Marantaceae plants are commonly collected by women and used and sold widely for wrapping food, kola nuts and other products (cf Falconer 1990; Agyemang 1996). The leaves are associated with particular sites and times within dynamic, variable forest and forest-savanna ecology. These include disturbed forest sites, moderately burnt forest, swamps, and abandoned cocoa farms and fallows, especially during the rainy season. The leaves become endowments - people gain rights over them - in different ways depending on whether they lie inside or outside government-reserved forest. Off reserve, the leaves are usually the common property of a village, with an actor's endowment mapping depending on village membership. Where they occur on farmland, collection rights are acquired through membership of, or negotiation with, the appropriate landholding family or farm household. On reserve, endowment mapping depends on the Forest Department's permit system, with women often using established trading relationships as a source of finance for permits. Without such a permit, leaf gathering is illegitimate from the state's perspective, although it may be sanctioned by customary tenure arrangements grounded in different definitions of reserved land as ancestral farmland. The set of entitlements derived from Marantaceae leaves may include direct use of the leaves or their sale for cash income. In practice, most women involved in gathering leaves prefer to sell them as an important source of seasonal income. In entitlements mapping, both labour and marketing issues are important. Women may have to negotiate with their husbands and co-wives - in relation to other farm work and domestic duties - for labour time to collect the leaves. They find leaf gathering in groups more effective, so collection depends on membership of a regular group or on impromptu arrangements among kin and friends. There is frequently competition between groups for the best sites, as well as competition for leaves among group members. When disputes arise, whether between individual women, collection groups, or with forestry officials, a 'queen mother of leaf gatherers' - appointed by each village or neighbourhood's women gatherers - helps to mediate them. Marketing effectively depends on establishing a regular relationship with village-based or visiting traders who will guarantee a reasonable price even at times of year when the market is flooded. Women frequently invest actively in maintaining such relationships, for instance collecting one type of leaf for one buyer, and another type of leaf for another buyer. The utilities derived from the cash sale of Marantaceae leaves contribute to a woman's capability to ensure that she and her children are well-fed and to satisfy other cash-dependent basic needs. In particular, the leaves offer a timely source of rainy season income when money is otherwise scarce. But whether a woman can keep control of the income, and how it is used, depends on intra-

22 household bargaining arrangements, such as negotiations with husbands and co-wives over expenditure priorities and responsibilities for providing food. Asking which combination of institutions makes the most difference to resource access and control for a set of social actors, or for the dynamics of resource use and management surrounding a particularly valued element of the landscape, may offer innovative practical insights. These are often overlooked in simplified analyses of people-environment interactions, which tend to concentrate on questions of per capita resource availability and ignore issues around different social actors' resource access. Exploring the institutional dimensions of resource use opens up a range of important issues for management and policy which are not addressed as part of conventional analyses based on resource availability alone. For example, in the Indian case study of a watershed development project, the NGO Seva Mandir had been working in the seven hamlets that make up the focal 'community' for at least eight years before it attempted any activities related to natural-resource management. In the earlier phase of its involvement in Nayakheda, Seva Mandir focused attention on adult literacy, maternal and child health care, leadership training and 'institutional strengthening', which principally referred to the creation of a village committee to oversee the implementation of project activities. These initiatives all aimed directly to build up capabilities among the relevant sets of social actors who benefited from these activities, to give them voice in local political struggles and, in time, to build 'social capital'.9 Although probably not intended at the time Seva Mandir began work in Nayakheda, since watershed development activities were not planned at that time, these investments in capabilities and social capital appear to have been instrumentally important in protecting and promoting the endowments and environmental entitlements of particular social actors. Thus, as illustrated by the feedback arrows in Figure 1, there may be potential for the enhanced capabilities of particular social actors to protect and promote both their endowments and their entitlements; this process might be accelerated by means of purposive action through projects and public policies. An institutional focus also highlights relations of power in the mapping processes. Such issues are notably absent in Sen's analysis (Watts 1991), suggesting the need to examine the degree to which different people can influence decisions about endowments and entitlements (cf Appadurai 1984). An extended entitlements approach therefore sees entitlements as the outcome of negotiations among social actors, involving power relationships and debates over meaning (cf Gore 1993: 452), rather than as simply the result of fixed, moral rules encoded in law. A further important addition to Sen's analysis offered by our institutional focus is the introduction of a dynamic, historical perspective, over different time scales. Mapping processes are not static; indeed the various elements of the framework as set out so far continuously change over time. In the process of actors gaining legitimate, effective command over a resource bundle, negotiations over labour or land may take place which in turn transform the nature of certain actors'

9 Ajay Mehta. Director of Seva Mandir, personal communication, Udaipur. August 1996.

23 land or labour rights. Over longer time frames, a process of commoditisation of certain resources might serve to increase the role of the market as a key institution in endowment and entitlement mapping. This dynamic and historical perspective therefore informs our interpretation of endowment and entitlement mapping processes and, in turn, our approach to the institutional analysis of environmental change, a subject to which we now turn.

6 Institutions

Since institutional arrangements shape the processes of endowment and entitlement mapping, the way we understand institutions and institutional change is central to our analytical framework. For example, several different kinds of both formal and informal institution emerged as being important in mediating access to and control over Marantaceae leaves in southern Ghana. Consider the case of property rights, which serve to allocate endowments to different individuals. On the one hand there are formal property rights, such as permits to gather non-timber forest products. These are issued by the Forest Department, legitimised by the state, and in principle could be defended in courts of law. On the other hand there are also informal or customary property rights, legitimised by social norms and codes of behaviour. These are legitimate in the eyes of those local resource claimants who regard government-reserved land as ancestral farmland, but illegitimate in the eyes of the state. Another informal institution, the authority vested in the queen mother of leaf gatherers, helps mediate in disputes between forestry officials and local resource claimants on forest-reserve land. The question of entitlement mapping, or who ultimately gets effective command over Marantaceae leaves, is influenced by the interplay of other formal and informal institutions, including the gender division of labour, cooperative work groups, and trading networks. So, how should we conceptualise institutions as part of the environmental entitlements framework? Several key themes emerge from recent work on institutions across a range of disciplines, and here we attempt to clarify our position in relation to these. First, we need to consider how institutions are to be defined. Many analysts in the fields of new institutional economics, new economic history and public choice theory view institutions as rules (Bromley 1989; North 1990; Ostrom 1986). This helps to distinguish institutions from organisations. If institutions are 'the rules of the game in society' (North 1990: 3), then organisations may be thought of as the players, or 'groups of individuals bound together by some common purpose to achieve objectives' (North 1990: 5). Some argue that organisations, such as schools, NGOs and banks, exist only because there is a set of 'working rules' or underlying institutions that define and give those organisations meaning (Bromley 1989; Commons 1968). This view is contested, however. Certain organisational forms may persist

long after the 'working rules' that may initially have given them meaning cease to be relevant (Powell and DiMaggio 1991). Many institutions have no single or direct organisational manifestation,

including money, markets, marriage, and the law, yet may be critical in endowment and entitlement

mapping processes. Viewing institutions merely as rules, however, raises other problems to which we return below.

24 In addition to a clarification of the institution-organisation distinction, perhaps the most enduring contribution of the new institutional economics is the focus on transaction costs as an important factor underlying institutional change. In this respect, while it may not offer a grand theory, the new institutional economics does yield insights that are useful in suggesting hypotheses to guide empirical research (Harriss et al 1995). For example, the high costs to the Rajasthani State Forest Department of overseeing and enforcing regulated access to state forest land in the Aravalli Hills in India has, it is argued, led to high levels of commercial exploitation and subsequent deforestation, suggesting that other types of institutional mechanism, with lower transaction costs, would be more appropriate if maintaining forest cover was a major objective. Similarly, in the former Transkei, South Africa, the type of tenure regime associated with different types of grazing can be related to the relative costs and benefits of managing exclusion. In high-value grazing sites, institutional forms with relatively high transaction costs may persist, while for low-value, highly variable grazing resources the opposite is most likely (cf Scoones forthcoming). However, some definitions of institutions derived from a transaction costs approach can be criticised for being tautologous and functionalist. At the extreme, this results in definitions along the following lines, as paraphrased by Harriss et al (1995: 7): 'existing institutions minimise transaction costs because transaction cost minimisation is their function'. The definition of institutions adopted by de Janvry et al (1993: 566) - 'complexes of norms, rules and behaviors that serve a collective purpose' - is problematic on account of both its functionalism and the tendency to lump together norms, rules and behaviour. In order to move beyond such problems to a more complete understanding of the dynamics of institutional change and the changing patterns of endowment and entitlement mapping, we must distinguish between institutional 'rules' and people's behaviour or practices, and clarify the relationship between them. Different bodies of work on institutions take different positions on this issue. Much new institutional economics, like earlier structural-functionalist anthropology, assumes that rules and behaviour are synonymous and can be reduced to one other (e.g. de Janvry et al 1993); a position we criticised in section 3 for denying people's conscious action and agency. Nevertheless, some of the work that defines institutions as rules recognises that such rules do not automatically determine people's behaviour. Ostrom (1986), for instance, argues that rules prescribe room for manoeuvre; behaviour is rule-bound rather than rile-determined. Others point to situations in which behaviour appears to contravene the rules. In an insightful critique of Sen's narrow view of the rules of entitlement, Gore (1993) refers to such behaviour as 'unruly' social practices, citing Thompson (1971); Scott (1976, 1985) and others on the ways that different forms of protest and resistance challenge legal rules governing people's ability to gain command over commodities. But such 'unruly' practices may well be bound by different sets of moral/informal rules (Gore 1993: 446). Such situations are those in which notions of legitimacy are contested, and therefore actual entitlements are influenced by the interplay of these competing rule sets in the context of prevailing power relations. Such an approach recognises that the law necessarily operates within a particular social context,

25 whereby, for example, the judiciary is able to bend the rule of law to favour selective class, gender or ethnic interests, particularly in weak states. The distinction between formal and informal institutions is highlighted by these analyses. Formal institutions may be thought of as rules that require exogenous enforcement by a third party organisation. The rule of law is an example, usually upheld by the state through such organisational means as law courts, prisons and so on. Informal institutions, however, may be endogenously enforced; they are upheld by mutual agreement among the social actors involved, or by relations of power and authority between them. Recent work on institutions stresses the socially 'embedded' nature of informal institutions, or the multiplicity of institutional relations in which people are engaged at any one time (Runge 1986; Mearns 1996b; Swallow 1996; White 1996). In short, institutions of various kinds, ranging from the informal (e.g. social norms) to the formal (e.g. the rule of law), interlock to form a matrix within which people live their lives. As argued in the burgeoning literature on social capital, trust and networks of civic engagement (e.g. Gambetta 1988, Putnam et al 1993; Stewart 1996; Humphrey and Schmitz 1996; Mearns 1996c; Dasgupta 1996), multiple involvement may promote mutual assurance among social actors, promoting cooperation and collective action. Yet this argument potentially neglects issues of power relations and the very different meanings that different institutions may carry for different actors. As Bates (1995) has emphasised, there is a need to ground institutional analysis in a theory of power. While this is currently lacking in much new institutional economics (but see Bowles and Gintis [1993] for a notable exception in the economics literature), it is strongly present in other strands of work, for instance in feminist analysis of institutions (e.g. Kabeer and Subrahmanian 1996; Goetz 1996). Many institutions patently do not serve a collective purpose, even if they may once have done, and different actors' perceptions of the 'collective good' depend very much on their social position. Equally, involvement in some groups may be a response to inequities in others. A view of institutions as simply coexisting in benign complementarity may be misleading. Women's investment in resource-sharing networks with neighbours, for instance, may relate to their lack of power within the household. Our analyses of institutions mediating environmental endowment and entitlement mapping processes within the case study settings leads us to view institutions not as the rules themselves, but as regularised patterns of behaviour that emerge from underlying structures or sets of 'rules in use'. While some new institutionalists adopt this perspective (e.g. Schotter 1981), it is more commonly associated with sociological and anthropological approaches (e.g. Giddens 1984). Such an approach

blurs the distinction between rule and practice. Rather than existing as a fixed framework, 'rules' are constantly made and remade through people's practices. This is a perspective developed in anthropological literature on customary law (Chanock 1985), and in the work of Berry (1989, 1993) who sees institutions as maintained by (and only existing because of) people's active 'investment' in them. Regularised practices, performed over time, eventually constitute institutions. Yet, as they consciously monitor the consequences of past behaviour and the actions of others, different social

26 actors may choose - or be forced - to act in irregular ways. Over time, perhaps as others similarly alter their behaviour, institutional change may occur. But owing to the embeddedness of informal institutions, institutional change in society may be a slow, 'path-dependent' (cf North 1990) process, even if formal institutions, such as legal frameworks, macroeconomic policies or political regimes, change quickly (Mearns 1996b). In order to understand the range of formal and informal institutions that mediate processes of endowment and entitlement mapping, an understanding of 'rules in use' is therefore required, where different actors' regularised, everyday practices are central to any analysis. This necessitates an actor- oriented approach to understanding institutions (cf Long and van der Ploeg 1994; Nuitjen 1992), one that takes seriously an analysis of difference and an appreciation of power relations. Our framework therefore seeks to investigate the embedded nature of formal and informal institutions, exploring institutional change in a historical perspective.

7 Structuration of the Environment

So far, the presentation of our framework has focused on how social actors access, use and benefit from different components of the environment, and how an extended entitlements approach, coupled with an analysis of institutions, helps to conceptualise these processes. But a key set of questions remains, concerning the ways in which the environment is, in turn, shaped and transformed through people's interactions with it (see the 'feedback loop' on the left in Figure 1). In debates around global environmental change, the degree to which particular landscapes have been shaped by their inhabitants as well as by processes operating at some remove is often underrecognised. As we have argued, community-based sustainable development narratives tend to conceive of local environments in terms of a baseline 'natural balance' or optimum which human intervention disturbs or degrades over time, but may be called upon to restore. Our discussion of new thinking in ecology (section 4) criticised this perspective, arguing for a different view in which environments are constantly transforming and emerging as the outcome of dynamic and variable ecological processes and disturbance events, in constant interaction with human use. In other words, environmental conditions at any given time can be seen as the product of both ecological and social history. Sections 5 and 6 explained the social history of environmental use in terms of institutional dynamics which shape the changing ways that differentiated social actors perceive, use and manage components of the environment as resources. Some of the ideas in structuration theory, which we have already addressed in relation to social and institutional dynamics, add significantly to our understanding of people-environment interactions (cf. Redclift and Woodgate 1994).

Seen in this way, the environment provides a setting for social action but is also a product of such action. People's actions and practices, performed within certain institutional contexts, may serve to conserve or reproduce existing ecological features or processes (e.g. maintain a regular cycle of fallow growth or protect the existing state of a watershed and its hydrological functions). But people may also act as agents who transform environments (e.g. shorten the fallow, altering soils and

27 vegetation, or plant trees in a watershed). Such agency may involve precipitating transitions of ecological state that push ecological processes in new directions or along new pathways. While some actions may be intentional, constituting directed management aimed at particular goals or transformations, others may be unintentional, yet still have significant ecological consequences. Over time, the course of environmental change may be strongly influenced by particular conjunctures of institutional conditions, or by the coming together of contingent events and actions. Practices and actions carried out at one time - under one set of institutional arrangements - may leave a legacy that influences the resources available to become endowments for actors at some future time. For instance, the farming practices of one group of people may enduringly alter soil conditions such that subsequent inhabitants may make use of these in their farming of different crops, whether or not they acknowledge this as the legacy of past farmers. Equally, past actions influence the possibilities for agency open to subsequent actors. As present practices build on the legacies of past ones, so the causality of environmental change may need to be seen as cumulative, sequential or path-dependent. The concept of landscape serves usefully to encompass these linked ecological and institutional dynamics, with landscape history referring to the reflexive relationship between environmental and social history. In this sense, our perspective articulates with a large literature that has explored the idea of landscape in cultural geography (e.g. Sauer 1925; Hoskins 1955; Glacken 1967; Cosgrove 1984; Duncan and Ley 1993) and in social anthropology (e.g. Bender 1993; Guyer and Richards 1996). Considering how regional, national or international processes articulate with local ones to shape landscapes over an historical time frame is also a shared feature of approaches in what has come to be termed political ecology (e.g. Blaikie and Brookfield 1987; Bassett 1988; Bryant 1992; Moore 1993; Peet and Watts 1996). However, with some exceptions (e.g. Rocheleau et al 1996) these have given rather less attention to issues of intra-community social difference. Equally, they have paid rather little attention to recent theoretical perspectives in ecology (but see Fairhead and Leach 1996; Zimmerer 1994). Landscapes, over time, then, may come to embody layer upon layer of the legacies of former institutional arrangements, and of the changing environmental entitlements of socially differentiated actors. This can be illustrated for the Ghana and South Africa case studies by the resource- and biodiversity-rich forest patches that are distinctive features of both landscapes (see Fairhead and Leach 1996 for other cases). Contrary to strong narratives suggesting a decline in forest cover, it seems that over the twentieth century tree cover has increased on specific sites within the settled landscapes of both the South African and Ghanaian case study areas. Shifts from grassland to wooded area have resulted from diverse interactions over time between social actors and ecological processes, influenced by institutions. Around settlement sites, for instance, tree growth has been encouraged by localised improvements in soil fertility resulting from inhabitants' everyday activities (gardening, waste deposition etcetera), by livestock management practices and by fire protection regimes such as early-burning. The form and composition of such woodland patches has been influenced by

28 management for tree crops (e.g. cocoa in the Ghana case), and by enrichment planting with indigenous and exotic species valued by particular social actors. Such management patterns have altered over time in response to changing market opportunities, labour relations and different social actors' priorities. Frequently, forest patches have endured even when settlements have been

abandoned, for instance as a result of migration or settlement consolidation, and at least in the Ghana

case they are sometimes actively preserved as burial and ancestral worship sites by the descendants of settlement founders. Understanding such changes in vegetation therefore requires insights into institutions influencing settlement patterns, labour organisation, fire and grazing management, tree product marketing and so on, and into the ecological legacies left by actions under each set of institutional arrangements for people's subsequent resource use and management.

8 Conclusions

In focusing attention on the mapping processes by which components of heterogeneous environments become endowments and entitlements of particular social actors, the framework outlined in this paper

has attempted to provide a dynamic perspective on the role of institutions in people-environment relations. Diverse institutions, both formal and informal, and often acting in combination, shape the ways in which differentiated actors access, use and derive wellbeing from environmental resources and services and, in so doing, influence the course of ecological change. As people interact with each other and with the environment in the context of these mapping processes, their actions may, over time, serve to reproduce particular institutions, but they may also serve to alter them, and thus to push institutionally influenced ecological dynamics along new pathways. By seeing people-environment relations in this way, the environmental entitlements framework offers some fundamental challenges to the ways in which 'community', 'environment' and the links between them are commonly portrayed in the policy narratives surrounding community-based

sustainable development. In so doing, it raises a number of implications for development planning and practice. By highlighting those institutional arrangements that support particular endowments and entitlements, the framework can help target particular institutions for external support, as a corrective to the current, narrow emphasis on community-level organisations. In particular, the entitlements approach can help to reveal institutional arrangements - including those of an informal nature - that are of central importance to the wellbeing of those social groups often marginalised by so-called community-level interventions. The framework also can help highlight which events, contingencies and institutional combinations contribute to particular types of ecological transformation, thus assisting a more focused approach to planned interventions intended to influence the course of environmental change. While much of our analysis has addressed local-level institutional dynamics, power relations and the like among sets of social actors that make up local communities, this does not mean that it is only at the local level that appropriate forms of external intervention may be targeted. On the contrary, it is often the interaction of macro- and micro-level institutions that influences the

29 entitlement mappings of particular social actors, or particular trajectories of environmental change. A valid criticism of many current approaches to community-based sustainable development is that they are heavily project-focused, and may not be generalisable. While an understanding of institutional dynamics requires detailed analysis at local level, more effective forms of public intervention are likely to require attention to policy frameworks rather than to individual projects. Through the comparative analysis of carefully selected, representative case studies, and by drawing attention to the ways micro and macro institutional variables interact, the environmental entitlements framework could inform a view of where policy initiatives are likely to yield most promising results, provided the objectives for public intervention are clearly specified at the outset. On the other hand, the framework may also prove useful to those interested in social impact evaluation of particular projects or discrete policy initiatives. Our India case study is an example of just such an application of the framework. More broadly, and as illustrated here for our three case study areas, the framework emphasises the importance of social difference and of ecological variability and dynamics in people-environment relations, and of the history of these relations in shaping current and future possibilities. Difference, dynamics and history are frequently ignored in development analysis and planning, yet, we would argue, should be brought much more forcefully to the fore. From this perspective, it becomes difficult, if not impossible, to identify an 'environmental ideal' to which community-based sustainable development should aspire: a more natural vegetation, for instance, or a more productive watershed, defined only in aggregate terms. Instead, the focus shifts to recognise that different groups of actors may give priority to different environmental resources or services, and that particular trajectories of landscape change will bring a different distribution of costs and benefits to different groups of people. In this context, collective choice and consensus, as emphasised in current approaches to community-based sustainable development, come to seem illusory goals. Landscape change is a fundamentally political process, involving negotiation and conflicts between actors with different priorities who are differently positioned in relations of power. Development interventions that seek to influence environments similarly need to take an approach in which plural perspectives are rendered explicit and outcomes are negotiated through processes that may well be prone to conflict. This may seem an uncomfortable conclusion to reach. However, making more explicit the political nature of landscape change and the varying environmental priorities of different social actors need not create a vacuum for public policy and action. The environmental entitlements

framework is offered first as an aid to understanding, to be used descriptively. Given carefully specified objectives, it may then be applied normatively. While this may appear rather top-down, in that it seems to imply that policy makers can and should judge between 'good' and 'bad' institutions, current approaches to community-based sustainable development may also reinforce socially or even ecologically undesirable outcomes by operating under false assumptions. Perhaps the most important conclusion is that much more broadly based citizen participation in processes of environmental problem-framing in the first place, and thereby in defining and negotiating particular solutions

30 (Gaventa 1995; Renn et al 1994), is more likely to lead to processes worthy of the epithet 'sustainable development'.

31 References

Abrams, P., 1982, Historical Sociology, Shepton Mallet: Open Books

Adams, J.S., and T.O. McShane, 1992, The Myth of Wild Africa: Conservation Without Illusion. Berkeley: University of California Press

Agyemang, M., 1996, 'The leaf gatherers of Kwapanin, Ghana', Forest Participation Series No 1, London: Forestry and Land Use Programme, IIED

Allen, T., and T. Hoekstra (eds), 1991, Ecological Heterogeneity, Berlin: Springer-Verlag

Appadurai, A., 1984, 'How moral is South Asia's economy?', Journal of Asian Studies 43: 481-97

Apthorpe, R., and D. Gasper (eds), 1996, Arguing Development Policy: Frames and Discourses, London: Frank Cass

Arnstein, S., 1969, 'A ladder of citizen participation', Journal of the American Institute of Planners 35: 216-24

Bailey, F.G., 1969, Stratagems and Spoils: A Social Anthropology of Politics, Oxford: Basil Blackwell

Baland, J.-M., and J.-P. Platteau, 1996, Halting Degradation of Natural Resources: Is There a Role for Rural Communities?, Oxford: Clarendon Press for FAO

Barth, F., 1992, 'Towards greater naturalism in conceptualising societies' in A. Kuper, (ed), Conceptualising Society, London: Routledge

Bassett, T.J., 1988, 'The political ecology of peasant-herder conflicts in the northern Ivory Coast', Annals of the Association of American Geographers 78 (3): 453-72

Bates, R.H., 1995, 'Social dilemmas and rational individuals: an assessment of the new institutionalism', in J. Harris, J. Hunter and C.M. Lewis (eds), The New Institutional Economics and Third World Development, London: Routledge: 27-48

Bebbington, A., 1994, 'Theory and relevance in indigenous agriculture: knowledge, agency and organisation', in D. Booth, (ed) Rethinking Social Development: Theory, Research and Practice, Harlow: Longman: 202-225

Bender, B., 1993, Landscape: Politics and Perspectives, Providence and Oxford: Berg Press

Berkes, F., 1995, 'Community-based management and co-management as tools for empowerment', in N. Singh and V. Titi (eds), Empowerment: Towards Sustainable Development, London: Zed Books: 138-146

Berry, S., 1989 'Social institutions and access to resources', Africa 59 (1): 41-55

Berry, S., 1993, No Condition is Permanent: The Social Dynamics of Agrarian Change in Sub- Saharan Africa, Madison: University of Wisconsin Press

Blaikie, P., and H. Brookfield, 1987, Land Degradation and Society, London and New York: Methuen

32 Borrini-Feyerabend, G., 1996, Collaborative Management of Protected Areas: Tailoring the Approach to the Context, Gland, Switzerland: IUCN

Botkin, D., 1990, Discordant Harmonies: A New Ecology for the Twenty-First Century, Oxford: Oxford University Press

Bowles, S., and Gintis, H., 1993, 'The revenge of homo economicus: contested exchange and the revival of political economy', Journal of Economic Perspectives 7 (10): 83-102

Bromley, D.W., 1989, 'Institutional change and economic efficiency', Journal of Economic Issues 23(3):735-59

Bryant, R.L., 1992, 'Political ecology: an emerging research agenda in Third World studies', Political Geographical Quarterly. 11 (1): 12-36

Bryant, C., and D. Jary 1991 'Introduction: coming to terms with Anthony Giddens', in C. Bryant, and D. Jary (eds), Giddens' Theory of Structuration: A Critical Appreciation, London: Routledge: 1-31

Carney, J., and M. Watts. 1991, 'Manufacturing dissent: work, gender and the politics of meaning in a peasant society', Africa 60 (2): 207-41

Carruthers, J., 1995, The Kruger National Park: A Social and Political History, Pietermaritzburg: University of Natal Press

Caughley, G., 1977 The Analysis of Vertebrate Populations: Chichester: John Wiley and Sons

Chanock, M., 1985, Law, Custom and Social Order: The Colonial Experience in Malawi and Zambia, Cambridge: Cambridge University Press

Cherrett, J. (ed), 1987, Ecological Concepts: The Contribution of Ecology to an Understanding of the Natural World, Oxford: Basil Blackwell

Clements, F., 1916, 'Plant succession: an analysis of the development of vegetation', Carnegie Institute Publications No 242: 1-512

Commons, J.R., 1968, Legal Foundations of Capitalism, Madison: University of Wisconsin Press (first published 1924)

Connell, J., and W. Sousa., 1983, 'On the evidence needed to judge ecological stability or persistence', American Naturalist 121: 789-824

Conroy, C., and M. Litvinoff (eds), 1988, The Greening of Aid: Sustainable Livelihoods in Practice, London: Earthscan Publications and LIED

Cosgrove, D.E., 1984, Social Formations and Symbolic Landscape, London and Sydney: Croom Helm

Dasgupta, P., 1996, 'Trust, social capital and economic analysis'. Paper presented at the Fourth World Bank Conference on Environmentally Sustainable Development, Washington DC, 25-27 September

33 de Janvry, A., E. Sadoulet, and E. Thorbecke 1993, 'State, market and civil organizations: new theories, new practices, and their implications for rural development', World Development 21 (4): 565-75

Devereux, S., 1988, 'Entitlements, availability and famine: a revisionist view of Wollo, 1972-74', Food Policy 13 (3): 270-82

Devereux, S., 1993, Theories of Famine Hemel Hempstead: Harvester Wheatsheaf

Devereux, S., 1996 'Fuzzy entitlements and common property resources: struggles over rights to communal land in Namibia', IDS Working Paper No 44, Brighton: IDS

Dreze, J., and A. Sen. 1989, Hunger and Public Action, WIDER Studies in Development Economics Series. Oxford: Clarendon Press

Duncan, J., and D. Ley (eds), 1993, Place/Culture/Representation, London: Routledge

Dunning, J., and D. Stewart 1995, 'Spatially explicit population models: current forms and future uses', Ecological Applications 5 (1): 3-11

Durkheim, E., 1895, The Rules of Sociological Method, Chicago: Free Press (edition published 1965)

The Ecologist, 1972'Blueprint for survival', The Ecologist 2 (1): 1-43

Ehrlich, P., and A. Ehrlich, 1991, The Population Explosion, London: Hutchinson

Elner, S., and P. Turchin, 1995, 'Chaos in a noisy world: new methods and evidence from time series analysis', American Naturalist 15 (3): 343-75

Escobar, A., 1995, Encountering Development: The Making and Unmaking of the Third World, Princeton: Princeton University Press

Fairhead, J., and M. Leach, 1996, Misreading the African Landscape: Society and Ecology in a Forest-Savanna Mosaic, Cambridge and New York: Cambridge University Press

Fairhead, J., and M. Leach, (forthcoming) Forests of Statistics: Reframing Environmental History in West Africa. m.s.

Falconer, J., 1990 'The major significance of 'minor' forest products: local people's uses and values of forest in the West African humid forest zone', Community Forestry Note No 6. Rome: FAO

Feeley, J., 1987, 'The early farmers of Transkei, southern Africa, before AD 1870', Cambridge Monographs in African Archaeology No 24, Cambridge: BAR International Series No 378

Ferguson, J., 1990, The Anti-Politics Machine: 'Development', Depoliticization and Bureaucratic State Power in Lesotho, Cambridge: Cambridge University Press

Fortes, M., and E. Evans-Pritchard (eds), 1940, African Political Systems, Oxford: Oxford University Press

Franzdn, L.-O., 1996, 'Trotset i Nayakheda' (Defiance in Nayakheda), in Dagens Nyheter, Stockholm

Frisby, D., and D. Sayer, 1986, Society, London: Tavistock

34 Gambetta, D., 1988, Can we trust trust?' in D. Gambetta (ed), Trust: Making and Breaking Cooperative Relations, Oxford: Basil Blackwell

Gasper, D., 1993, 'Entitlements analysis: relating concepts and contexts', Development and Change 24: 679-718

Gaventa, J., 1995, 'Citizen knowledge, citizen competence and democracy building', The Good Society 5 (3): 28-35

Ghai, D., and J. Vivian (eds), 1992, Grassroots Environmental Action: People's Participation in Sustainable Development, London: Routledge for UNRISD

Ghai, D. (ed), 1994, 'Development and environment: sustaining people and nature', special issue of Development and Change 25 (1)

Giddens, A., 1984, The Constitution of Society: Outline of the Theory of Structuration, Oxford: Polity Press

Glacken, C., 1967, Traces on the Rhodian Shore: Nature and Culture in Western Thought from Ancient Times to the End of the Eighteenth Century, Berkeley: University of California Press

Goetz, A. M., 1996, 'Understanding gendered institutional structures and practices at the macro, meso and micro levels', mimeo, Institute of Development Studies: Brighton

Gore, C., 1993, 'Entitlement relations and "unruly" social practices: a comment on the work of Amartya Sen', Journal of Development Studies 29 (3): 429-60

Guyer, J., and P. Richards, 1996, 'The invention of biodiversity: social perspectives on the management of biological variety in Africa', Africa 66 (1): 1-13

Harriss, J., J. Hunter and C.M. Lewis, 1995, 'Introduction: development and significance of NIE', in J. Harriss, J. Hunter and C.M. Lewis (eds), The New Institutional Economics and Third World Development, London: Routledge: 1-13

Harvey, D., 1979, 'Population, resources, and the ideology of science', in S. Gale, and G. Olsson (eds), Philosophy in Geography, Dordrecht: D.Reidel: 155-185 (first published in Economic Geography Vol 50 [1974])

Hastings, A., C. Hom, S. Ellner, P. Turchin and C. Godfray, 1993, 'Chaos in ecology: is mother nature a strange attractor?' Annual Review of Ecology and Systematics 24: 1-33

Hirschmann, A.O., 1968, Development Projects Observed, Washington DC: Brookings Institution

Hoben A., 1995, 'Paradigms and politics: the cultural construction of environmental policy in Ethiopia', World Development 23 (6): 1007-21

Hobley, M., 1992, 'Policy, rights and local forest management:,the case of Himachal Pradesh, India', Rural Development Forestry Network Paper No 13b, London: ODI

Holmberg, J., K. Thompson and L. Timberlake 1993, Facing the Future: Beyond the Earth Summit, London: IIED/Earthscan

Hoskins, W.G., 1955, The Making of the English Landscape, London: Hodder and Stoughton

35 Humphrey, J., and H. Schmitz, 1996, 'Trust and economic development', IDS Discussion Paper No 355: Brighton: IDS

IUCN/WWF/UNEP, 1991, Caring for the Earth: a Strategy for Sustainable Living, Gland, Switzerland: IUCN, WWF and UNEP

Jenkins, H., 1997, 'Social response to environmental injustice: legal and moral gateways to entitlement', mimeo. Department of Law, School of Oriental and African Studies, University of London

Kabeer, N., and R. Subrahmanian, 1996, 'Institutions, relations and outcomes: frameworks and tools for gender-aware planning', IDS Discussion Paper No 357: Brighton: IDS

Kolasa, J., and S. Pickett (eds), 1991, Ecological Heterogeneity, Ecological Studies No 86. Berlin: Springer-Verlag

Leach, M., 1994, Rainforest Relations: Gender and Resource Use among the Mende of Gola, Sierra Leone, Edinburgh: Edinburgh University Press

Leach, M., and R. Mearns. (1991) Poverty and Environment in Developing Countries: an Overview Study. Report to ESRC, Global Environmental Change Programme; and Overseas Development Administration. Swindon: ESRC

Leach, M., and R. Mearns (eds), 1996, The Lie of the Land: Challenging Received Wisdom on the African Environment, Oxford: James Currey

Long, N., and A. Long (eds), 1992, Battlefields of Knowledge: The Interlocking of Theory and Practice in Social Research and Development, London: Routledge

Long, N., and J.D. van der Ploeg 1989, 'Demythologizing planned intervention: an actor perspective', Sociologia Ruralis XXIX (3/4): 227-49

Long, N. and J.D. van der Ploeg 1994, 'Heterogeneity, actor and structure: towards a reconstitution of the concept of structure', in D. Booth, (ed), Rethinking Social Development: Theory, Research and Practice, Harlow: Longman

Lukes, S., 1973, Individualism, Key Concepts in the Social Sciences Series. Oxford: Basil Blackwell

Lukes, S., 1974, Power: A Radical View, London: Macmillan

May, R., 1976, 'Simple mathematical models with very complicated dynamics', Nature 261: 459-67

May, R., 1977, 'Thresholds and breakpoints in ecosystems with a multiplicity of stable states', Nature 269: 471-77

May, R., 1986, 'When two and two do not make four: nonlinear phenomena in ecology', Proceedings of Royal Society, B228: 241-266

MacArthur, R., and E. Wilson, 1967, The Theory of Island Biogeography, Princeton: Princeton University Press

McIntosh, R., 1985, The Background of Ecology: Concept and Theory, Cambridge: Cambridge University Press

36 Mearns, R., 1995a, 'Institutions and natural resource management: access to and control over woodfuel in East Africa', in T. Binns,(ed), People and Environment in Africa, Chichester: John Wiley & Sons: 103-114

Mearns, R., 1995b, 'Environmental entitlements: towards empowerment for sustainable development', in N. Singh and V. Titi (eds), Empowerment: Towards Sustainable Development. London: Zed Books: 37-53

Mearns, R., 1996a, 'Environmental entitlements: pastoral natural resource management in Mongolia' Cahiers des Sciences Humaines 32 (1): 105-131

Mearns, R., 1996b, 'Community, collective, action and common grazing: the case of post-socialist Mongolia', Journal of Development Studies 32 (3): 297-339

Mearns, R., 1996c, 'Commons and collectives: the lack of social capital in Central Asia's land reforms', IDS Working Paper No 40: Brighton: IDS

Moore, D.S., 1993, 'Contested terrain in Zimbabwe's Eastern Highlands: political ecology, ethnography and peasant resource struggles', Economic Geography 69 (4): 380-401

Mosse, D., 1994, 'Authority, gender and knowledge: theoretical reflections on the practice of participatory rural appraisal', Development and Change 25: 497-526

Mosse, D. (forthcoming), 'Colonial and contemporary ideologies of community management: the case of tank irrigation development in South India', Modern Asian Studies 31

Nolan, P., 1993, 'The causation and prevention of famines: a critique of A.K. Sen', Journal of Peasant Studies 21 (1): 1-28

North, D.C., 1990, Institutions, Institutional Change and Economic Performance, Cambridge: Cambridge University Press

Nuitjen, M., 1992, 'Local organization as organizing practices: rethinking rural institutions', in N. Long, and A. Long (eds), Battlefields of Knowledge: The Interlocking of Theory and Practice in Social Research and Development, London: Routledge

Odum, E., 1953, Fundamentals of Ecology, Philadelphia: Saunders

Osmani, S, 1995, 'The entitlement approach to famine: an assessment', in K. Basu, P. Pattanaik and K. Suzumura (eds), Choice, Welfare and Development: a Festschrift in Honour of Amartya K. Sen, Oxford: Clarendon Press

O'Neill, R., D. De Angelis, J. Waide, and T. Allen (eds), 1986, A Hierarchical Concept of Ecosystems, Princeton: Princeton University Press

Ostrom, E., 1986, 'An agenda for the study of institutions', Public Choice 48: 3-25

Peet, R., and M. Watts (eds), 1996, Liberation Ecologies: Environment, Development and Social Movements, London: Routledge

Pimbert, M., and J. Pretty, 1995, 'Parks, people and professionals: putting "participation" into protected area management, UNRISD Discussion Paper No 57

37 Pimm, S., 1991, The Balance of Nature? Ecological Issues in the Conservation of Species and Communities, Chicago: University of Chicago Press

Powell, W.W., and P.J. DiMaggio (eds), 1991, The New Institutionalism in Organizational Analysis, Chicago: University of Chicago Press

Pretty, J., and I. Guijt, 1992, 'Primary environmental care: an alternative paradigm for development assistance', Environment and Urbanisation 4 (1): 22-36

Putnam, R.D., R. Leonardi and R.Y. Nanetti, 1993, Making Democracy Work: Civic Traditions in Modern Italy, Princeton: Princeton University Press

Radcliffe-Brown, A.R., 1952, Structure and Function in Primitive Society, London: Cohen and West

Rappaport, R.A., 1968, Pigs for the Ancestors: Ritual in the Ecology of a New Guinean People, New Haven: Yale University Press

Redclift, M., and G. Woodgate, 1994, 'Sociology and the environment: discordant discourse?' in M. Redclift, and T. Benton (eds), Social Theory and the Global Environment. London: Routledge: 51-66

Renn, 0., T. Webler and P. Wiedemann (eds), 1994, Fairness and Competence in Citizen Participation: Evaluating Models for Environmental Discourse, Dordrecht: Kluwer Academic

Rocheleau, D., et at (eds), 1996, Feminist Political Ecology: Global Issues and Local Experiences London: Routledge

Roe, E., 1991, "'Development narratives" or making the best of development blueprints', World Development 19 (4): 287-300

Runge, C.F., 1986, 'Common property and collective action in economic development', World Development 14 (5): 623-635

Sarin, M., 1995, 'Regenerating India's forests: reconciling gender equity with joint forest management', IDS Bulletin 26 (1): 83-91

Sauer, C.O., 1925 'The morphology of landscape', University of California Publications in Geography 2 (2): 19-53

Schotter, A., 1981 The Economic Theory of Social Institutions, Cambridge: Cambridge University Press

Schumacher, E.F., 1973, Small is Beautiful: A Study of Economics as if People Mattered, London: Blond and Briggs Ltd

Scoones, I. (in preparation), 'Dynamic ecology and social science: prospects for partnership', mimeo, Institute of Development Studies: Brighton

Scoones, I. (forthcoming), 'Ecological dynamics and grazing resource tenure: a case study from Zimbabwe', in M. Niamir-Fuller (ed), The Role of Transient Institutions in Common Property Resource Management in the Arid Lands of Africa, Stockholm: Beijer Institute of Ecological Economics

38 Scott, J., 1976, The Moral Economy of the Peasant, New Haven: Yale University Press

Scott, J., 1985, Weapons of the Weak: Everyday Forms of Peasant Resistance, New Haven: Yale University Press

Sen, A. (1981) Poverty and Famines: an Essay on Entitlement and Deprivation. Oxford: Oxford University Press

Sen, A.,1984, 'Rights and capabilities', in A. Sen, Resources, Values and Development, Oxford: Basil Blackwell: 307-24

Sen, A., 1985, Commodities and Capabilities, Amsterdam: Elsevier Science Publishers

Stewart, F., 1996, 'Groups for good or ill', Oxford Development Studies 24 (1): 9-26

Swallow, B.M., 1996, 'Understanding the multiple functions of common property regimes: examples from rangelands'. Paper presented at the Sixth Conference of the International Association for the Study of Common Property, Berkeley, California, 5-8 June

Thompson, E.P., 1971, 'The moral economy of the English crowd in the eighteenth century', Past and Present 50: 76-136

Turner, M., 1989, 'Landscape ecology: the effect of pattern on process', Annual Review of Ecology and Systematics 20: 171-197

Vincent, J., 1978, 'Political anthropology: manipulative strategies', Annual Review of Anthropology 7: 175-94

Watts, M., 1991, 'Entitlements or empowerment? Famine and starvation in Africa', Review of African Political Economy 51: 9-26

WCED, 1987, Our Common Future. Report of the World Commission on Environment and Development. Oxford: Oxford University Press

Western, D., M. Wright and S. Strum (eds), 1994, Natural Connections: Perspectives in Community-Based Conservation, Washington DC: Island Press

White, T.A., 1996, 'Private exchange and social capital: multiple functions of common property regimes in Haiti'. Paper presented at the Sixth Conference of the International Association for the Study of Common Property, Berkeley, California, 5-8 June

Williams, M., 1994, 'The relations of environmental history and historical geography', Journal of Historical Geography 20: 3-21

Worster, D., 1985, Nature's Economy: a History of Ecological Ideas, Cambridge: Cambridge University Press

Worster, D., 1990, 'The ecology of order and chaos', Environmental History Review 14: 1-18

Worster, D., 1993, The Wealth of Nature: Environmental History and the Ecological Imagination. New York and Oxford: Oxford University Press

Zimmerer, K., 1994, 'Human geography and the "new ecology": the prospect and prospect of integration, Annals of the Association of American Geographers 84 (1): 108-25

39

Designing Institutions for Environmental and Resource Management

4. The Institutional Analysis and Development Approach

Elinor Ostrom

1. INTRODUCTION'

In designing systems to improve the governance and management of natural resources, a framework to facilitate the organization of diagnostics, analysis, and prescriptions is useful. Frameworks can also aid in the compilation of empirical knowledge and in the assessment of past reforms. For example, environmental problems are said to result from both market failure and state failure. But recommendations for reform of these institutions, unless based on systematic performance comparisons, may reflect only naive beliefs about what makes institutions "good" or "bad". The design and evaluation of institutional arrangements for environmental problems must draw on multiple theories and methods of analyzing institutions and their performance (see, for example, Hurwicz, 1973; 1994). The Institutional Analysis and Development (IAD) approach provides a framework for design. Its focus is on how rules, physical and material conditions, and community attributes shape action arenas and incentives faced by individuals, and hence how these conditions combine to determine outcomes. The components of the IAD framework and some representative applications are described in this chapter.

2. COMPONENTS OF THE IAD FRAMEWORK

The basic conceptual unit for the analysis, prediction, and explanation of behavior within institutional arrangements is the action arena. An action arena is a set of variables, including the action situation, the actors, the structural rules, the community attributes, and the material conditions that shape the action arena. The action arena is the social space within which individuals interact-exchange goods and services, solve problems, dominate

68 The Institutional Analysis and Development Approach 69 one another, or fight. In field settings, there may be overlapping action arenas, and it may be hard to separate one from another. The action arenas experienced by individuals as they move from home to market to work can be viewed as a seamless web. The actors and the action situation in an action arena can themselves depend on other, deeper features such as the rules used by participants to order their relationships, the states of the world acted upon in the arena, and the structure of the more general community within which the arena is placed (see Kiser and Ostrom, 1982). Underlying any model of an action arena are implicit assumptions about the rules individuals use to order their relationships, the states of the world, and the nature of the community within which the arena is located. Different analysts focus on different features of action arenas. For instance, sociologists are often interested in value systems and how humans organize their relationships. Environmentalists address the interactions of physical and biological systems and how these facilitate, constrain, or are otherwise affected by human actions. Institutional analysts focus on rules and incentives and how they affect outcomes.

2.1 Action Situations

The action situation is the structure that includes immediate choices and outcomes. It is typically described by the following sets of variables:

(1) the set of participants, or relevant actors; (2) each participant's set of allowable actions, as defined by the rules; (3) each participant's information about the action situation; (4) the outcomes, or states of the world that can result from the action situation; (5) the technologies or other links by which actions determine outcomes; and (6) the costs and benefits of actions and outcomes for each participant.

Which actions are possible, which outcomes can be produced, how actions are linked to outcomes, and what is contained in the actors' information sets may depend on the physical aspects of a situation, social structures, and past decisions. For example, in the analysis of problems such as overharvesting a common pool resource, describing the action situation means answering the following questions: 70 Designing Institutions for Environmental and Resource Management

(1) Which individuals (and how many) withdraw resource units (e.g., fish, water, fodder) from this resource system? (2) What positions or roles do individuals fill (e.g., member of an irrigation association, water distributor-guard, chair)? Do participants act on their own initiative, or do they confer with others? (For example, does an appropriator obtain a permit before harvesting?) (3) Are appropriators informed about the condition of the resource itself, the cost and benefit functions of other appropriators, and mechanisms by which appropriators' actions combine to produce joint outcomes? (4) What geographic region and what events in that region are affected by the actions of the participants in these positions? (5) Which harvesting technologies are used? (For example, are chain saws used to harvest timber? Are there open and closed seasons? Do fishers return to the water all fish smaller than some limit?) (6) What are the direct costs and benefits of each appropriator's possible actions, and what group benefits result from the various outcomes?

2.2 Actors

An actor is any individual, or group functioning as a unit (corporate actor), whose behavioral choice is relevant to the outcome of an action situation. An action is a behavior to which the actor attaches a subjective and instrumental meaning. An actor is specified by four clusters of variables:

the resources that the actor brings to the situation; the actor's valuations of states of the world and of actions; the actor's knowledge and information, and capabilities for acquiring, processing, retaining, and using knowledge and information; and the actor's method of action selection.

All analyses of behavior at the micro level are based on implicit or explicit theories or models of actors in action situations. The analyst makes assumptions about participants' values; their resources, information, and beliefs; their information-processing capabilities; and their internal strategic- choice mechanisms. The aim is to infer the likely behavior of each actor in a situation, and thus the joint behavior pattern. For many problems, it is useful to accept the classical political economy view that an actor's choices depend on how he or she perceives and weighs the benefits and costs of various strategies and their likely outcomes (Radnitzky, 1987). One formal model of an actor, developed in neoclassical economics and game theory and now well-established in institutional analysis, is homo economicus. To use homo economicus, one assumes that actors have The Institutional Analysis and Development Approach 71 complete and well-ordered preferences and complete information, and that. they choose so as to maximize their net expected returns. All of these assumptions are controversial. Many institutional analysts adopt a broader conception of individual actors, including for example reputations for reliability and trustworthiness among their perceived benefits (Breton and Wintrobe, 1982), or time and resources for establishing and maintaining relationships among their perceived costs (Williamson, 1979). Actors are frequently modeled as if they had complete information about all relevant variables. In highly competitive environments, individuals who survive are often assumed to be maximizers of some key variable associated with survival, such as profits or fitness (Alchian, 1950; Dosi and Egidi, 1987). Constrained maximization may be an adequate description of behavior only in simple decision situations with complete and accurate information generated by institutions. Alternatively, one could assume that the individuals who evaluate benefits and costs are fallible learners who can make mistakes. Institutional incentives may or may not encourage them to learn from their mistakes. Models of the individual that include fallibility and the capacity to learn can be viewed as more general than homo economicus. Different institutional arrangements for governing and managing common pool resources can offer different opportunities to learn.

2.3 Community Attributes

A third set of variables affecting the structure of an action arena concerns the community. Community attributes relevant to the structure of an action arena include behavioral norms, the level and nature of the common understanding shared by potential participants, the extent to which those living in the community have homogeneous preferences, and the distribution of resources. The term culture is frequently applied to this bundle of variables. When all appropriators of a common pool resource share a common set of values and interact within a complex set of arrangements, there is a much greater probability that they will develop adequate rules and norms to manage resources (Taylor, 1987). If keeping one's word is important in such a community, the need for costly monitoring and sanctioning mechanisms is reduced. Conversely, if the appropriators of a resource represent different communities, or are distrustful, then the task of devising and sustaining effective rules is substantially more difficult. Whether a written vernacular language is available to express ideas, develop common understandings, and share learning, is a crucial culture- related variable for institutional analysis (V. Ostrom, 1997). Without a written vernacular language, individuals face considerably greater difficulties 72 Designing Institutions for Environmental and Resource Management in accumulating knowledge in a form suitable for transmission from one generation to the next.

2.4 Rules

Rules serve to specify the action possibilities of actors, their levels of control over outcomes, and available information about the structure of the action situation. Rules also embody individual incentives and disincentives within the group context. Choices within a set of rules may be contrasted with choices among sets of rules. When a "boss" says to an "employee", "How about changing the way we do X?" and the two then discuss alternatives and agree on a better way, they have shifted from actions within previously established rules to decisions about the rules structuring future actions. In IAD language, they have shifted to a collective choice arena.

2.5 Physical and Material Conditions

The same set of rules may yield entirely different action situations if the physical and material conditions are different. The relative importance of the rule configuration and the physical characteristics of a situation can vary widely. In some games, like chess, the rule configuration dominates. But imagine the consequences of switching balls in the games called "football" in North America and elsewhere in the world. In these games, the strategies available to players are strongly affected by the physical attributes of the ball, the equipment, the playing field, etc. Although attributes of the world shape the options for organizing a legislature, rules predominate. Voting behavior depends on whether voting is carried out by raising hands, by counting paper ballots, by calling for ayes and nays, by marching legislators before an official counter, or by polling each legislator's computer terminal. The principle that only one person at a time is to be heard improves legislators' capacity to communicate effectively with one another (see V. Ostrom, 1987).

2.6 Outcomes

The outcome of an action situation is determined by the rules, the states of the world, and the nature of the community. To illustrate, an important mathematical model of harvests from an open access, common pool resource predicts that, acting independently, appropriators of the resource will race to use it up, leading to high social costs (E. Ostrom et al., 1994a). The model supposes that the net benefit to a user increases with own harvesting effort The Institutional Analysis and Development Approach 73 only if the total harvesting effort of other users is low, and it decreases with own effort if total harvesting effort by others is high. This is a stable but deficient equilibrium: too many fishers chase too few fish, too many users pump groundwater, and hunters take more than a sustainable yield of wildlife. Both field research and laboratory experiments strongly support this prediction of overuse and potential destruction of common pool resources with open access (ibid.). Yet sometimes this overuse prediction does not accord with reality. Tendencies to exploit may be mitigated when the resource appropriators have access to arenas within which the open access structure can be modified, or when individual decision-makers are embedded in communities with norms of fairness and conservation that influence strategies. In some situations, the institutional analyst can make only weak inferences about the possible outcomes. But even weak inferences, such as identification of what will not occur, can be important in understanding general tendencies.'

3. THE CONCEPT OF RULES

Rules specify which actions are required, permitted, or prohibited. Rules result from implicit or explicit efforts to achieve order and predictability in human interactions.3 Here we focus on rules in some detail, because they are the fundamental unit that the social designer can use to change social outcomes. The rules that individuals follow in everyday life have many sources (Schlager and Ostrom, 1992), particularly in an open and democratic society. Rules come from central government legislation and regulation; from regional, local, and special governments; and from norms, customs, and other informal arrangements. Within private firms and voluntary associations, individuals adopt rules defining membership, the sharing of profits or benefits, and group decision-making procedures. Families also constitute rule-making bodies.

3.1 Nesting-of Rules

Which rules are important for institutional analysis? Since myriad rules are used in structuring complex action arenas, scholars have been trapped into endless cataloguing of rules, sometimes in ways unrelated to theoretical explanations. But classification is a necessary step to develop a science. A useful classification of rules must be more than a method of imposing superficial order onto an extremely large set of seemingly disparate elements. 74 Designing Institutions for Environmental and Resource Management

We seek to develop an understanding of how rules affect the structure of action situations and thereby affect the way individuals act and the results they achieve. A self-governing society is one in which individuals may participate in crafting rules at many different scales: the family, the neighborhood, the consumers of particular common pool resources or local public goods, the workplace, the region, the country as a whole, and even the international arena. Rules interacting at different levels are called nested rules. Rules are generally nested. That is, it is typical for one set of rules to define how other sets of rules can be changed. Two basic principles governing institutional change can be phrased in terms of changes in rules, as follows:

(1) changes in the rules that order action at one level must be carried out in accordance with a deeper-level set of rules; (2) changes in deeper-level rules are usually more difficult and more costly to implement, so deeper-level rule sets are more stable.

Three levels of rules together determine the actions that can be taken, and the outcomes obtained, in any setting (Kiser and Ostrom, 1982). Operational rules govern day-to-day decisions. Collective choice rules affect how operational rules are to be changed, and who can change them, thereby indirectly affecting operational activities and results. Constitutional choice rules are the rules to be used in crafting collective choice rules that in turn regulate the operational rules. A "metaconstitutional" level underlies all the others. Institutional change is a continuous process of adjustment across these nested levels of rules. Sometimes individuals who are participating in multiple layers of rules will design new rules. Many such rules originate through problem-solving, as individuals interact to figure out how to do a better job in the future than they have in the past. Colleagues in a work team are crafting their own rules when they say, "How about if in the future you do A, and I do B, and we agree that we don't do C before we discuss it and make a joint decision." This collaboration is normal problem-solving behavior for individuals in a democratic society. (Of course, democracies also contain other, less fluid arrangements, such as elections to select legislators.)

3.2 Working Rules

Working rules are informal rules that participants would refer to if asked to explain and justify their actions to fellow participants. While following a rule is a "social habit or norm", individuals may explicitly decide to adopt a The Institutional Analysis and Development Approach 75 different rule; over time, behavior in conformity with the new rule again becomes habitual (Shimanoff, 1980; Toulmin, 1974; Harre, 1974). Humans use complex cognitive systems to order their own behavior, often at a subconscious level. In a system governed by "rule of law", the general legal framework in use will have its source in actions taken in constitutional, legislative, and administrative settings, augmented by decisions taken by individuals. Its rules-in-form will be consistent with its rules-in-use (Sproule-Jones, 1993). In a system not governed by "rule of law", central laws may exist, and there may even be considerable enforcement effort, but individuals will typically attempt to evade rather than obey the law. The stability of rule-ordered actions depends on the shared meaning assigned to the words used to formulate the underlying rules. Rules, formulated in human language, suffer from the lack of clarity, misunderstanding, and change that typify language-based phenomena (V. Ostrom, 1980; 1997). If there is no shared meaning, then there will be confusion about which actions are required, permitted, or forbidden. Action regularities cannot result if the meaning of a rule within an action situation is susceptible to multiple interpretations. Because "rules are not self- formulating, self-determining, or self-enforcing" (V. Ostrom, 1980, p. 342), human agents must formulate them, apply them where appropriate, and attempt to enforce them. Even if shared meaning exists at the time a rule is accepted, transformations in technology, social norms, or other circumstances may change the events to which rules apply. The working rules used by participants to order their relationships within an action arena should be conceived as the minimum set that can explain action selection and results. But, rule-following or conforming actions are not as predictable as biological or physical behavior explained by scientific laws. Because the state of the world, its possible transformations, and the nature of the community all affect the structure of an action situation, working rules alone explain this structure completely. Seven types of working rules can be said to affect the structure of an action situation: entry and exit rules, position rules, scope rules, authority rules, information rules, aggregation rules, and payoff rules. Together, rules of these types play a major role in setting the values of variables that determine the structure of an action situation. Entry and exit rules affect the number of participants, their attributes and resources, as well as the conditions they face on entering or leaving the arena. Position rules establish positions or roles in the situation. Scope rules delimit the outcomes that can be affected and, working backwards, the actions linked to specific outcomes. Authority rules assign sets of actions that participants in positions at particular nodes must, may, or may not take. Combined with 76 Designing Institutions for Environmental and Resource Management scientific laws about the states of the world being acted upon, authority rules determine the form of the decision tree-the action-outcome linkages and number of iterations. Information rules affect the contingent information of the participants, and aggregation rules the level of control that a participant in a position exercises in the selection of an action at a decision node. Payoff rules determine the benefits and costs assigned to particular combinations of actions and outcomes, thereby establishing incentives and deterrents. The set of working rules is a rule configuration in the sense that the effects of changing one rule may depend on the other rules-in-use. An analyst assessing the use of a common pool resource could ask a series of questions to help identify the rules-in-use that structure the action situation, including the following:

Who can appropriate this resource? Are appropriators limited to local residents; to one group defined by ethnicity, race, caste, gender, or family structure; to those who win a lottery; to those who have obtained a permit; to those who own required assets (such as land, or a fishing berth); or to some other bounded class of individuals? Can new participants join the group by some kind of initiation, or by paying an entry fee? Must an appropriator give up harvesting rights after migrating to another location? How does an appropriator change roles in the group, say from being just an ordinary "member" to having a specialized task, such as a water distributor-guard? What understandings do the appropriators, and others, have about the authorized or forbidden geographic and functional domains? Are there maps that show who can appropriate from which region? Are some resource units "off-limits"? (For example, in some areas of Africa, there were rules against cutting down particular acacia trees growing on privately owned or communal land.) Do appropriators have understandings about mandatory, authorized, or forbidden harvesting technologies? For fishers, is there a minimum dimension for net mesh? For forest users, are some cutting tools required or prohibited? What action choices are available to monitors? Are there. rules concerning the aggregation of individual choices? (Do certain actions require prior permissions or agreements, for example?) What information must be held secret and what information must be made public? What sanctions are imposed for breaking any of the rules identified above? How is compliance to rules monitored? Who is responsible for sanctioning nonconformers? How reliably are sanctions imposed? Are positive rewards offered for actions or roles? (For example, is an elected official relieved of labor duties?) The Institutional Analysis and Development Approach 77

The problem for the field researcher is that rules-in-use are usually not written down. Nor can the field researcher carry out a survey, asking a random sample of appropriators and others about their rules. Many rules-in- use are not even conceptualized as rules because, for instance, participants view them as prescriptions from external authorities. If the rules-in-use have evolved over a long period of time, or if participants understand them implicitly rather than explicitly, then the institutional analyst can gain an appreciation of the rule configuration only from long and careful observations of behavior, and from the answers to nonthreatening, context-specific questions.'

3.3 Rules as Related to Important Physical Attributes

Specific physical attributes are frequently used to distinguish goods and services that are more effectively provided by particular institutional arrangements. "Public goods" are those that are nonsubtractive (the benefits can be enjoyed jointly and simultaneously by many users) and nonexcludable (additional appropriators are difficult to exclude). "Common pool" resources are subtractive-generally, the total available for harvesting is finite-but nonexcludable.

3.3.1 Excludability and the free-rider problem It is frequently assumed that if the consumers or harvesters of a good are difficult to exclude, then that good must be provided publicly rather than privately. When - the benefits of a good are available to all members of a group, whether or not they contribute to its provision, that good is characterized by nonexcludability. Whenever exclusion is costly, the providers of a good or service face a potential free-rider or collective-action problem (Olson, 1965). Individuals who gain from an irrigation system, for example, may avoid labor or tax contributions to maintenance, hoping that others will bear the burden. This is not to say that all individuals will free- ride whenever they can. But the incentive to free-ride will always be strong, whenever potential beneficiaries of a good or service cannot easily be excluded for failing to contribute. Profit-seeking entrepreneurs-who must recoup their investments through quid pro quo exchanges-have little incentive to provide a good or service when the cost of excluding individuals from the benefits is high. The excludability characteristics of common pool resources thus lead to free- riding, which in turn leads to underinvestment in capital and maintenance. Often, goods with excludability problems become public goods, as the public sector has authority to finance the good using taxes or fees and to regulate its use. Public sector involvement, however, raises difficult questions concerning 78 Designing Institutions for Environmental and Resource Management whether the design of collective-choice mechanisms accurately reflects the preferences of the beneficiaries of the good, and whether the means to finance the construction and maintenance of the good are appropriate and fair. A variety of institutional arrangements can help alleviate free-riding. One approach is provision by an appropriate level of government. The problems that plague public sector provision of common pool resources include determination of preferences and organization of finance. Private groups (clubs) that can control their own membership can avoid some of the problems of collective action.' Clubs reduce free-riding by assuring their members that: (1) the benefits they receive will exceed the costs they pay; (2) the collective benefit can be achieved only through member contributions; and (3) most beneficiaries will contribute their share of needed inputs (Popkin, 1981; Frohlich and Oppenheimer, 1970, 1974; Frohlich et al., 1971). When a supplier can exclude at low cost, preferences are revealed as a result of quid pro quo transactions. Producers learn consumers' preferences through their willingness to pay for various goods offered for sale. When exclusion is costly, mechanisms to obtain truthful information about beneficiaries' preferences (willingness to pay) are complex, whether a good is provided by a public or a private supply. When the group of participants is small, face-to-face discussions of preferences and constraints can produce a rough consensus for larger groups, decisions (about infrastructure, for example) are made through mechanisms such as voting or the delegation of authority to public officials. However, making collective choices that adequately reflect individual preferences may be difficult as discussed in the extensive literature on voting systems (Arrow, 1951; Shepsle, 1979; Buchanan and Tullock, 1962). Some goods-such as public spraying against insects-have the property that, once the good is provided, consumers have no choice about the amount they consume. An individual who opposes such a good will have a strong incentive not to comply with a general tax levy. Compliance with a tax to finance a public good thus depends on the legitimacy of the public-choice mechanism used to make provision decisions.

3.3.2 Subtractivity For pure public goods, consumption by one individual does not subtract from the flow of services available to others. The flow of services from a jointly used infrastructure facility, on the other hand, can be entirely subtractive. Most agricultural uses of water are fully subtractive. For example, the withdrawal of any quantity of water from an irrigation canal by one farmer reduces by the same amount the quantity available to other users. However, many other uses of water-such as for power generation or navigation-are The Institutional Analysis and Development Approach 79 not subtractive, because, for instance, most of the water that passes through a turbine to generate power can be used again downstream. When services are scarce relative to demand, and use of the flow of services by one individual subtracts from what is available to others, users will be tempted to take as much of the flow as they can for fear that it will not be available later. Effective rules are required to allocate scarce, fully subtractive service flows productively. The imposition of metering and use-based charges constitutes one such allocation mechanism. Sometimes, however, the pricing of subtractive services is counterproductive, leading some individuals to grab as much of the flow as possible and resulting in noneconomic uses of the services and high levels of conflict among users. Allocation rules also affect users' incentives to maintain a system. Farmers at the tail end of an irrigation system that lacks effective allocation rules have little motivation to contribute to maintenance because they receive their share of the water only occasionally. Similarly, farmers at the head end of such a system are not motivated to provide maintenance services voluntarily, because they can obtain enough water even if the system is poorly maintained (E. Ostrom, 1996). Consequently, for common pool resources with highly subtractive flows, institutional arrangements related to flow allocation are intimately tied to sustainability. Sustainability is almost certainly unattainable without careful attention to the efficiency, fairness, and enforceability of the rules specifying who can appropriate how much of the flow, when, where, and under what conditions. Furthermore, beneficiaries will resist rather than accept responsibilities that are not clearly linked to benefits.

3.3.3 Additional attributes In addition to the general attributes of excludability or nonexcludability, and subtractivity or nonsubtractivity, resource systems are characterized by other physical and material conditions that affect the ability of rules to generate positive or negative incentives. Whether resource units are mobile or stationary, and whether storage is feasible, are reflected in the problems of governing and managing common pool resources (Schlager et al., 1994). A lobster fishery, for example, is simpler to regulate than a salmon fishery. Similarly, predictable and efficient water allocation is easier to achieve in a system with storage than in a run-of-the-river system. Some natural resource systems, such as many groundwater basins, are renewable but characterized by very long cycle times. If the time horizon for sustainability is distant, then appropriate allocation rules may be the difference between creating a sustainable conjunctive use system and destroying the resource. To devise effective rules for regulating withdrawal from an oil pool, on the other hand, requires that the optimal extraction path be 80 Designing Institutions for Environmental and Resource Management determined. The cost of withdrawing the last few units of oil will be much higher if producers have not coordinated their withdrawal patterns, yet the fmiteness of the resource may not generate adequate incentives for regulation early in the development phase. The size of a resource system can also have a major impact on the incentives facing participants. The length and slope of the main canal of an irrigation system affects not only its maintenance cost but also the strategic bargaining between headenders and tailenders (Lam, 1998; E. Ostrom, 1996). When the number of participants increases, so do transaction costs, though how steeply depends to a large extent on the rules-in-use and the degree of heterogeneity of the users. The productivity, predictability, and patchiness of a resource affect the likelihood that successful management can be achieved by private-property or common-property arrangements (Netting, 1982). Similarly, the resilience of a multispecies ecosystem affects its sensitivity to the particular rules used to govern it, and to changes in economic or environmental conditions elsewhere (Holling, 1994). All of these attributes are slowly being integrated into a body of coherent theory about the impacts of physical and material conditions on the structures of action situations and their resulting incentives and behavior. Analysts diagnosing resource problems should be aware of the enormous diversity of resource settings and the importance of tailoring rules to the particular combination of attributes that applies. Rules that are uniform across resources, or throughout a geographical or political unit, are rarely optimal.

4. EVALUATING RULES AND OUTCOMES

The institutional analyst geared toward designing rules must evaluate alternative institutional arrangements according to their likely outcomes. Evaluation criteria may include economic efficiency, equity, accountability, conformance to cultural norms of morality, and adaptability. Each of these criteria is described below. To compare the performance of different institutional arrangements, the institutional analyst can evaluate their success in meeting these criteria (see E. Ostrom et al., 1993, ch.5). Meticulous comparison of the relative success of real-world institutions on each criterion can be an immense task for the analyst. Another consideration is how real-world institutions embody the deliberate trade-offs among criteria, particularly between efficiency and equity. Therefore, comparison of alternative institutions also requires assessment of the social desirability of possible trade-offs among criteria. The Institutional Analysis and Development Approach 81

4.1 Economic Efficiency

A private, competitive market is generally inefficient at allocating nonexcludable or nonsubtractive goods because the theoretical conditions for economic efficiency for such goods are not satisfied. Thus a need for government intervention is created. Economic efficiency effects of intervention are described by the magnitude of net benefits associated with a reallocation of resources. However, institutions that permit opportunistic behavior like shirking can also lead to an inefficient allocation of resources. In evaluating alternative institutional arrangements, it is crucial to consider how revisions in the rules will alter participants' behavior and, consequently, the allocation of resources.6

4.2 Equity

Outcomes can be evaluated using two different concepts of equity-allocational and fiscal. Fiscal equity has to do with public finance. The fundamental tenet of fiscal equity is that, in an exchange economy, those who benefit from a good or service should bear the burden of financing it. An obvious corollary is that those who derive greater benefits from a public good should pay more for it. Perceptions of greater or lesser degrees of fiscal equity can affect the willingness of individuals to contribute to the development and maintenance of resource systems. Allocation policies that redistribute resources are particularly important, when distributions of wealth are highly skewed, as in developing countries. Although efficiency might dictate using scarce resources where they produce the greatest net benefit, allocational equity goals temper this objective by shifting resources toward programs that benefit needy groups. Allocational equity may conflict with fiscal equity. For example, if a water pump benefits a community of users, the fiscal equity principle holds that each individual in the community should be required to pay the marginal costs associated with his or her use of the pump. But under an ability-to-pay principle, such fees may be inappropriate because they would mean that low- income individuals would be unable to access the pump.

4.3 Other Criteria

To achieve efficiency and equity goals, officials should be accountable to citizens concerning the development and use of public facilities and natural resources. Without adequate accountability, targeted groups are less likely to receive the benefits of projects, and resources are more likely to be wasted. In addition to accountability, one can evaluate the kind of general morality 82 Designing Institutions for Environmental and Resource Management fostered by a particular set of institutional arrangements. Do cheaters receive high payoffs if undetected? Are those who keep promises more likely to be rewarded and advanced in their careers? What kind of behavior is learned when individuals interact repeatedly over the long term? Adaptability requires institutional arrangements to respond as environments change. In rural areas of developing countries where crises and natural disasters are often very localized, an institutional arrangement must be flexible to succeed. For example, if an irrigation system allocates only a pre-specified level of resources to periodic maintenance, it may not be able to respond when a canal section is destroyed by flood. Institutions subject to greater local control are more likely to have rules that facilitate quick and appropriate crisis response.

5. APPLICATIONS OF THE IAD FRAMEWORK

The IAD framework has been used extensively, both as a vehicle for teaching (see, for example, E. Ostrom et al., 1996) and as a metalanguage for analyzing diverse theories. In the early 1970s, the IAD framework was developed as part of a program to understand how the paradigms of political science conditioned conceptions of public administration and metropolitan organization (V. Ostrom and E. Ostrom, 1971; E. Ostrom, 1972). For a decade and a half, we used the nascent framework as a foundation for the many empirical studies of police service delivery in metropolitan areas.' During the past decade, the IAD framework has provided the language to link formal models of resource appropriation with empirical work conducted in both laboratory and field settings (E. Ostrom et al., 1994a).' Colleagues at Indiana University have developed a series of theoretical models of appropriation from a common pool resource and tested them in laboratory experiments (see E. Ostrom et al., 1992; 1994a; Walker and Gardner, 1992; Hackett et al., 1994). When laboratory subjects are not allowed to communicate, their behavior closely approximates the predictions of the theory of finitely repeated, noncooperative games. But when communication or sanctions are allowed, the behavior observed in the lab is not consistent with these theoretical models; rather, it is closer to what has been observed in field settings. Consequently, we developed a theory of how boundedly rational individuals use heuristics such as "measured responses" to stabilize agreements achieved in settings where there are no external enforcers to impose rules on participants (E. Ostrom et al., 1994a). Three major databases for the study of common pool resources and diverse property regimes have been founded on the IAD framework. The first "CPR The Institutional Analysis and Development Approach 83

Database" drew on cases produced for an NAS panel and on many individual case studies provided by historians, sociologists, engineers, political scientists, anthropologists, and environmental scientists (Martin, 1989; 1992; Hess, 1996). The IAD framework was used explicitly to create a structured database for appropriation and for collective choice arenas. About 50 common property resource systems were assessed by Schlager (1990; 1994) (who studied in-shore fisheries) and Tang (1991; 1992) (who studied irrigation systems) to isolate key rules associated with better performance. In Governing the Commons (E. Ostrom, 1990), I drew on these and other case studies to elucidate some design principles that characterize robust, self- organized institutions that have fostered sustainable resource use, and to develop an initial theory of institutional change. The second database focuses entirely on irrigation systems (Benjamin et al., 1994). Containing more than 175 irrigation systems in Nepal, it has enabled us to carry out convincing tests of many propositions concerning institutional performance (see Lam, 1998; E. Ostrom et al., 1994b; E. Ostrom and Gardner, 1993; E. Ostrom, 1996). This enterprise has enabled us to challenge many of the standard assumptions of development scholars who have presumed that farmers are unable to self-organize to engage in costly collective action unless rules are imposed by external authorities (see also Thomson, 1992). Controlling for factors such as size of group, length of canal, and type of terrain, we have found that, measured by agricultural productivity, farmer-managed irrigation systems in Nepal outperform agency- managed systems (Lam, 1998). The third database is an integral part of the International Forestry Resources and Institutions (IFRI) research program, an ongoing research program of the Workshop in Political Theory and Policy Analysis and of the Center for the Study of Institutions, Population, and Environmental Change. This research program is designed to address gaps in knowledge about how institutions affect the incentives of forest users, resulting in substantial deforestation in some locations while improving forest conditions elsewhere. Six collaborative research centers have now been established in Bolivia, Ecuador, India, Mali, Nepal, and Uganda and more will be established soon (E. Ostrom and Wertime, 1994; Jerrells and E. Ostrom, 1995). Many scholars have drawn on the IAD framework to produce empirically grounded studies demonstrating that diverse sets of institutional arrangements are needed to generate appropriate incentives for the successful management of natural resource systems. Banana and Gombya-Ssembajjwe (1996) have shown that, in Uganda, deforestation is minimized when local institutional arrangements include extensive monitoring, and are viewed as legitimate by local residents. In their study of a comuna in Ecuador, Becker and Gibson (1996) have documented the importance of distance from a forest in 84 Designing Institutions for Environmental and Resource Management determining villagers' costs for active monitoring and enforcement, even when villagers have full authority to devise and enforce their own rules. Agrawal (1996) challenges empirically the presumption of many scholars that collective action becomes progressively more difficult as the size of the group increases. He shows that in India, villages of moderate size are better able to generate the labor needed to protect local forests than are very small villages. Schweik's study (1996) shows that neither local population levels nor the predictions of optimal foraging theory explain the geographic distribution of Shorea robusta, a highly valued species in Nepal. The most robust explanation for the observed distribution lies in institutional rules that allow higher-caste villagers to access their "own" forests, as well as forests near lower-caste villages, but not vice versa. From our studies, it is clear that neither market nor state institutions can, by themselves, cope adequately with the vast array of environmental problems found in developing and developed countries. Moreover, we have found that, to be successful, intentional institutional design requires a deep sensitivity to the interaction of rules with physical and biological attributes. Rules that in some circumstances generate positive incentives and effective environmental and resource management may lead to deficiencies under other conditions. Institutional designers seek the rules most likely to enhance human welfare in a particular context. The IAD approach provides a framework for design, enabling us to analyze how rules interact with the physical and biological world, and culture, to condition the behavior of individuals, and produce social and environmental outcomes. Our preferences over these outcomes will influence our prescriptions about rules to bring about change.

NOTES

1. An earlier version of part of this paper was presented at the 1996 Annual Meetings of the American Political Science Association, San Francisco, CA, USA, 29 August - 1 September 1996, and to the Economic Development Institute of the World Bank, Curriculum Development Workshop, Washington, DC, 6-7 December 1995. The author is appreciative of the support provided by the National Science Foundation (Grants SBR-93 19835 and SBR-95 21918) and the Ford Foundation (Grant 940 0518). Useful comments by Kathryn Firmin-Sellers, Roy Gardner, Maurice Garner, Clark Gibson, Vincent Ostrom, Roger Parks, Margaret Polski, Eric Rasmusen, James Walker, Tjip Walker, and Xin Zhang on an earlier draft are deeply appreciated. Patty Dalecki's careful review of this paper is greatly appreciated. 2. In laboratory experiments on common pool resources, subjects who can communicate generally place greater value on favorable joint outcomes (see E. Ostrom et al., 1994a and citations therein). In field settings, helping individuals The Institutional Analysis and Development Approach 85

engage in face-to-face discussions usually makes improved outcomes more likely, depending on historical and other factors. 3. Rules create classes of persons (positions) who are then required, permitted, or forbidden to select actions from classes that may depend on the state of the world (Crawford and Ostrom, 1995; V. Ostrom, 1991). 4. The International Forestry Resources and Institutions (IFRI) research program has faced this problem in developing research protocols to enable a network of research scholars to gather the "same" information from a multicountry sample of forestry sites. The recording forms can be structured and filled in by the research team after in-depth group and individual discussions, but there cannot be a standard way of asking the questions. Anthropologists generally look upon the individuals with whom they talk as "informants"; this is the stance one must take to elucidate information about rules-in-use (see E. Ostrom and Wertime, 1994). 5. The theory of clubs evolved in analyses of situations where strictly private arrangements sufficed to overcome free-rider problems (see Buchanan, 1965; Sandler and Tschirhart, 1980; Cornes and Sandler, 1986). 6. For example, according to a well-known pricing efficiency rule, price should equal marginal cost. But this rule implies that the efficient price for nonsubtractive goods is zero, because the marginal cost of another user is zero. Yet a zero user price would provide perverse resource use incentives. It would also require that resource use be financed entirely through taxes. 7. See E. Ostrom and Whitaker, 1973; E. Ostrom et al., 1973; 1978; Parks and Ostrom, 1981; Parks, 1984; E. Ostrom, 1975; McIver, 1978; McDavid, 1974. 8. In the early 1980s, the National Academy of Sciences organized a research panel on the study of common property. Ronald Oakerson's core paper (1992) was used to organize a series of case studies on how diverse groups had devised institutional arrangements related to common pool resources (see also Thomson et al., 1992; E. Ostrom, 1992). Oakerson's presentation of the framework has influenced studies of common-property regimes in many sectors around the world. The intellectual productivity stimulated by the NAS panel led to the formation of the International Association for the Study of Common Property. Only a few years later, the 1996 meeting of the IASCP was attended by more than 500 scholars.

REFERENCES

Agrawal, A. (1996). `Group size and successful collective action: A case study of forest management* institutions in the Indian Himalayas', Forests, Trees and People Programme, Phase II, Food and Agriculture Organization of the United Nations, Rome, Italy. Alchian, A.A. (1950). `Uncertainty, evolution, and economic theory', Journal of Political Economy 58, 211-221. Arrow, K. (1951). Social Choice and Individual Values, 2nd ed., Wiley, New York, NY, USA. Banana, A.Y. and W. Gombya-Ssembajjwe. (1996). `Successful forest management: The importance of security of tenure and rule enforcement in Ugandan forests', 86 Designing Institutions for Environmental and Resource Management

Forests, Trees and People Programme, Phase II, Food and Agriculture Organization of the United Nations, Rome, Italy. Becker, C.D. and C.C. Gibson. (1996). `The lack of institutional supply: Why a strong local community in western Ecuador fails to protect its forest', Forests, Trees and People Programme, Phase II, Food and Agriculture Organization of the United Nations, Rome, Italy. Benjamin, P., W.F. Lam, E. Ostrom, and G. Shivakoti. (1994). Institutions, Incentives, and Irrigation in Nepal, Associates in Rural Development, Burlington, VT, USA. Breton, A. and R. Wintrobe. (1982). The Logic of Bureaucratic Conduct: An Economic Analysis of Competition, Exchange, and Efficiency in Private and Public Organizations, Cambridge University Press, Cambridge, UK. Buchanan, J.M. (1965). `An economic theory of clubs', Economica 32, 1-14. Buchanan, J.M. and G. Tullock. (1962). The Calculus of Consent, University of Michigan Press, Ann Arbor, MI, USA. Corns, R. and T. Sandler. (1986). The Theory of Externalities, Public Goods, and Club Goods, Cambridge University Press, Cambridge, UK. Crawford, S.E.S. and E. Ostrom. (1995). `A grammar of institutions', American Political Science Review 89, 582-600. Dosi, G. and M. Egidi. (1987). `Substantive and procedural uncertainty: An exploration of economic behaviours in complex and changing environments', presentation, International Workshop on Programmable Automation and New Work Modes, Paris, France. Frohlich, N. and J.A. Oppenheimer. (1970). `I get by with a little help from my friends', World Politics 23, 104-120. Frohlich, N. and J.A. Oppenheimer. (1974). `The carrot and the stick: Optimal program mixes for entrepreneurial political leaders', Public Choice 19, 43-61. Frohlich, N., J.A. Oppenheimer, and O. Young. (1971). Political Leadership and Collective Goods, Princeton University Press, Princeton, NJ, USA. Gregg, P.M. (1974). `Units and levels of analysis: A problem of policy analysis in federal systems', Publius 4, 59-86. Hackett, S., E. Schlager, and J. Walker. (1994). `The role of communication in resolving commons dilemmas: Experimental evidence with heterogeneous appropriators', Journal of Environmental Economics and Management 27, 99-126. Harre, R. (1974), `Some remarks on "rule" as a scientific concept', in T. Mischel (ed.), Understanding Other Persons, Basil Blackwell, Oxford, UK. Hess, C. (1996). Common Pool Resources and Collective Action: A Bibliography, vol. 3, Workshop in Political Theory and Policy Analysis, Indiana University, Bloomington, IN, USA. Holling, C.S. (1994). `An ecologist's view of the Malthusian conflict', in K. Lindahl- Kiessling and H. Landberg (eds), Population, Economic Development, and the Environment, Oxford University Press, New York, NY, USA, 79-103. Hurwicz, L. (1973). `Design mechanisms for resource allocation', American Economic Review 63, 1-30. Hurwicz, L. (1994). `Economic design, adjustment processes, mechanisms, and institutions', Design 1, 1-14. The Institutional Analysis and Development Approach 87

Jerrells, J. and E. Ostrom. (1995). `Current developments in a relational database for biological and social science research', presentation, IUFRO World Congress, Tampere, Finland, 8-11 August. Kiser, L.L. and E. Ostrom. (1982). `The three worlds of action: A metatheoretical synthesis of institutional approaches', in E. Ostrom (ed.), Strategies of Political Inquiry, Sage, Beverly Hills, CA, USA, 179-222. Lam, W.F. (1998). Institutions, Engineering Infrastructure, and Performance in the Governance and Management of Irrigation Systems: The Case of Nepal, Institute for Contemporary Studies Press, San Francisco, CA, USA. Lasswell, H. and A. Kaplan. (1950). Power and Society. A Framework for Political Inquiry, Yale University Press, New Haven, CT, USA. Martin, F. (1989). Common Pool Resources and Collective Action: A Bibliography, vol. 1, Workshop in Political Theory and Policy Analysis, Indiana University, Bloomington, IN, USA. Martin, F. (1992). Common Pool Resources and Collective Action: A Bibliography, vol. 2, Workshop in Political Theory and Policy Analysis, Indiana University, Bloomington, IN, USA. McDavid, J.C. (1974). `Interjurisdictional cooperation among police departments in the St. Louis metropolitan area', Publius 4, 35-58. McIver, J.P. (1978). `The relationship between metropolitan police industry structure and inter-agency assistance: A primary assessment', Policy Studies Journal 7, 406- 413. Netting, R. McC. (1982). `Territory, property, and tenure', in R. McC. Adams, N.J. Smelser, and D.J. Treiman (eds), Behavioral and Social Science Research: A National Resource, National Academy Press, Washington, DC, USA, 446-501. Oakerson, R.J. (1992). `Analyzing the commons: A framework', in D.W. Bromley (ed.), Making the Commons Work: Theory, Practice, and Policy, Institute for Contemporary Studies Press, San Francisco, CA, USA, 41-59. Oakerson, R.J. (1994). `The logic of multi-level institutional analysis', presentation, "Workshop on the Workshop" Conference, Workshop in Political Theory and Policy Analysis, Indiana University, Bloomington, IN, USA, 15-19 June. Olson, M. (1965). The Logic of Collective Action: Public Goods and the Theory of Groups, Harvard University Press, Cambridge, MA, USA. Ostrom, E. (1972). `Metropolitan reform: Propositions derived from two traditions', Social Science Quarterly 53, 474-493. Ostrom, E. (1975). `On righteousness, evidence, and reform: The police story', Urban Affairs Quarterly 10, 464-486. Ostrom, E. (1990). Governing the Commons: The Evolution of Institutions for Collective Action, Cambridge University Press, New York, NY, USA. Ostrom, E. (1992). `The rudiments of a theory of the origins, survival, and performance of common-property institutions', in D.W. Bromley (ed.), Making the Commons Work: Theory, Practice, and Policy, Institute for Contemporary Studies Press, San Francisco, CA, USA, 41-59. Ostrom, E. (1996). `Incentives, rules of the game, and development', Annual World Bank Conference on Development Economics 1995, The World Bank, Washington, DC, USA, 207-234. 88 Designing Institutions for Environmental and Resource Management

Ostrom, E. and R. Gardner. (1993). `Coping with asymmetries in the commons: Self- governing irrigation systems can work', Journal of Economic Perspectives 7, 93- 112. Ostrom, E., R. Gardner, and J. Walker. (1994a). Rules, Games, and Common Pool Resources, University of Michigan Press, Ann Arbor, MI, USA. Ostrom, E., W.F. Lam, and M. Lee. (1994b). `The performance of self-governing irrigation systems in Nepal', Human Systems Management 14, 87-108. Ostrom, E., V. Ostrom, and M. McGinnis. (1996). `A course of study in institutional analysis and development, 1995-96', Workshop in Political Theory and Policy Analysis, Indiana University, Bloomington, IN, USA. Ostrom, E., R.B. Parks, and G.P. Whitaker. (1973). `Do we really want to consolidate urban police forces? A reappraisal of some old assertions', Public Administration Review 33, 423-432. Ostrom, E., R.B. Parks, and G.P. Whitaker. (1978). Patterns of Metropolitan Policing, Ballinger, Cambridge, MA, USA. Ostrom, E., L. Schroeder, and S. Wynne. (1993). Institutional Incentives and Sustainable Development: Infrastructure Policies in Perspective, Westview Press, Boulder, CO, USA. Ostrom, E., J. Walker, and R. Gardner. (1992). `Covenants with and without a sword: Self-governance is possible', American Political Science Review 86, 404-417. Ostrom, E. and M.B. Wertime. (1994). `IFRI research strategy', Workshop in Political Theory and Policy Analysis, Indiana University, Bloomington, IN, USA. Ostrom, E. and G.P. Whitaker. (1973). `Does local community control of police make a difference? Some preliminary findings', American Journal of Political Science 17, 48-76. Ostrom, V. (1980). `Artisanship and artifact', Public Administration Review 40, 309- 317. Ostrom, V. (1987). The Political Theory of a Compound Republic: Designing the American Experiment, 2nd ed., University of Nebraska Press, Lincoln, NE, USA. Ostrom, V. (1991). The Meaning of American Federalism: Constituting a Self- Governing Society, Institute for Contemporary Studies Press, San Francisco, CA, USA. Ostrom, V. (1997). The Meaning of Democracy and the Vulnerability of Democracies, University of Michigan Press, Ann Arbor, MI, USA. Ostrom, V. and E. Ostrom. (1971). `Public choice: A different approach to the study of public administration', Public Administration Review 13, 203-216. Parks, R.B. (1984). `Linking objective and subjective measures of performance', Public Administration Review 44, 118-127. Parks, R.B. and E. Ostrom. (1981). `Complex models of urban service systems', in T.N. Clark (ed.), Urban Policy Analysis: Directions for Future Research, Urban Affairs Annual Reviews 21, Sage, Beverly Hills, CA, USA, 171-199. Popkin, S.L. (1981). `Public choice and rural development-free riders, lemons, and institutional design', in C.S. Russell and N.K. Nicholson (eds), Public Choice and Rural Development, Johns Hopkins University Press, Baltimore, MD, USA, 43-80. Radnitzky, G. (1987). 'Cost-benefit thinking the methodology of research: The "economic approach" applied to key problems in the philosophy of science', in G. The Institutional Analysis and Development Approach 89

Radnitzky and P. Bernholz (eds), Economic Imperialism: The Economic Approach Applied Outside the Field of Economics, Paragon House, New York, NY, USA, 283-334. Sandier, T. and J. Tschirhart. (1980). `The economic theory of clubs: An evaluative survey', Journal of Economic Literature 18, 1481-1521. Schlager, E. (1990). `Model specification and policy analysis: The governance of coastal fisheries', Ph.D. dissertation, Indiana University. Schlager, E. (1994). `Fishers' institutional responses to common pool resource dilemmas', in E. Ostrom, R. Gardner, and J. Walker (eds), Rules, Games, and Common Pool Resources, University of Michigan Press, Ann Arbor, MI, USA, 247-266. Schlager, E., W. Blomquist, and S.Y. Tang. (1994). `Mobile flows, storage, and self- organized institutions for governing common pool resources', Land Economics 70, 294-317. Schlager, E. and E. Ostrom. (1992). 'Property-rights regimes and natural resources: A conceptual analysis', Land Economics 68, 249-262. Schweik, C.M. (1996). `Social norms and human foraging: An investigation into the spatial distribution of Shorea robusta in Nepal', Forests, Trees and People Programme, Phase II, Food and Agriculture Organization of the United Nations, Rome, Italy. Shepsle, K.A. (1979). `The role of institutional structure in the creation of policy equilibrium', in D.W. Rae and T.J. Eismeier (eds), Public Policy and Public Choice, Sage Yearbooks in Politics and Public Policy 6, Sage, Beverly Hills, CA, USA, 249-283. Shimanoff, S.B. (1980). Communication Rules: Theory and Research, Sage, Beverly Hills, CA, USA. Sproule-Jones, M. (1993). Governments at Work: Canadian Parliamentary Federalism and Its Public Policy Effects, University of Toronto Press, Toronto, ON, Canada. Tang, S.Y. (1991). `Institutional arrangements and the management of common pool resources', Public Administration Review 51, 42-51. Tang, S.Y. (1992). Institutions and Collective Action: Self-Governance in Irrigation, Institute for Contemporary Studies Press, San Francisco, CA, USA. Taylor, M. (1987). The Possibility of Cooperation, Cambridge University Press, New York, NY, USA. Thomson, J.T. (1992). A Framework for Analyzing Institutional Incentives in Community Forestry, Forestry Department, Food and Agriculture Organization of the United Nations, Rome, Italy. Thomson, J.T., -D. Feeny, and R.J. Oakerson. (1992). `Institutional dynamics: The evolution and dissolution of common-property resource management', in D.W. Bromley (ed.), Making the Commons Work: Theory, Practice, and Policy, Institue for Contemporary Studies Press, San Francisco, CA, USA, 129-160. Toulmin, S. (1974). `Rules and their relevance for understanding human behavior', in T. Mischel (ed.), Understanding Other Persons, Basil Blackwell, Oxford, UK, 185- 215. Walker, J. and R. Gardner. (1992). `Probabilistic destruction of common pool resources: Experimental evidence', The Economic Journal 102, 149-161. 90 Designing Institutions for Environmental and Resource Management

Williamson, O.E. (1979). `Transaction cost economics: The governance of contractual relations', Journal of Law and Economics 22, 233-261.

ANALYSIS OF FISHERIES CO-MANAGEMENT ARRANGEMENTS: A RESEARCH FRAMEWORK

Prepared by Fisheries Co-Management Project

Core Staff at the

International Center for Living Aquatic Resources Management (ICLARM)

And

North Sea Center (NSC)

September 1996

i 1. INTRODUCTION

The Fisheries Co-management Research Project is a collaborative project between International Center for Living Aquatic Resources Management (ICLARM), Institute of Fisheries Management (IFM)-The North Sea Center and National Aquatic Resource Systems (NARS) partners in Asia and Africa. It is based on a mutual interest to gain practical experience in research on fisheries co-management. There are two components to the research:

(1) Comparative case studies of fisheries co-management strategies on the basis of existing literature and country research.

(2) Co-management models based on (1) and tested at pilot sites.

This paper describes a research framework which has been developed by the project to carry out the above research. The aim is to provide a common analytical framework which will enable comparison between case studies, country research and pilot-tested co-management models. This will allow data to be analysed in a systematic way and allow generalisations to be made about conditions which facilitate successful fisheries co-management. The framework is in draft form because it is anticipated that once it has been applied in the analysis of case studies and country research, there might be a need to modify and/or expand it.

This paper is divided into two main sections. The first section briefly describes the theoretical background to the research framework, exploring the concepts of common property, fisheries co- management, institutional analysis and rights and rules. For more detailed coverage of the theoretical background, readers are encouraged to consult the bibliography. The second half of the paper describes the framework itself.

2. THEORETICAL BACKGROUND

2.1 Common property resources

The "commons" include natural resources, such as fisheries, wildlife, forests, irrigation waters and pasture lands, which by their physical nature are now owned by individuals but are shared by a community or group of users, such as fishers. The "commons" has come to connote inevitable resource degradation. Many accepted that fishery resources which are held in common are subject to overexploitation and degradation. The main rationale for this was based on Hardin's theory on The Tragedy of the Commons (Hardin, 1968) which concluded that "freedom of the commons brings ruin to all". The assumption was that when resources are limited and publicly owned, it is rational for each individual to overexploit them, even though this behaviour ultimately results in tragedy for the group (Acheson, 1989). Hardin's solution was either to privatise the commons or keep them as public property, to which rights of entry and use could be allocated i.e. privatisation or government control.

In more recent years, social scientists have observed that not all common property resources are subject to such a 'tragedy' and are not overexploited. This has led to considerable discourse on the subject and consequent rejection of the notion that it is the common property nature of the 2 resource which is the problem. What is important is not the type of resource i.e. common property, but the property rights regime in combination with the resource it is subject to, namely open access, private property, communal property and state property. The following definitions are given by Feeny et al (1990) to describe these regimes:

Open access: the absence of well defined property rights. Access to the resource is unregulated and free and open to anyone.

Private property: the rights to exclude others from using the resource and to regulate the use of the resource are vested in an individual or group. They are usually recognised and enforced by the state and are usually exclusive and transferable.

Communal property: the resource is held by an identifiable community of interdependent users who exclude outsiders while regulating use amongst members. The rights are unlikely to be exclusive or transferable and are often rights of equal access and use. Some inshore fisheries and shellfish beds are managed as communal property. The rights of the group may be legally recognised or de facto.

State property: rights to the resource are vested exclusively in the government which makes decisions concerning access to the resource and the level and nature of exploitation.

This separation between the nature of the resource and the property regime it falls under shows that Hardin's theory was correct inasmuch as it predicted a situation of a common property resource under an open access regime. However, other property regimes can and have also led to overexploitation, indicating that the provision of property rights alone is not enough. New methods of management are being investigated, in an attempt to take on the best aspects of state control, private and communal property. Largely from the management experiences gained in certain fisheries, and other common property resources such as forests and groundwater, it is recognised that what is needed is a more dynamic partnership using the capacities and interests of local fishers and communities, complemented by the ability of the state to provide enabling policies and legislation as well as enforcement and other assistance. This has been termed co- management.

2.2 Fisheries Co-Management

Co-management is defined as the sharing of responsibility and/or authority between the government and local resource users to manage a specified resource, e.g. fishery, coral reef. Lying between two management strategies - centralized control and self- or community management - co-management covers a a broad spectrum of management arrangements (Figure 1). The amount of responsibility and/or authority that the government and local resource users have will differ and depend upon country- and site-specific conditions. For this reason, co- management covers a number of arrangements ranging from government (fisheries administration) instructing user groups to user groups informing government on management arrangements they have developed. Co-management should not be viewed as a single strategy to solve all problems of fisheries management. Instead, it should be seen as a set of alternative management strategies, appropriate for certain areas and situations (Pomeroy and Williams 1994).

3 Figure 1 Spectrum of co-management arrangements (adapted from McCay, 1993 and Berkes, 1994)

Gove nment bas )d mans ement

User roup base mans ement

2 3 4 5 6 7

Government Instructs Consults Cooperate Advises Is informed Self Centralistic management management

Co-mana ement

2.3 Institutional Analysis

In conducting research on fisheries co-management we are essentially interested in understanding how rules affect the behavior and outcomes achieved by fishers using fisheries resources. Institutional analysis, which focuses on the institutional arrangements, the set of rights and rules by which a group of fishers and government organizes resource governance, management and use in collective action situations, will provide the framework for the research. It should be noted that not every detail of institutional analysis will be discussed in this paper. The reader is encouraged to make use of the highlighted publications in the reference list for more specific discussion of several aspects of institutional analysis.

The purpose of institutional analysis is to separate the underlying rules (institutions) from the strategy of the players (organizations). Institutional analysis examines how institutiona arrangements affect user behavior and incentives to coordinate, cooperate and contribute in the formulation, implementation and enforcement of. management regimes. When carrying out institutional analysis, it is also important to examine some aspects of organizations because their strategies can influence, or lead to change in, institutions.

Institutions are the "rules of the game" in a society and are affected by economic, social and political factors. Institutions can be both formal and informal, created or evolved. Any human interaction is governed by both formal rules (i.e. those that are written down) and informal rules or codes (i.e. those which everyone knows about but are not formalised in any way). North (1990) uses the analogy of American football to describe institutions. In football there are formal rules and usually unwritten codes of conduct which underlie and supplement formal rules such as 4 not deliberately injuring a player from an opposing team. The effectiveness of the rules is determined by whether they are enforced, the cost of enforcement and the severity of punishment.

Organisations, on the other hand, are groups of individuals bound by some common factors to achieve particular objectives. The origin of organisations and how they evolve is influenced by the institutional framework and in turn organisations influence how the institutional framework evolves. Organisations can be political such as a local council, economic such as a cooperative, social such as a church, or educational such as a school.

In general, institutional arrangements are defined by authority relationships that specify who decides what in relation to whom. The institutional arrangements can be be subdivided into several attributes: operational rules for governing resource use, collective choice rules for determining, enforcing and altering operational rules, and organizational authority relationships.

Ostrom (1994) states, "Action situations are perceived to be nested within at least three relevant tiers of action. Operational level actions are decisions which occur whenever individuals directly affect variables in the world by doing such things as harvesting products, worshipping at a forest shrine, planting seeds, building fences, patrolling the borders of a forest or feeding leaves to their animals. Collective choice actions constitute a group's decisions about operational activities, e.g., the actions taken at an annual meeting of a forest users' association to keep a forest closed for the harvest of a particular product except for a specified time. Constitutional choice actions are decisions about how collective choice actions will be made. An example is the resolution of a forest users' association to create an executive committee that will meet once a month to determine joint activities to be undertaken. Constitutional choices are frequently made without recognition that they are indeed creating a future structure to make rules about an operational action."

Institutional arrangements are sets of rights that fishers possess in relation to the fishery and the rules that define what actions they can take in utilizing the fishery. In order to organize their harvesting, for example, fishers must develop rules to establish how rights are to be exercised. Rules give substance to rights, structure a situation, define the behavior of the group members, and reduce conflict. Rules may create different incentive structures that affect cooperation or conflict among fishers. Rules structure human behavior into four categories: compulsory, permitted, authorized and non-authorized (Thomson 1992). The types of rules that are devised will depend upon the severity of the problem the fishers face, the level of information they possess, socio-cultural traditions, the extent of the bundle of rights they hold, the level of opportunistic behavior, and the ease with which actions can be monitored and enforced. Rules can provide stability of expectations, and efforts to change rules can rapidly reduce their stability (Ostrom 1990). It should be noted that the institutional arrangements fishers develop and use may not always be the same as formal laws and regulations. The fishers may develop institutional arrangements to meet their needs which are not recognized and legitimized by government. These informal or traditional rights and rules may be as or more important and credible to the local fishers than the formal fisheries laws and regulations.

The terms "rights" and "rules" are often used interchangeably in referring to the uses of natural resources. "Rights" refer to particular actions that are authorized (Ostrom 1990). A right is a claim to a benefit stream that is consciously protected, in most cases by the state. The possessor of the right has the expectation in both law and practice that their claims will be respected through the assignment of duty to others who may interfere with the benefit stream (Bromley 1991). To possess a right implies that someone else has a commensurate duty to observe this right (Commons 1968). Rights define the uses which are legitimately viewed as exclusive and the penalties for violating those rights. The specification of a right does not define how the right is to be exercised. How rights are exercised are defined by rules. Rules define specifically what acts are required, permitted, and forbidden in exercising the authority provided by the right. For 5 every right that an individual holds, rules exist that authorize particular actions in exercising the right. For example, a right provides the authority for a fisher to operate on a specific fishing ground. How the fisher exercises that right through the fishing activity is specified by rules which may dictate the type of fishing gear used or the time of year when the fishing gear can be used. Thus rules specify both rights and duties. The important aspect of rules in terms of institutional analysis is that they may create different incentives which affect cooperation among users. Schlager (1990) cited the example of two groups of fishers in an identical set of rights but with different rules determining how these rights can be exercised. These two groups may have different incentives to cooperate so that the result may be totally different outcomes. The more complete the set of rights, the less exposed the fishers are to the actions of others, and the less risk that the fishers face in organizing themselves in groups (Ostrom 1990).

For common-pool resources, the most relevant day-to-day or operational-level property rights are "access" and "withdrawal" rights. These are defined as:

Access: the right to enter a defined physical property.

Withdrawal: the right to obtain the "products" of a resource; e.g., catch fish.

Individuals who have access and withdrawal rights may or may not have a more extensive set of rights which authorize their participation in defining operational-level rights or what is referred to as collective-choice actions. In regard to common-pool resources, these additional rights include management, exclusion and transfer. These rights are defined as:

Management: the right to devise operational-level rights of withdrawal.

Exclusion: the right to devise operational-level rights of access.

Transfer: the right to sell or lease all or part of the above collective-choice rights (Schlager and Ostrom 1993).

The sources of the rights of access, withdrawal, management, exclusion and transfer are varied. These property rights nay originate by government which explicitly grants rights to fishers. These de jure rights are given formal and legal recognition. Property rights may also originate from fishers. Such indigenous or de facto rights, devised and enforced by fishers, are not usually recognized by government. These two types of property rights may overlap, complement or conflict with each other. While de facto rights may eventually be given recognition by government, until they are formally legitimized, they are less secure than dejure rights (Schlager and Ostrom 1993). While most authorities tend to ignore de facto rights, many have proven to be efficient and equitable in fisheries management. Ostrom (1991) identifies three levels of rules which are all closely linked. Operational rules govern and regulate resource use. Operational rules directly affect the day-to-day decisions made by the fisher concerning when, where and how to harvest fish; who should monitor the actions of others and how; what information must be exchanged or withheld, and what rewards or sanctions will be assigned to different combinations of actions and outcomes. Operational rules can be formal (written, legitimized) or informal (unwritten, customary/traditional). In both circumstances they are understood by those to whom they apply.

Several kinds of operational rules have been identified (Gardner and Ostrom 1991, Ostrom, Gardner and Walker 1994). Position rules specify what position a fisher may hold. Boundary rules specify who can enter the fishery. Allocation rules specify the procedures or actions for harvesting. Authority rules specify the authorized actions fishers may take independently. Scope rules specify the characteristics of fish that can be harvested. Information rules specify the information that fishers must reveal to others. Aggregation rules specify the procedures that fishers must follow for making decisions which involve multiple individuals. Penalty rules 6 specify punishment for non-compliance. Payoff rules specify the rewards or penalties that may be assigned to certain actions or outcomes. Input rules specify fishers requirements in terms of time, money and/or materials for management and participation.

Ostrom, Gardner and Walker (1994) state that in the IAD methodology, "... one has a conceptual tool for inquiry about how rules affect a given situation. For each variable identified in the action situation, the analyst interested in rules needs to ask what rules affect the variable as specified. For example, in regard to the number of participants, the analyst asks: Why are there N participants? How did they enter? Under what conditions can they leave? Are some participants forced into entry because of their residence or occupation?" "Answers to these sets of questions are formalized as a set of relations that, combined with the structure of a physical world and the type of community involved, produces the particular values of the variables of the situation." Ostrom, Gardner and Walker (1994) further state that "... a second advantage of examining the rules that directly affect ... an action situation is that doing so leads to a relatively natural classification system for sets of rules. Classifying rules by what they initially affect enables us to identify rules that all directly affect the same working part of the situation."

Operational rules are neither self-generating nor self-enforcing. Institutional arrangements are needed to adjudicate conflicts, enforce decisions, formulate and change operational rules, detect and sanction against rule violation, and hold officials accountable to users. These institutional arrangements are called collective-choice rules (Tang 1992). These are the rules used by fishers, their officials, or external authorities in making day-to-day decisions - the operational rules - about how the fishery should be managed. In a broad sense, collective-choice rules include qualifications for participation in the management organization and whether membership is compulsory. They state what proportion of the group of fishers must agree before a rule may be adopted. Of critical importance are the arrangements for monitoring and enforcing compliance with the operational rules and for settling disputes.

There may be multiple levels of collective-choice entities depending upon the situation. In some situations only one entity, such as a fisher's association, may be constituted to adopt and enforce their own collective-choice and operational rules. In another situation, multiple collective-choice entities, at national, regional and/or local levels, may subject fishers to multiple sets of operational rules. For example, national-level regulations may overlap with local-level regulations which may overlap with customary or traditional practices. Different levels of collective-choice entities may be constituted to deal with problems of different scopes. Issues of coordination and control must be addressed when multiple levels of collective-choice entities are in place (Tang 1992).

Finally, constitutional-choice rules affect operational rules by determining who is eligible to participate in the system and by establishing the process and rules by which collective-choice rules are created, enforced and modified. Constitutional-choice rules include, for example, the national fisheries policy and legislation which establishes a national fisheries agency and the fisheries administrative and management structure. This national institution has the mandate to empower lower level institutions to establish rules.

Ostrom, Gardner and Walker (1994) state, "Action situations are also linked across several levels of analysis. All rules are nested in another set of rules that, if enforced, defines how the first set of rules can be changed." "What can be done at a higher level will depend on the capabilities and limits of the rules at the level and at a deeper level. Changes in the rules used to order action at one level occur within a currently "fixed" set of rules at a deeper level. Changes in deeper-level rules usually are more difficult and more costly to accomplish, thus increasing the stability of mutual expectations among individuals interacting according to a set of rules."

Operational or working rules are nested within collective choice rules which are in turn nested within constitutional rules. In other words. the rules affecting operational choice are made within

7 a set of collective choice rules that are themselves made within a set of constitutional choice rules. To further complicate matters, all levels of rules operate at different levels: community, district, province, national, regional, and international. This means that the constitutional choice rules for a micro-setting are affected by the collective and constitutional choice rules for larger jurisdictions. A full understanding of behaviours and outcomes in managing a resource is only possible when information is analysed on how institutional arrangements at the operational level (resource use, monitoring and enforcement) relate to institutional arrangements in collective choice (policy, management and adjudication) and constitutional choice (governance, adjudication and modification).

Institutional arrangements are not organizations, and may not be associated with a formal organization. Examples of institutional arrangements operating independent of any formal organization do exist (Schlager 1990). Yet, organizations are important since they can serve to operationalize institutional arrangements.

A fishery management system may involve several discrete organizations, both formal, that is legitimately recognized, and informal, and the organizations may perform different management and service functions at different levels of management. There can be a variety of organizational forms for managing and governing the fishery which may range from fisher's organizations, to cooperatives, to nongovernmental organizations, to government sponsored organizations (Pollnac 1988). The organization will establish an agenda and goals which are to be achieved. This may include an identification of the problem or issue to be addressed and management and adjudication procedures. The authority system to ensure that fishers expectations are met is normally inherent in the organization. The analyst must recognize the mixture of organizations, both governmental and nongovernmental and internal and external to the community, and be able to identify the expansion and contraction of their roles under a co-management system.

A group of fishers can be regarded as organized to perform fishery management functions whenever it shares common understanding about:

(a) who is and is not a member; (b) the rights of fishers to access and use the fishery as conveyed by membership; (c) how decisions are made concerning the development of coordinated strategies of managing the fishery; (d) leadership roles; (e) membership responsibilities to sustain the organization; and (f) how conflicts are to be resolved (Ostrom 1992).

8 3. INSTITUTIONAL ANALYSIS RESEARCH FRAMEWORK FOR FISHERIES CO- MANAGEMENT

Based on the theoretical and conceptual framework described above in Section 2, an analytical framework has been developed for use by project researchers on fisheries co-management. The purpose of the framework is to describe and characterise the key factors which influence the institutional and organisational aspects of fisheries co-management arrangements. Such an analysis can then be used to make generalisations about the type of co-management arrangements appropriate for different situations. In particular, the analysis would enable:

(1) The identification of the existing property rights system in order to determine who defines rights to exploit the resource, who has access to the resource and whether any of these rights are transferable.

(2) The scale and level of user group involvement in order to determine the ways in which user groups do or can participate in co-management. Scale refers to the types of tasks which can be carried out by user groups, whilst level refers to the political level at which user groups are involved such as local, regional or national. Scale is related to level in the sense that different tasks can be carried out at different levels.

(3) The nature of the representation of user groups in the decision-making process in order to determine the participants in the co-management arrangement, which user groups are legitimate participants in the decision-making process and who can claim rights to participate (eg. fishermen, fish processors, consumers, environmentalists).

(4) The type of management organisation (existing or possible) in order to determine the type of co-management arrangement most appropriate for a particular fishery.

Ostrom, Gardner and Walker (1994) state, "At the conceptual level of a framework, theorists identify the broad working parts and their posited relationships that are used in an entire approach to a set of questions. Frameworks help to organize diagnostic and prescriptive inquiry." The institutional analysis research framework provides for a structured approach to document and evaluate the origin, current status, operation, impact and performance of fisheries co- management institutions and for designing modified or new co-management institutions. Oakerson (1992) has stated that, "Such a framework must be specific enough to offer guidance in the field, yet general enough to permit application to widely variable situations." The framework allows for the essential elements of the action situation to be identified and examined. The research framework is used to collect and organize information on key contextual variables which characterize collective action situations at multiple levels. The same set of contextual variables are used to describe and analyze all situations. These contextual variables will take on different values in different situations. By utilizing the same set of contextual variables, it is possible to conduct a systematic and comparative analysis of diverse situations and identify relationships among variables for evaluative, diagnostic and design purposes. As an evaluative tool, the framework can be used to describe a colleciive action situation. As a diagnostic tool the framework can be used to prescribe solutions to modify an action situation. As a design tool the framework can be used to develop and implement a new action situation (Oakerson 1992). It should be noted that the framework is not a causal model in which data is imputed and an output is generated. Rather, the framework is a method for logically arranging information, examining relationships among attributes and considering or describing outcomes. The framework can be used in different situations at varying levels of complexity and completeness as required.

The research framework is adapted from theoretical and empirical work on the Institutional Analysis and Development (IAD) framework developed by researchers at 9 the Workshop in Political Theory and Policy Analysis at Indiana University, USA. The research framework uses concepts from economics, political science, anthropology, biology and law and relies of methods described by Kiser and Ostrom 1982, Ostrom 1986 and 1990, and Oakerson 1992. The theoretical foundations of institutional analysis are based on game theory, neoclassical microeconomic theory, institutional economics, political economy, transaction cost economics, and public choice.

As specified for the fisheries co-management project, the institutional analysis research framework consists of three parts:

1 Institutional Arrangements Analysis: This component links contextual variables characterizing key attributes of the resource (biological, physical) and resource user (technology, market, social, cultural, economic, political) with the local fisheries management institutional arrangements (rights and rules). The contextual variables are referred to as a set since each is composed of a number of attributes. A causal relationship exists among and between the contextual variables, the local institutional arrangements (the focus of the analysis) and the resulting transactional (action) situations (Figure 2). The local institutional arrangements, structured by the contextual variables, affect the actions of the resource users by shaping the incentives and disincentives they face to coordinate and cooperate in resource governance, management and use, the incentives, in turn, shape the patterns of interaction that result when resource users select and implement fishing strategies; these interactions result in outcomes. These outcomes may, in turn, affect other outcomes. Time is a critical element. All the contextual variables can change through time. This causes change in institutional arrangements which, in turn, affects incentives, patterns of interaction and outcomes.

The six sets of contextual variables are:

a. Fisher or community institutional and organizational arrangements; b. Biological, physical and technological; c. Market (supply and demand); d. Fisher, stakeholder and community; e. External institutional and organizational arrangements; and f. Exogenous (macroeconomic, social, political, natural).

2. Institutional and Organizational Performance: The outcomes of co-management institutional arrangements can be evaluated in terms of performance, that is, the meeting of management objectives and the impact on the resource and its users. Performance can be evaluated in terms of different criteria such as sustainability, efficiency, equity, replicability, adaptability, and manageability.

3. Characteristics of Successful Fisheries Co-management Institutional Arrangements: The most important aspect of this analysis is the specification of what conditions bring about successful long-enduring, locally-sustainable and self-governing fisheries co- management institutions. From the analysis we can identify a list of principles and propositions about conditions for the development and sustainability of effective fisheries co-management institutions. The three parts of the research framework build on each other. The first part provides a framework for describing what is occurring in the collective action situation and specifying relationships between organizational and institutional arrangements. The second part is an assessment of how well the management institution is performing and what are its impacts. The third part is a determination of what characteristics or conditions of the management institution lead to successful co-management. For example, once described, a performance evaluation of a co-management institution may conclude that it is biologically sustainable but not equitable. As a result those attributes of the action situation which lead to sustainability can be identified and

10 used to develop conditions for successful co-management institutions. Those conditions which lead to inequity can be identified and prescriptions made to improve the situation.

The analysis of an action situation focuses on how rules combine with various contextual variables to structure the action situation and to generate particular types of outcomes. Implicit and explicit assumptions about rules and the contextual variables all influence the way an action situation is analyzed (Ostrom, Gardner and Walker 1994). The institutional arrangements analysis begins with an identification of variables affecting the action situation; the institutional, biophysical, technological, market, socio-cultural, economic, and political attributes and conditions of the resource users and the resource. These variables form the context within which fishers and other resource stakeholders coordinate and cooperate to establish institutions and organizations to govern, manage and use the fishery; from which emerges some patterns of interaction. Patterns of interaction result directly from the mutual choice of strategies by members of the group. These patterns of interaction result in outcomes.

In analyzing institutional arrangements, the basic strategy is to separate and dissect the parts of the action situation - contextual variables, incentives, patterns of interactions and outcomes (Figure 2); identify and collect data on the attributes and conditions of each part; and examine the relationships between and among the attributes and conditions of each part. The purpose of dissecting the action situation is to examine relationships between and among the parts. Each part of the framework has a causal relationship with other parts, some stronger and some weaker depending upon the involvement of human choice in the relationship. Biophysical and technological attributes can have a direct affect on outcomes, for example, high levels of fishing effort can lead to overexploitation, regardless of whether or not institutional arrangements are in place. Institutional arrangements, on the other hand, have an indirect affect on outcomes as they lead to changes in human behavior and choice, which affect interactions and outcomes (Oakerson 1992). Different combinations of these parts can be examined depending upon the situation. These relationships can be analyzed forward or backward depending upon if one is using the framework as an evaluative, diagnostic or design tool. Explicit and implicit assumptions about the relationships help structure and guide the analysis.

The various biophysical, socioeconomic and institutional attributes of the contextual variables which may affect collective action situations are identified. These attributes include who is involved in a situation, what their stakes and resources are, the types of actions the can take, the costs of those actions, what information is available to them, how much control individuals can exercise, how individuals and actions are linked to one another and to outcomes, the outcomes that can be achieved through what actions, and how rewards and punishments are allocated to particular combinations of actions and outcomes. Similar to the various parts of the action situation described above, the attributes are interrelated, a change in one attribute may create a new structure of incentives, resulting in different outcomes. Based on the contextual variables, one tries to explain or predict the patterns of relationships and the interactions and outcomes that are most likely to occur for an action situation, given the incentive structure. These explanations and predictions for what is observed are then verified in an empirical setting. The performance of the outcomes are then evaluated using established criteria (Ostrom 1990, Tang 1992, Ostrom, Schroeder and Wynne 1993).

As mentioned earlier, the institutional arrangements analysis framework can be used as an evaluative and diagnostic tool by working backwards through the relationships (Oakerson 1992). Thus, the analysis focuses on outcomes. What is happening with the fish stocks? Are they increasing, stable or decreasing? Are returns from the fishery increasing, stable or decreasing? Are they being distributed equitably among the users? Is the quality of life of the fishers and their families improving, stable or worsening from previous years? The next question is why? Oakerson (1992) states, "A first order answer can be obtained by examining patterns of interaction among resource users. Are members of the community competing with one another to maximize their individual 'take' from the commons? The inquiry cannot stop, however, with

11 patterns of interaction. The question of why recurs. Second-order answers depend on how ..." contextual variables, together with institutional arrangements, jointly affect patterns of interaction. Do fishers face a high level of uncertainty over fish availability and are they highly dependent upon the resource with limited livelihood options which would provide an incentive for low compliance with rules and non-conforming behavior?

"Outcomes disclose the effect of a difficulty that is manifested behaviorally in patterns of interaction" (Oakerson 1992). The reason for the difficulty is that the institutional arrangements do not adequately address the problems generated by the contextual attribute, for example, the biophysical characteristics of the fishery. This "lack of a good fit" between the biophysical attributes of the resource and the institutional arrangements used to manage and govern its use leads to a non-optimal incentive structure resulting in counterproductive patterns of interaction and undesirable outcomes (Oakerson 1992).

The framework can also be used as a design tool to modify or develop collective action situations. Design requires the analyst to work forward through the framework. The analyst must identify key contextual attributes and what problems exist in the community. The analyst then evaluates what institutional arrangements are needed to address these issues. This is followed by a series of questions such as: What incentives for interaction will these institutional arrangements produce? What strategies will fishers select based on the incentives? How will this affect the resource and the fisher?

In a short-run analysis of an action situation, the contextual variables are assumed to be unchanging. Over a longer period, however, change will occur in them. Yields may increase, gear type may change or the day-to-day rules may be restructured. A dynamic element can be introduced into the framework. One approach treats institutional changes as exogenous; the aim is simply to understand how a series of changes in resource attributes or institutional arrangements affects patterns of interaction and outcomes. Another approach examines long- term relationships between attributes and institutional arrangements in an iterative and causal fashion. For example, outcomes can affect patterns of interactions resulting in a process of learning by the fishers; causing, in turn, individuals to modify their strategies. These relationships can be traced through the framework to identify factors which cause the strategies to change (Oakerson 1992).

4. THE RESEARCH FRAMEWORK

The following sections, describe each of the main framework components in more detail. These sections focus on what are considered to be the key attributes which affect incentives and patterns of interaction in a fisheries co-management situation. It should be emphasised that the attributes are not definitive and might vary according to situation. As research gets underway, other key attributes might be identified and existing ones rejected. However, the purpose is to focus research on critical attributes to ensure that considerable time is not spent collecting an exhaustive amount of information which ultimately turns out to be of little use for diagnostic or prescriptive purposes for fisheries co-management.

12 1. Institutional Arrangements Analysis

1. Contextual Variables

(1) Biological, physical and technological attributes.

Problems and constraints over resource use most often originate in the biological and physical attributes of the resource and in the harvesting technology used. The nature of interactions among fishers are commonly structured by the biophysical and technological environment of the fishery. Fishers vulnerability to scarcity and uncertainty in fish supply and its effects impact upon their incentives to engage in collective action. Collective action situations have been shown to develop when a group of individuals are highly dependent on a resource and when availability of the resource is uncertain or limited. If the resource availability problem is repeatedly experienced, such as low or no catch, and if it exists within a single community of users, the fishers are likely to develop institutional arrangements to deal with the problem. To understand the actions fishers have taken and the institutions that they have developed requires an understanding of the fishing grounds, fish stocks, fishing activity, boundary conditions and fishing technology. Key questions to be considered are shown in Table 1.

Oakerson (1992) has identified three considerations for analyzing these attributes:

(a) The relative capacity of the fishery to support many fishers simultaneously without mutual interference and/or without diminishing the aggregate yield of the fishery for the group (subtractability). The harvesting activity of an individual fisher subtracts from the amount of fish available for other fishers to withdraw. The catch of one fisher affects the amount of fish that can be harvested by other fishers utilizing the same fishing ground. Oakerson (1992) states, "The analysis should specify as precisely as possible the 'limiting conditions' that pertain to natural replenishment or maintenance of the resource. Physical limits established by nature or technology provide critical information for devising rules to maintain jointly beneficial use."

(b) The degree to or relative ease with which access to the fishery is limited (exclusion). The physical nature of fishing grounds means that exclusion (or limiting access) of fishers is both difficult and costly. A single fisher would find it difficult to exclude other fishers, therefore, fishing grounds are subject to joint use. Oakerson (1992) states, "Two types of exclusion can be distinguished: (1) access may be fully regulated on an individual basis, or (2) it may be partially regulated and applied only to those outside the immediate community. This distinction is related to the potential exposure to increases in demand. Within a definite community of users, increases in aggregate demand derive mainly from expanded operations. If there is open access, however, increases in the number of users can also contribute to an increase in total demand on the resource." Thus, a fishery is characterized by high levels of interdependence among fishers. The action of one fisher affect the actions and outcomes of other fishers. These interactions can lead to conflicts among fishers over space and amount of fish (Schlager 1990).

(c) The spatial boundaries of the fishery, which determine the minimal scale on which effective coordinated resource management can occur (indivisibility). Physical boundaries having to do with divisibility of the fishery derive from nature, human design and technology. Fishing gear type, terrestrial and oceanographic features, customs, culture, government, organizations and scale all dictate the division or partitioning of the fishery into smaller units for management purposes (Oakerson 1992).

The basic institutional forms for fisheries management are fundamentally shaped by these three characteristics of the resource. In addition to the above, two other concerns are important. Technological problems occur when fishers physically interfere with each other in the fishing

13 activity. Gear conflicts may occur or the placement of gear may interfere with the flow of fish, often referred to as crowding. Assignment problems occur when fishers, desiring to fish the most productive spots, fail to allocate themselves efficiently across spots, leading to conflicts (Schlager and Ostrom 1993).

If the attributes of subtractability and exclusion are arranged in a simple matrix, four types of goods can be identified: (1) private goods, (2) public goods,(3) toll goods and (4) common pool goods (Savas 1987, Thomson 1992, Tang 1992).

(1) If a resource is consumed individually and it is possible to exclude some people from access to the resource, it is a private good. An example of this is a fish consumed by a fish farmer which has been produced in her fish pond on property that she owns.

(2) Public goods are those to which access cannot be easily controlled (exclusion is infeasible), especially free riders, and consumption is joint. An example of this is improved environmental quality in a region, such as air, resulting from maintaining the environmental health of the fishery.

(3) If a resource is consumed jointly but it is still possible to exclude access to it, such as for free riders, then it is called a toll good. An example of this is a water bottom area for clam harvest for a group of fishers who purchase use rights. The group harvests clams in the area together and exclude others from fishing in the area.

(4) Common pool goods are those to which access cannot be easily controlled and consumption is separable. An example of this is a small-scale fishery where there are a relatively large number of fishers using different fishing gear types in a situation where effective regulation is not feasible. Common pool goods can be subdivided into common property and open access goods. Common property goods are managed to some extent. Access to and exploitation rates of the common property fishery are controlled by an identified group of users. Open access goods are not managed. Neither access nor exploitation are controlled.

A fishery could be either a private, public, toll or common pool good or a combination of these goods, depending upon the local biophysical, technological, and social setting. A fishery may be a private good at one point (such as along a shoreline) and a common pool good at another point (such as the coastal waters). Which type of good a fishery resembles tends to determine (within broad parameters) how social or elaborate the fishery management institution will need to be, and to what extent the government is likely to be involved in management. Ostrom (1990) points out that changes in biological, physical and technological attributes can potentially transform a resource from one type of good to another, which in turn is likely to lead to pressures for other basic institutional changes. Diverse biophysical factors, gradual trial-and-error experiments and negotiating lead to the emergence of site specific institutional forms, which continue to evolve over time. Characteristics of excludibility and subtractability or jointness of consumption which characterize different types of goods strongly influence the incentives for fishers to use, manage and govern the goods.

14 Table 1 Biological, physical and technical attributes and indicators

Biological, physical and technical Key questions attributes Multi-species or single species fishery What species are caught?

Migratory or sedentary fishery resources Are the fish resources sedentary or migratory?

Level of stock exploitation Have total catches been falling over time? Is catch per unit effort increasing or decreasing? What do stock assessments say? Status of habitat What is the % coverage of coral (where applicable) Are the fish and plant resources healthy? Is the water polluted? Boundaries Are there geographical boundaries for the fishery? How are they defined? Are there restrictions on who can enter the fishery? How are these defined? Are there any other boundaries relevant to the exploitation of the fishery? What are they? Single or multiple gear fishery What types of gears are being used?

Artisanal or industrial fishery' What gear and vessel types are there? What is the range of fishing operations? Can the fishery be divided into artisanal and industrial? How are these terms defined? Level and mix of technology What are the gear and vessel types? What are the technologies for preservation/processing of fish? Dispersed or localised fishing patterns Is fishing a seasonal activity? Does fishing take place in a localised area? Is fishing carried out offshore?

(2) Market (supply and demand) attributes

Resource problems are often market-based. Market attributes (price, structure, stability) can effect the incentives for resource use activities, effort levels and compliance with rules. These attributes include those related to the operation and function of the market and those related to fisher and fish trader relationships. Some of these market attributes include:

(a) stability of supply and demand in terms of price and quantity. Are there periods of scarcity of supply and/or high market demand? What effect does this have on price?

(b) market availability and location. Are there ready and consistent markets for product. Is the fishing community isolated? Do fishers sell primarily to local markets or are there regional, national and international markets?

i There is no standard definition of artisanal and industrial. Smith (1979) pointed out that it is more useful to talk about ranges of the technical and socio-economic characteristics of fishing activities. What is considered artisanal in one country may be considered industrial in another. For the purposes of the analysis presented in this paper, what is meant by the term artisanal is lower technology fishing with limited fishing ranges, often, but not always, for subsistence needs. What is meant by the term industrial is higher technology fishing with greater fishing ranges., predominantly for commercial purposes.

15 (c) market structure. How competitive is the market at different levels within the market channel? How free is the flow of market information?

(d) credit/market relationships. Are there strong credit and marketing linkages between fishers and fish traders and between fish traders and other business people?

(e) changes in market. Have there been recent changes or "shocks" to the market in terms of structure, price or stability?

(f) market operation. Are there rules in terms of who can participate in the market and formal or informal operating procedures?

Key questions to be considered are shown in Table 2.

Table 2 Market Attributes and Indicators

Market Attributes Key questions

Subsistence or market oriented fishery What proportion of the catch sold/consumed?

Market structure Are there many buyers? Are there many sellers? What are the power relations between buyers and sellers? Do women service other market segments than men? What are these? Market orientation Are the fish/fish products sold in local, domestic or international markets? Value of products Is the value of fish/fish products high or low?

(3) Fisher, stakeholder and community attributes

Fisher, stakeholder and community attributes can all affect incentives to cooperate. These attributes include religious beliefs and practices, traditions and customs, sources of livelihood, the degree of social, cultural, economic and locational heterogeneity or homogeneity, asset ownership, community mores, level of community integration into the economy and polity, and others. Whether individual attributes or in combination with others, each of these attributes potentially affects collective action and outcomes. General assumptions about fishers and stakeholders are related to how they behave both individually and in groups. Stakeholders, those individuals indirectly dependent upon the fishery for their livelihood, such as fish traders, processors and transporters, are also included since their relationship with fishers can provide incentives or disincentives for the fishers to cooperate (Table 3).

Institutional analysis involves attempting to understand the behavior a representative individual will adopt in a certain situation in light of their interests. Fishers and stakeholders "are assumed to be self-interested, rational individuals who operate within a context of rules which specify lawful and unlawful behavior. Individuals are also assumed to make decisions under conditions of uncertain information and to adopt maximizing strategies." (Thomson 1992). Some literature suggest two key attributes which lead to incentives to cooperate (Ostrom 1990, 1992, Runge 1992): (1) if a community of fishers exhibits a high degree of social, cultural and economic homogeneity in terms of kinship, ethnicity, religion, interests, beliefs, customs, livelihood strategies, etc.; and (2) if there is a high dependence or reliance of fishers on the 16 fishery for their livelihood and the number of alternative livelihoods available to fishers in the community is low. If the fishers are highly dependent upon the fishery and if the availability of the resource is uncertain or limited, fishers are more likely to facilitate collective action to deal with the problem.

Table 3 Socio-economic Attributes and Indicators

Socio-economic attributes Key questions

Homogeneity/heterogeneity of users How many ethnic groups are there among users? Are there differences in wealth, religion, gear types, residency amongst users? What are the vessel/gear/technology ownership structure? Who are the users/stakeholders (female/male)? Dependence on the fishery for livelihood What proportion of household/family income comes from fisheries? Motivation of users Are users exploiting the fishery for subsistence or commercial reasons? Attitudes towards: risk, innovation, Are these attitudes: Strong/weak/indifferent? collective action Is their heterogeneity in these attitudes? Is there any social/political mechanism, that prevent risk taking, innovation and collective action? Level of information and knowledge on What kind of knowledge (indigenous or scientific)? the fishery and management

4. Fisher and Community Institutional and Decision-making Arrangements

Decision-making arrangements are concerned with how institutional arrangements, rights and rules, are made. There are three important aspects of decision-making: representation, relevance and enforceability.

(1) Representation is the extent to which users and stakeholders participate in rule making.

(2) Relevance is the extent to which the rules are considered relevant to the management problem.

(3) Enforceability is the extent to which the rules are enforced.

The key questions which are considered critical are given in Table 4.

17 Table 4 Decision-making arrangements and indicators

Decision-making Arrangements Key questions

Leadership/power structure of user What are the main characteristics of the leadership? groups Are the leaders considered legitimate? How are decisions taken eg. consensus, majority? Main types of rules (operational, What are the formal and informal rules on access rights to the fishery? collective choice, constitutional What are the rules on harvesting rights? choice) and at what level of What boundary conditions (social, economic, physical, legal) apply to applicability (local, regional, the regulation of the fishery? national). Decision-making process for What are the main factors determining the way in which decisions are operational and collective choice made for (one or two examples of) operational and collective choice rules rules? How are decisions made (consensus/majority/autocratic)? Which types of decisions are made through consensus/majority/autocratically Level of representation in the Who are the users/stakeholders? decision-making processes at Are they represented in the decision-making process and how? different levels (local, regional, What is the level of participation of user groups/stakeholders in the national) decision-making process? Are women represented and how? Relevance of rules What are user and stakeholder attitudes/values towards decision- making processes? What are user and stakeholder attitudes/values towards rule-breaking? How much legitimacy do the rules have with users? Enforcement of rules and How are formal and informal rules enforced? regulations/sanctions What sanctions are used? How do users perceive the utility of the rules? What is the level of compliance/non-compliance e.g. number of convictions, user attitudes; obedience/obstruction? What resources are available for monitoring and enforcement?

(5) External institutional and organizational arrangements attributes

Institutional and organizational arrangements external (international, national, regional, district or municipal levels) to the fisher or community level are relevant and can affect fisher or community institutional arrangements. However, the connection can vary widely. Some community-level arrangements may be subject to constitutional approval (constitutional-choice rules) to engage in collective choice. A community-level arrangement may be dependent on government for enabling legislation or enforcement of operational rules. An outside party may be needed to settle disputes between fishers. Fishers or communities may require outside assistance to facilitate and strengthen organizational capacity or to strengthen market forces. In some instances, fishers or community may need outside intervention to address problems outside its institutional mandate which impact upon; local institutional arrangements and conditions (Oakerson 1992). There may be nested multiple layers of external institutional arrangements, with and without formal linkages, at both governmental and nongovernmental levels.

18 (6) Exogenous (macroeconomic, social, political, natural) attributes

A variety of factors exogenous to the fishery resource, fisher and community may have an impact on fisher or community institutional arrangements. These are factors which are beyond the control of the fishers and community, and at times higher level entities. These are surprises or shocks to the community or management system, brought about by macroeconomic, social, political or natural occurrences or interventions which affect the survival of the institutional arrangements. They may include typhoons, civil unrest, elections and inflation. They should be viewed in an historical and dynamic perspective.

These factors can provide an indication of how well the institutional arrangements are functioning and surviving through its capacity or resiliency to accommodate sudden change. Bromley (1992) states that all large changes in exogenous factors threaten the capacity of individuals to learn about change fast enough to make adaptive responses. The faster and greater the change, the higher the probability that the institutional arrangements will not be able to respond rapidly enough.

II. Incentives to Cooperate and Coordinate

The contextual variables, and specifically the institutional and decision-making arrangements, shape the incentives for individuals to take certain actions (cooperate, engage in collective action, coordinate activities) to achieve desired outcomes. The focus of the analysis is on how rules are devised, what the rules contain, whether the users consider them legitimate, and whether they are enforced.

An individual member of a group relies on reciprocal behavior from other members of the group regarding their adherence to the agreed-upon rules for management. An individual's choice of behavior in a collective action situation will depend upon how he or she weighs the benefits and costs of various alternatives and their likely outcomes. An individual's choices are often affected by limited information and trust in the actions of others which leads to uncertainty and by the level of opportunistic behavior or self-interest that individual fishers can expect from other fishers. Uncertainty often creates obstacles for exchange and cooperation among fishers(Ostrom 19911, Ostrona, Gardner and Walker 1994). 'Individual's also have differing discount rat: s; many poor fishers, for example, attributing less value to benefits that they expect in the future, and more value to those expected in the present.

In some situations, individuals may have incentives or opportunities, resulting from unevenly distributed information and power among the group involved in an action situation, to adopt opportunistic strategies to circumvent the rules and to obtain disproportionate benefits at the cost of others (Runge 1992). Three types of opportunistic behavior may occur: free riding, corruption and rent seeking. Free riders respond to incentives to engage in other activities while other members of the group work. Corruption can occur when incentives exist for rules to be changed for an individual through, for example, the provision of illegal payments. Rent seeking, the gaining of excess profits, can occur when an individual's assets, for example, property rights, increase in value through special advantages (Ostrom 1992, Tang 1992). The imperative of the group is to develop institutional arrangements which reduce or minimize transaction costs, reduce uncertainty and counteract opportunistic behavior.

For institutional arrangements to be maintained over time, it is important to develop workable procedures for monitoring the behavior of fishers, enforcing against non-conforming behavior with sanctions, and settling conflicts. The ease and costliness of monitoring rules devised to organize the fishing activity depend upon the physical nature of the resource, the rules-in-use, and the level of conformance to the rules (Ostrom 1990). The number of times that non- conformance must be measured affects the cost of monitoring. The ease and cost of monitoring 19 will also depend upon whether the fishers can monitor compliance themselves, as they fish or through self-monitoring incentives, or if they must establish more elaborate arrangements, such as external authorities.

Tang (1992) states, "Both transaction cost economics and institutional analysis are concerned with identifying appropriate institutional arrangements that can counteract perverse incentives inherent in various transaction situations. Whereas transaction cost economics approaches the problem by examining the characteristics of different transaction situations, institutional analysis explicitly identifies a higher level of analysis by delineating the contextual attributes that shape various action situations. At the contextual level of analysis, one examines how rules, physical attributes, and attributes of the community shape various action situations."

III. Patterns of interactions among resource users

While incentives to cooperate might exist, this does not guarantee that stakeholders and users will cooperate. Much will depend on the way resource users interact with each other and their behaviour both as individuals and as a group. (Oakerson, 1992). There is some overlap between incentives and patterns of interaction but the purpose in separating them is to assist the analyst in determine the likely reasons for a lack of cooperation when the incentives are thought to be in place.

To understand how fishers react according to incentives in the action situation, the analyst must systematically analyze the contextual attributes which combine to shape the incentive and constraint structure faced by the fishers. Given these contextual attributes and the incentives and constraints they generate for collective action, individuals make choices and develop strategies of action, from which emerges some patterns of interaction. Institutional arrangements do not generate patterns of interaction, but they shape and generate the choices and strategies individuals make in relation to others members of a group (Oakerson 1992). individual fishers will develop strategies to maximize their own self-interest within the bounds set by the institutional arrangements. Individual fishers will take advantage of the opportunities available to them and will comply with the rules to the extent required by those who enforce the rules (Thomson 1992). Individuals will adopt strategies which change with the level of information available to them and the level of reciprocal interaction from others. An individual's strategy to cooperate will depend upon assurance that others will do likewise. Patterns of mutual action will emerge. Thus, in the first period an individual will monitor the behavior of others. If he observes successful collective action, he will revise his strategy in the next period to increase the level of cooperation and compliance. He will continue to revise his strategy over time based on new information about the resource and its users (contextual attributes), the degree of reciprocity, and the effectiveness of monitoring and enforcement.

IV. Outcomes

Outcomes are produced as a result of patterns of interaction which are, in turn, the result of the strategies adopted by the fishers. The consequences of the patterns of interaction affect both those involved directly in the action situation and those indirectly involved. Various contextual variables can be associated with different outcomes. They may affect outcomes either positively or negatively depending upon the association. Patterns of interaction can produce biophysical, technical, social, economic, institutional and managerial outcomes. These include, for example, current resource yield, distribution of benefits and costs, livelihood and quality of life of fishers, knowledge of the fishery and its users, and degree of rule compliance. The specific outcomes may vary among fishers and communities and through time. Outcomes can be evaluated using different criteria. Of particular interest in evaluating outcomes in the question: How do predicted outcomes conform to evaluative criteria (Ostrom, Gardner and Walker 1994).

20 2. Institutional and Organizational Performance

The outcomes of co-management institutional arrangements can be evaluated in terms of performance, that is, the meeting of management objectives and the impact on the resource and its users. It is expected that in certain situations co-management institutional arrangements will perform better than other types of fisheries management institutional arrangements, such as centralized management. It is also expected that co-management will lead to more sustainable, efficient and equitable fisheries management.

The performance of co-management institutional arrangements can be evaluated at two levels. The first level of evaluation relates to overall institutional performance of co-management in meeting its stated advantages versus other types of management arrangements. These advantages include equity, more economical in terms of administration and enforcement, increased sense of ownership of the resource by users, self-management, higher degree of acceptability and rule compliance, improved information about the resource, improved social cohesion in the community, and more participatory (Pomeroy and Williams 1994). A comparative assessment of the performance of different co-management institutional arrangements can be conducted at this level of evaluation.

The second level of evaluation relates to performance in meeting specific management objectives and impacts at the operational level. Each individual co-management system has objectives established by the participants, both resource users and government, that are to be achieved. Performance of these systems is conducted to determine how well they achieved the objectives and what the impacts of the management activity are on both the human and biophysical environment.

The most common evaluative criteria are efficiency, equity and sustainability. Other criteria can also be used. These abstract evaluative criteria must be converted into operational measures of value for use in appraising specific outcomes (Oakerson 1992). Feeny (1992) has identified four methodological standards that operational measures of performance should meet. Reproducibility and reliability measures mean that the criteria should, for example, be able to be used by different observers to evaluate the same situation and get answers in close agreement. In addition, the measure should be applicable in different settings. "Validation may be demonstrated by showing that the operational measure that has been obtained inexpensively is highly correlated with an accepted and previously validated measure - the 'gold standard"'. Lastly, a measure should be responsive and "... able to capture change, if in fact it has occurred, while providing stable scores for situations in which there has been no change."

Efficiency

There are various measures of efficiency. The first aspect of efficiency is whether fishers have achieved an optimal rate of use of the fishery (Oakerson 1992). A less rigorous criterion is that fishers are not exceeding the sustainable yield of the fishery. A second measure of efficiency has to do with the flow of benefits resulting from the co-management institutional arrangements and the costs (such as transaction costs) of establishing and maintaining such arrangements. Ostrom, Schroeder and Wynne 1993) state, "Economic efficiency is determined by the magnitude of the change in the flow of net benefits with an allocation or reallocation of resources." A minimal efficiency criterion is that the benefits of operating and maintaining such arrangements exceed the full set of direct and indirect costs. A comparative efficiency criterion is that the difference between the benefits and the costs of co-management institutional arrangements in one setting is the same as or greater than those of similar arrangements in a similar setting elsewhere (Ostrom 1992).

21 As management processes are established to achieve particular objectives, the cost-effectiveness of the process compared to others has to be evaluated. One of the purported advantages of co- management compared to centralised management is that it will reduce transaction costs - the costs of information-gathering and processing, coordination of decision-makers/ user groups and regulation enforcement. Some of these costs remain fixed regardless of the management regime, such as information which is required by law. Other transaction costs vary with the quality of data and the process used to make decisions. Hanna (1994) points out that a centralised approach is often associated with low programme design costs but high implementation, monitoring and enforcement costs as the management regime may have little legitimacy with user groups. A co-management approach, on the other hand, is associated with high programme design costs as effective participation is time-consuming and therefore costly. However, co-management is likely to lead lower implementation, monitoring and enforcement costs as legitimacy of the regime is greater (Hanna 1995).

Equity

Equity (fair treatment for all people involved in managing, governing and using the resource) has four main components (Hanna 1995):

(a) Representation: a more equitable management regime should represent the range of interests in the fishery and accommodate the full diversity of those interests.

(b) Process clarity. the management process should have a clear purpose and a transparent operation.

(c) Homogenous expectations: the extent to which participants have similar expectations concerning the management process and it's objectives

(d) Distributive effects: the management process should address the distributional changes embedded in the options under consideration

Equity can be measured in several ways. First, is the distribution of an individual's return on contributions to the management and governance effort roughly similar to the benefits they receive? Oakerson (1992) states, "The presence of inequities may lead to the collapse of reciprocity, resulting in less efficient use. Equity problems are apt to be aggravated by asymmetries (unequal proportions) among users, which create opportunities for some benefit at others' expense. This, in turn, can lead to costly conflict where all parties lose." Corruption and abuse of authority may contribute to inequities. A second measure is to determine if there are patterns of redistribution that fishers wish to achieve at this level of institutional arrangements. Questions to be asked include: (a) whether most fishers in the community are relatively satisfied with the institutional arrangements; (b) whether any fishers have been consistently disadvantaged as a result of the institutional arrangements; (c) whether the distribution of resources and wealth in the community has changed; (d) whether allocation of membership and rights in the organization is fair; and (e) whether individual's are receiving benefits based on differential abilities to pay.

Sustainability

Sustainability can be divided into stewardship and resilience. Stewardship, the tendency for resource users to maintain productivity and ecological characteristics of the resource, is divided into three components: time horizons, monitoring and enforcement. To promote resources stewardship, the management process should expand time horizons beyond the short term. A 22 sense of stewardship will be more likely if the effects of the management regime can be monitored and where necessary, enforcement measures taken.

The other aspect of sustainability is resilience. This is the ability of the system to absorb and deal with changes and shocks. The three components of resilience are rule flexibility, structural adaption and market adaption. Rules should be flexible enough to respond quickly to changing conditions. The management regime should be able to adapt to both changes in the structure of the industry as well as changes in the market (Hanna 1995).

Ostrom (1992) identifies several other criteria by which performance can measured. One is concerned with the "match" of the membership of the management and governance organization and the fishers. Ostrom (1992) states, "A key factor that affects the long-run performance of an organizational arrangement is whether organizations can be established and maintained whose boundaries are roughly coterminous with those of the..." resource and its users. Mismatches can take two forms. One is a case where an organization is considerably larger than the fishery or the number of fishers. This might result in a total indifference to the problems of regulating the fishery. A second would arise where the organization is substantially smaller than either the territory of the fishery or number of fishers. If only a small number of the total fishers are involved in the organization than the gains from the arrangements would be minimized. Operational patterns of the organizational and institutional arrangements must be examined before concluding a mismatch exists.

A second measure involves the knowledge generated by fishers about the resource and about user preferences, benefits and costs. It is highly improbable that an organization could survive in the long-run and perform efficiently and equitably without detailed information about the characteristics of the fishery and use patterns. Without this information, rules to structure resource allocation would not achieve desired results. Thus, performance can be measured by the increase in information about the resource and the exchange of information among resource. users.

A third measure is the level of rule compliance. The willingness of fishers to regularly follow operational-level rules reflects the viability of the rules as coordinating devices. This measure may be included with other measures in a group of performance criteria called "governance". In addition to the level of rule compliance other measures of governance performance may include overall reduction in conflict, existence of an effective conflict resolution mechanism, and existence of practical and implementable enforcement procedures.

3. Characteristics of Successful Fisheries Co-management Institutional Arrangements

The most important research task for an institutional analysis of fisheries co-management is to specify conditions and propositions for successful development of co-management institutions and which arrangements are most favorable for maintaining it. A number of questions need to be answered: Why are some co-management arrangements successful, while others fail? Why do some co-management arrangements endure for long periods of time? How can we improve the success rate for implementation of co-management arrangements?

The success of co-management institutional arrangements may be related to specific contextual variables or attributes, and these may affect the development and maintenance of the arrangements. The research framework provides a means to identify attributes which lead to successful co-management from those which lead to failure. For example, specific biophysical or fisher attributes, such as boundary definitions or fish stock characteristics or social homogeneity of the community, may be critical factors for success of co-management. They may be more critical than the institutional arrangements themselves. By identifying these attributes and then examining their relationship with patterns of interaction and outcomes, it is possible to specify

23 conditions and propositions which can lead to successful development and maintenance of fisheries co-management institutional arrangements.

Over the last decade, research done at different locations around the world has documented many cases of co-management and community-based management in fisheries and other natural resource systems. From the results, certain conditions are emerging which appear to be central to the chances of developing and sustaining successful co-management institutional arrangements. Ostrom (1990, 1992) and Pinkerton (1989) have made useful contributions to our existing knowledge about key conditions for successful fisheries co-management. These key conditions are discussed in Pomeroy and Williams (1994). These conditions should not be taken as complete as continued research is needed to reveal more about co-management arrangements and the factors leading to successful performance.

5. CONCLUSIONS

There are a number of research frameworks and alternative theories available for the study of fisheries co-management institutions. The institutional analysis research framework has been selected for use in this project because it is both specific enough in character for organizing inquiry but general enough to be useful in a wide variety of situations. The institutional analysis research framework has been used for analysis of diverse situations and no problems have yet been encountered. It is hoped that its use in this project will lead to further methodological strengthening of the research framework.

24 BIBLIOGRAPHY

Acheson, J. M., 1975. The lobster fiefs: economic and ecological effects of territoriality in the Maine lobster industry. Human Ecology, 3, pp. 183-207.

Berkes, F., 1989. Co-Management and the James Bay Agreement. In Co-Operative Management of Local Fisheries: New Directions for Improved Management and Community Development, edited by Evelyn Pinkerton, pp. 189-208.

Berkes, F., D. Feeny, B.J. McCay and J.M. Acheson, 1989. The benefits of the commons. Nature Vol.340, No.6229.

Berkes, F., 1994. Co-Management: Bridging the Two Solitudes. Northern Perspective,Vol. 22, No. 2-3.

Bromley, D.W. 1991. Environment and Economy: Property Rights and Public Policy. Blackwell: Cambridge, Massachusetts.

Bromley, D.W. 1991. The Commons, Property and Common-property Regimes. In D.W. Bromley (ed) Making the Commons Work. Institute of Contemporary Studies Press: San Francisco.

Commons, J.R. 1968. Legal Foundations of Capitalism. Madison. Wisconsin: University of Wisconsin Press.

Feeny, D., F. Berkes B.J. McCay and J. M. Acheson, 1990. The Tragedy of the Commons: Twenty-Two Years Later. Human Ecology, Vol. 18, No. I

Feeny, D. 1992. Where do we go from here? Implications for the Research Agenda. In D.W. Bromley (ed) Making the Commons Work. Institute of Contemporary Studies Press: San Francisco.

Gardner, R. and E. Ostrom. 1991. Rules and Games. Public Choice. 70(2):121-149.

Hanna, S. A. 1990. The Eighteenth Century English Commons: A Model for Ocean Management. Ocean and Shoreline Management 14, pp. 155-172. Elsevier Science Publishers Ltd. UK.

Hanna, S.A, 1994. Co-management. In Gimbel, K.L., Limiting Access to Marine Fisheries: Keeping the Focus on Conservation. Centre for Marine Conservation and World Wildlife Fund US.

Hanna, S.A. 1995. Industry Participation and Fishery Management Performance. Presented at the Fifth Conference of the International Association for the Study of Common Property, Bodo, Norway, 24-28 May 1995. To appear in Ocean and Coastal Management 1995.

Hilton, R.M. 1992. Institutional Incentives for Resource Mobilization: An Analysis of Irrigation Systems in Nepal. Journal of Theoretical Politics 4(3): 283-308.

25 Jentoft, S., 1989. Fisheries co-management. Delegating government responsibility to fishermen's organizations. Marine Policy April 1989, Butterworth & Co. (Publishers) Ltd.

Jentoft, S. and T. Kristoffersen, 1989. Fishermen's Co-management: The Case of the Lofoten Fishery. Human Organization, Vol. 48, No. 4.

Jentoft, S. and B.J. McCay, 1995. User participation in fisheries management. Lessons drawn from international experiences. Marine Policy, Vol. 19., No. 3, pp. 227-246. Butterworth- Heinemann Ltd.

Kiser, L.L. and E. Ostrom. 1982. The Three Worlds of Action: A Metatheoretical Synthesis of Institutional Approaches. In E. Ostrom (ed) Strategies of Political Inquiry. Beverly Hills, California: Sage.

Korten, D.C., 1986. Community-based resource management. In D.C. Korten (ed) Community management: Asian experiences and perspectives. Kumarian Press, West Hartford, Connecticut, USA, pp. 1-15.

McCay, B. J. and J. M. Acheson, 1987. Human ecology of the commons. In The Question of the Commons: Anthropological Contributions to Natural Resource Management, ed. B. McCay & J. Acheson. University of Arizona Press, Tucson, Arizona, pp. 1-34.

McCay, B. J., 1993. Management Regimes. Property Rights and The Performance of Natural Resource Systems. Background Paper prepared for the September 1993 Workshop, The Beijer International Institute of Ecological Economics.

North, D.C. 1990. Institutions, Institutional Change and Economic Performance. Cambridge Series of Political Economy of Institutions and Decisions. Cambridge University Press, Cambridge.

Oakerson, R.J. 1992. Analyzing the Commons: A Framework. pp. 41-59. In D.W. Bromley, (ed) Making the Commons Work: Theory, Practice and Policy. Institute for Contemporary Studies Press, San Francisco.

Ostrom, E. 1986. A Method of Institutional Analysis. In F-X Kaufmann, G. Majone and V. Ostrom (eds). Guidance, Control and Evaluation in the Public Sector. pp. 459-475. Berlin and New York: de Gruyter.

Ostrom, Elinor. 1990. Governing the Commons: the Evolution of Institutions for Collective Action. Cambridge University Press, Cambridge.

Ostrom, E. 1992. Crafting Institutions for Self-governing Irrigation Systems. Institute for Contemporary Studies Press, San Francisco.

Ostrom, E., L. Schroeder and S. Wynne. 1993. Institutional Incentives and Sustainable Development: Infrastructure Policies in Perspective. Westview Press, Boulder, Colorado.

Ostrom, V. D. Feeny and H. Puht (eds) 1993. Rethinking Institutional Analysis and Development: Issues, Alternatives and Choices. 2nd edition, Institute for Contemporary Studies Press, San Francisco.

26 Ostrom, E., R. Gardner and James Walker. 1994. Rules, Games and Common Pool Resources. University of Michigan Press, Ann Arba.

Ostrom, E. 1994. Self-organizing Resources Regimes: a Brief Report on a Decade of Policy Analysis. In The Common Property Resource Digest No. 31, October pp. 14-19.

Ostrom, E. 1994. Neither Market Nor State: Governance of Common-pool Resources in the Twenty-first Century. Lecture presented at the International Food Policy Research Institute, Washington D.C.

Pinkerton, E. (ed) 1989. Cooperative Management of Local Fisheries. Vancouver, University of British Columbia Press.

Pinkerton, E. ,1993. Local Fisheries Co-Management. A Review of International Experiences and Their Implications for Salmon Management in British Columbia. A Discussion Paper, University of British Columbia.

Polinac, R. 1988. Evaluating the Potential of Fisherman's Organizations in Developing Countries. iCMRD. University of Rhode Island, Kingston.

Porneroy, R.S. and M.J. Williams. 1994. Fisheries Co-Management and Small-Scale Fisheries: a Policy Brief. ICLARM: Manila.

Raakja;r Nielsen, J., 1994. Participation in fishery management policy making. Marine Policy, Vol. 18, No. 1.

Raakja;r Nielsen, J. and T. Vedsmand, 1995. Fisheries Co-management: An alternative strategy in fisheries - cases from Denmark (draft). An issue paper for the OECD study on the efficient management of living resources.

Ruddle, K. 1989. The Organization of Traditional Inshore Fishery Management Systems in the Pacific. Rights Based Fishing. Kluwer Academic Publishers, The Netherlands.

Runge, C.F. 1992. Common Property and Collective Action in Economic Development. In D.W. Bromley (ed) Making the Commons Work. Institute of Contemporary Studies Press: San Francisco.

Savas, E.S. 1987. Privatization: the key to better government. Chatam, W.J.:Chatam House.

Schlager, E. 1990. Model Specification and Policy Analysis: the Governance of Coastal Fisheries. Workshop in Political Theory and Policy Analysis. Indiana University, Bloomington, Indiana. Unpublished Ph.D. thesis.

Schlager, E. and E. Ostrom. 1993. Property Rights Regimes and Coastal Fisheries: an Empirical Analysis. pp. 13-41. In T.L. Anderson and R.T. Simmons (eds). The Political Economy of Customs and Culture: Informal Solutions to the Common's Problem. Rowan and Littlefield publishers, Lanham, Maryland.

27 Tang, Shui Yan. 1992. Institutions and Collective Action Self-governance in Irrigation. Institute for Contemporary Studies Press, San Francisco.

Thomson, James T. 1992. A Framework for Analyzing Institutional Incentives in Community Forestry. Community Forestry Note 10. FAO, Rome.

Townsend, R. E., 1995. Transferable dynamic stock rights. Marine Policy, 1995, Vol. 19, No. 2, pp. 153-158. Butterworth-Heinemann Ltd.

Vedsmand, T. and J. Raakjaer Nielsen, 1995. Devolved fisheries management systems: A discussion on implementation of alternative fisheries co-management models in Denmark. Paper prepared for ICES Annual Science Conference, 21-29 September 1995, Aalborg, Denmark.

Williamson, O.E. 1985. The Economic Institutions of Capitalism: Firms, Markets, Relational Contracting.

28

Institutional analysis

Definition

Institutional analysis is the identification of various resource users, stakeholders and organizations involved in community- based coastal resource management (CBCRM). It also involves an examination of the institutional arrangements, the set of rights and rules for CBCRM in a community.

An institutional analysis is usually conducted early in the CBCRM process during the planning phase. The level of detail can range from a simple description of the existing coastal resource management system to a very detailed analysis of the management system in terms of equity, efficiency and sustainability. Purpose

To identify existing legislation, policies and regulations for coastal resource management at different levels of government (village, municipal, district, province, regional, national, international) and community (customary, traditional). To identify existing property rights and tenure arrangements in order to determine who defines rights to exploit the resources, who has access to the resources, and whether any of these rights are transferable, and the identification of the rules that must be followed. To evaluate the existing level of involvement of resource user groups in managing resources in order to determine the ways in which user groups can participate in CBCRM. A CBCRM program may attempt to alter existing power structures to allow wider participation in management of resources. This is a delicate process where it is essential to be aware of the existing structures and what likely effects proposed changes will have.

To assess the mandate and structure of existing CBCRM organizations.

Requirements

Human resources government officials facilitator with expertise in political science, sociology, resource management or economics

Institutional analysis and be I-. for for a may primary primary as research and and D1 these both publications policies, methods methods questions questions and for and from questions leaders leaders laws, and guide (NGOs) Sources guide reports Community-based purpose purpose the in markers provides provides information information community community the management Use Management and the on documents, colored organizations Methods organizations. sources. pens collecting collecting trains trains resource and from analysis analysis Resource officials officials in and approach and institutions. use data paper coastal to facilitator facilitator secondary Participatory Coastal Collect about regulations non-government academic framework. institutional institutional them and notebooks The The government government brown ®j 1. Suggested Materials Guide questions

Stakeholders

Who are the resource users and stakeholders?

Organizations at the local level What village-level organizations exist in the area? Which are engaged in CBCRM? Which are formal (legally recognized) groups, and which are informal? For formal groups, to which category do they belong (I) LGUs; ()) NGOs; (3) community-based organizations; (4) private interest groups and (5) others? What are the organization's mandates or objectives and administrative structure? How long has the organization been in existence, and what is its historical development? Is the membership increasing or decreasing? What are the organization's technical, personnel and financial resources? How is the organization affiliated with other organizations vertically and horizontally? What are the characteristics of the leadership/power structure of the group? How are group decisions made (consensus/majority/autocratic)? What is the level of representation and participation of resource users and stakeholders in decision-making?

Institutional arrangements at the local level

What are the property rights in terms of access. management, exclusion and transfer? What are the formal and informal (traditional and customary) rules?

Institutional analysis Guide questions., continued

What are the operational rules that pertain to boundary, allocation, authority and equity?

What are the regulatory mechanisms (e.g., quota, closed season, etc.) and incentives (e.g., taxation, licensing, etc.)?

What are the management rules, such as adjudication and enforcement?

How is the rulemaking body formed in terms of leadership, membership and representation? What are boundaries (i.e., political, gear type, traditional/custom- ary, organizational, physical), their size/clarity, ownership, geo- graphical coverage and changes over time?

How are rules enforced and what sanctions are used? How legitimate and relevant are the rules to resource users?

Organizations above the local level Which organizations exist in the area above the village level? Which organizations are engaged in CBCRM? For the relevant organizations, what are the formal policies, programs, regulations, laws and legislation related to CBCRM?

Participatory Methods in Community-based Coastal Resource Management Guide questions.. continued

Which are formal (legally recognized) groups, and which are informal?

For formal groups, to which category do they belong (I) local government or other state-level bodies; (2) NGOs; (3) POs; (4) private interest groups; (5) national government agencies and other regional agencies; (6) bilateral/regional bodies; (7) international agencies and (8) others?

What are the organization's mandates or objectives and administra- tive structure?

At what level does the organization operate: (I) international; (2) regional; (3) national/central; (4) regional; (5) province/state; or (6) district/municipal/town?

How long has the organization been in existence, and what is its historical development?

What are the organization's technical, personnel and financial resources?

How is the organization affiliated with other organizations vertically and horizontally?

What is the organization's awareness of the conditions of the fisheries/marine resources?

Institutional arrangements above the local level

How do national policies, programs, regulations, laws and legisla- tion affect CBCRM at the local level? How do the other national policies, programs, regulations, laws and legislation on economic development and general public administra- tion affect CBCRM?

Source: Pido, et a/. / 996.

2. Complement and validate the secondary data collection by collecting primary data. A variety of participatory techniques and tools can be used. These include

Institutional analysis structured and semi-structured interviews, focus group discussion, resource mapping, historical timelines, flow patterns, case studies and venn diagrams. Again, the guide questions should serve as a framework.

3. Collect and sort the data, focus on the relationships between and among the various institutional arrangements and organizations involved in CBCRM.

4. Identify complementarities, conflicts, overlaps and gaps in the institutional arrangements and organizations which support or hinder effective CBCRM at various levels of government and within the community.

5. Identify what is needed to support CBCRM, such as new regulations, new organizations and enforcement mechanisms.

6. Recommend strategies for implementing patterns of relationships in space, time, flow and decision using various tools such as transects, maps, timelines, venn diagrams and matrix.

4 Participatory Methods in Community-based Coastal Resource Management Note

Institutional and organizational arrangements can change and should be analyzed over time.

Example

Matrix of nested organizations for coastal resource management in Binunsalan Bay, Philippines

Administrative Government Local Non-government People's level agency government organization organization unit

National Department of Environment and Natural Resources

Provincial Provincial Provincial Environment Government and Natural (Palawan) Resources Office

Municipal Community Municipal Environment Government and Natural (Puerto Resources Princesa City) Office

Village Village Council - Ligaya ng Bu/tay - Christian Multi-

(Mangingisda) - Binunsalian Bay purpose Cooperative Foundation, Inc. - Charity Women's Association - Fisherman's Ministerial Fellowship

7. Analyze rules.

There are three levels of rules which are very closely linked and need to be clearly identified during the analysis.

Operational rules govern and regulate the day-to-day decisions and operations of the resource user con- cerning when, where and how to harvest the resource.

Institutional analysis Management rules are used by resource users and government to formulate and change operational rules, deal with conflicts, enforce decisions, and detect and sanction against rule violation. Legal and policy rules establish the process for re- source management. Legal and policy rules include, for example, the national fisheries policy and legisla- tion which establishes a national fisheries agency. National laws and policies are translated into man- agement rules, which in turn are further translated into operational guidelines. In other words, the rules affecting operation are made within a set of manage- ment rules that are themselves made within a set of legal and policy rules.

8. Validate. It is important for the community members and government officials to participate in validation of the institutional analysis to ensure accuracy and to fill in any data gaps. This can be done through meetings with key leaders, focus group discussions, and/or with the community as a whole.

Participatory Methods in Community-based Coastal Resource Management 9. Produce the final report. Following is the suggested outline.

Background/Rationale for Institutional Analysis

Objectives of Institutional Analysis

External Institutional and Organizational Arrangements (national. provincial, district, municipal) - policy, legislation, regulation, programs - government administrative agencies (mandate, functions, structure, objectives) - non-government organizations (mandate, functions, structure, objectives) - services - nested relationships

Community Institutional and Organizational Arrangements - stakeholders - community organizations (mandate, functions, membership, structure, resources) - boundaries (political, physical/natural, gear, customary, fishing spot) - property rights/tenure arrangements - rules: formal/informal; operational, collective choice, constitutional - decision making and conflict resolution - monitoring and enforcement - nested relationships

Analysis and Diagnosis - Institutional analysis of the coastal resource - Management systems - Pattern analysis (Space, time, flow, decision)

Summary and Recommendations

Institutional analysis Output

Ilr A report containing descriptions, maps and figures that analyzes the formal and informal coastal resource management systems that operate in the community. The outputs are of use to resource users, stakeholders and government for dialogue and debate about coastal resource management and CBCRM.

Venn diagram showing nested arrangements for different levels of fishing rules in the Philippines

Legal and policy rules

National fisheries PD 704 Code of 1975 Mana ement rules

Municipal marine reser and sanctuary rules eratwn (for V Only handlines maybe used in the reserve ora rules

rp,

Community level

No fishing in reserye during fishbreeding season of January to March

No fishing activities allowed in sanctuary

Local government code of 1991

Participatory Methods in Community-based Coastal Resource Management Strengths

Depending upon the level of detail generated, the' institutional analysis provides information which may not otherwise be available or commonly understood by the community and government. Can be conducted in a relatively short time period (one or two weeks) and at a low cost.

Provides an opportunity for resource users to share their knowledge and understanding about resource management and use.

Allows for the synthesis of bio-physical, socio-economic and institutional information about coastal resource use and management.

Limitations

Institutional analysis can be relatively complex. Due to this complexity, it requires a trained facilitator to undertake the analysis.

The analysis requires an understanding of informal rights and rules at the community level. The team members must be able to probe deeply to get this type of information in an accurate manner.

Institutional analysis in kLA Example

In San Miguel Bay, Philippines, an institutional analysis identified coastal resource use patterns and both formal and informal manage- ment systems. Weaknesses were identified in the formal manage- ment system of the government because the various municipalities surrounding the Bay were implementing fisheries management regulations in different ways and there was no coordination of management. This resulted in use conflicts and overexploitation of the fishery. Recommendations were made to develop a Bay-wide management council to coordinate management systems and improve overall enforcement. The San Miguel Bay Management Council was established. Fisher organizations were directly involved in the management council providing for a participatory and "bottom-up" management system.

References

ICLARM/IFM. 1996. Analysis of Fisheries Co-management Arrangements: A Research Framework. Working Paper

No. 1. Fisheries Co-management Project. International Center for Living Aquatic Resource Management, Manila, Philippines.

Pido, M.D., R.S. Pomeroy, M.B. Carlos and L.R. Garces. 1996. A Handbook for Rapid Appraisal of Fisheries Management Systems (Version 1). ICLARM Educ. Ser. 16. Manila, Philippines.

Prepared by Robert Stephen Pomeroy

Participatory Methods in Community-based Coastal Resource Management Socioeconomic and Gender Analysis Programme SEAGA Field Handbook

Compiled by Vicki L. Wilde

Available at http://www.fao.org/sd/seaga/SEfh0001.htm

Chapter Six: The Development Context Toolkit

Development context tool 5: Venn diagrams Development context tool 6: Institutional profiles

All SEAGA materials are available from: Women in Development Service, Women and Population Division, Sustainable Development Department, Food and Agriculture Organisation, Viale delle Terme di Caracalla, 00100 Rome, Italy. Phone: 39-6-52255102, Fax: 39-6-52252004, E-mail: [email protected] Development Context Tool 5: Venn Diagrams

PURPOSE

The Venn Diagram is a tool that helps us to learn about the importance of local groups and institutions. This can be useful for clarifying decision-making roles and identifying potential conflicts between different socio-economic groups. It is also helpful for identifying linkages between local institutions and those at the intermediate- and macro-levels.

PROCESS

Organise separate focus groups of women and men, including a mix of socio- economic groups. Be sure that the poorest and most disadvantaged (by ethnicity or caste etc.) are included, or have their own groups, as appropriate.

The Venn Diagram can be traced on the ground, but it is especially clear (and fun!) if coloured sticky paper circles are used on a large sheet of flip chart paper. It is helpful to cut out circles in different sizes and colours ahead of time.

Start by asking the participants to list the local groups and organisations, as well as outside institutions, that are most important to them. Then, ask the participants to decide whether each organisation deserves a small, medium or large circle (to represent its relative importance). The name (or symbol) of each organisation should be indicated on each circle. (Make sure each organisation has a different colour, if possible.)

Ask which institutions work together or have overlapping memberships. The circles should be placed as follows:

X separate circles = no contact

X touching circles = information passes between institutions x small overlap = some co-operation in decision making

X large overlap = a lot of co-operation in decision making Discuss as many institutions as possible and ask the participants to position them in relation to each other. There may be a lot of debate and repositioning of the circles until consensus is reached.

81 Vicki L. Wilde March '98 MATERIALS

Flip chart paper, markers, sticky paper (in several colours) and scissors.

NOTES TO THE RA TEAM

In general it is important to understand in what ways the different participants are satisfied or dissatisfied with the groups or institutions available to them. It is also important to understand if certain kinds of people, e.g. women, poor or a certain ethnic group, are excluded from participation in certain institutions. Use the SEAGA Questions to deepen the discussions.

Finally, be sure to discuss and compare the Venn Diagrams produced by the different groups of participants. If one group has given a certain institution a large circle and another has given it a small circle, find out why. How is that institution relating differently to different members of the village? Note also whether one group has included fewer organisations in its diagram.

Some SEAGA Questions to Ask While Facilitating

Development Context Tool 5: Venn Diagrams

Are there local groups organised around environmental issues? e.g. forest users group, water users group.

Are there local groups organised around economic issues? e.g. credit, agriculture production.

Are there local groups organised around social issues? e.g. health, literacy, religion.

Are there groups from which women are excluded? Which ones? Why? What do they lose due to their lack of participation?

Are there groups exclusively for women? If so, what is the focus of these groups? What do women gain from them?

Are the poor excluded from any of the local groups? Which ones? Why? What do they lose due to their lack of participation?

What are the links between local groups or organisations and outside institutions? e.g. NGOs, political parties, government institutions.

Vicki L. Wilde March '98 82 EXAMPLE

Development Context Tool 5 shows the importance of gathering information from both women and men about institutions. Focus group discussions in El Zapote, Honduras showed that women and men ranked the relevance of community groups for local welfare very differently. This example focused on gender differences. Further differentiation is possible across socio-economic lines.

Development Context Tool 5 Venn Diagrams Example: Men's and Women's Perceptions of Institutions in El Zapote, Honduras

is Health ;t;natioro

School 1agtartD

Mothers' School Lvn:h Parents Club Committee Association Church Wh Crvb

Mrsn's Perceptions

Source: Urban and Rojas, (1993) Shifting Boundaries: Gender, Migration and Community Resources in the Foothills of Choluteca, Honduras, An ECOGEN Case Study, Worcester, MA: Clark University

Vicki L. Wilde March '98 83 Development Context Tool 6: Institutional Profiles

PURPOSE

Institutional Profiles are tools that help us to learn more about the nature of the institutions identified in the Venn Diagrams. An analytical chart is created for each institution in the community to examine what they have accomplished and what they need to foster development work. If local communities are to implement development activities which they can sustain, explicit attention must be paid to their capacities.

PROCESS

Work with the same groups of participants that produced the Venn Diagrams. For each local group or institution identified in the Venn Diagrams, discuss at least four kinds of information: founding and goals, management, achievements and needs. Pre-prepare a chart on flip chart paper for each institution (see example). Use the SEAGA Questions to deepen the discussion. Be sure to ask about leadership, membership, activities, decision-making processes, and interactions or conflicts with other groups or institutions, including those from the intermediate- and macro-levels.

MATERIALS

Venn Diagrams, flip chart paper and markers.

NOTES TO THE RA TEAM

Whereas the Venn Diagrams reveal the importance of local institutions and the degree of interaction between them, the Institutional Profiles show details about how these institutions function and for what purposes. Together these tools facilitate learning about the local institutional context. This information will be very important when the community is planning development activities.

Vicki L. Wilde March '98 84 Some SEAGA Questions to Ask While Facilitating

Development Context Tool 6: Institutional Profiles

How many local groups or institutions are there? Who participates in them? e.g. elders, women, farmers. What are their purposes?

Are leadership positions dominated by a particular social group, (e.g. high caste wealthy elderly men)?

Do women occupy leadership positions in any of the local institutions? If so, which women?

Which institutions have achievements related to meeting community development

needs?

Which local institutions have links with outside institutions? For what purposes?

EXAMPLE

In the case of Jeded Village in Somalia there were only three community institutions elders, women and youth. Institutional Profiles for each are shown in Figure A6. Discussions revealed that Jeded's Council of Elders was by far the most important and that it worked closely with the other groups in the village. Later, when designing their Community Action Plan it was decided that the Council of Elders would retain a leading role, but a Steering Committee also would be formed consisting of three members each, elders, women and youth. Further, responsibility for certain development priorities was assigned to the different groups: education and soil erosion to the Youth Committee, human health and income generation to the Women's Organisations, and animal health and jobs to the Council of Elders. Water issues would be overseen by both the elders and women.

Vicki L. Wilde March '98 85 Development Context Tool 6 Institutional Profiles Example: Institutional Profiles of Jeded Village, Somalia

Council of Elders Group Foundation and Management Achievements Needs Goals Council of Founded in 1954 The Council elects ..Maintaining ..Office equipment Elders a Chair for a peace in the and stationary Goals:..to solve flexible term; village community Criteria for ..Training problems membership on the ..Borehold water council include age, management ..Transport ..to manage water wisdom, and and other significant ..Sanitation ..Petty Cash community affairs experience ..Education ..to develop a water schedule for nomads

to adjudicate disputes

Women's Organization Group Foundation and Management Achievements Need Goals Women's Founded in 1991 Chairwoman Helped to Training Organization elected in a resettle families Goals:..solve Congress of coming from the Space and Women's problems women of Jeded Civil War in the equipment South ..advocate rights of Annual elections for Income generating women and Chair and other Sanitation activities children leaders activities Created income ..participate in Any women 20 generating implementation of years or older may projects such as development be a member weaving mats projects Membership fee is Fund raising for ..solve problems 1000 Somali business among themselves Shillings activities

..serve as link Meets once a between women of month Jeded and aid organizations Links with women's groups in other ..initiate income villages generating projects

..care for displaced families

Vicki L. Wilde March '98 86 Example: Institutional Profiles of Jeded Village, Somalia

Youth Organization Group Foundation and Management Achievements Needs Goals Youth Founded on 19 Chair Security Training Organization June 1991 School Meeting room and Objectives include construction office maintaining Vice Chair security; solving Volunteer Stationery and routine problems; teachers operational tools encouraging community Central Committee Help with health Communication and cooperation and post transport income generation Resettlement of Sub-Committees; displaced people

Founded by Security Planting and educated young conservation of people in the village Justice Relief and soil Rehabilitation Road Sports, Health, maintenance Education Water Livestock Repairing borehole Displaced Persons Sports Agriculture

Range and Forestry

Chair selected by the Central Committee according to ability and knowledge; elections held every four years

Source: Ford, Adam, Abubaker, Farad and Barre, (1994) PRA) with Somali Pastoralists: Building Community Institutions for Africa's Twenty-first Century, Clark University/GTZ/Gardo

Vicki L. Wilde March '98 87 + '1L Yvrid r; &° i Oil d 5-4. c'nac IN-right CHAPTER 4 Transforming Customary Law and Coastal Management Practices in the Maluku Islands, Indonesia, 1870-1992 Charles Zerner

Leave the pinang alone! Leave the coconuts alone! Leave the bananas alone! Leave the trochus there! From the mountains to the coast, Leave it all alone! Ritual chant, Saparua Island

In 1989, I sat in the office of the head of Nolloth village on Saparua Island, one of the Lease Islands of the Central Maluku of eastern Indonesia. There Mr. Metekohy, one of two chief kewangs of the village, chanted a portion of the an- nouncement he recites each year to communicate that the period of prohibitions on entry, harvest, or hunting in community-controlled areas, known throughout the islands as sasi, is in effect. In his role as kewang, Metekohy combines the responsibilities of a resource monitor with the power to enforce community resource-management rules. While walking the paths on the periphery of the community-management territory, or petuanan, he announces the temporary closing of the area. "As I circumambulate the outermost limits of each community-management area," Metekohy explained, "I stop wherever paths intersect to repeat my ritual chanting." His chants are ac- companied by the lowing, plaintive sounds of an assistant blowing on a conch shell trumpet. Among many agricultural and fishing communities throughout the Maluku Is- lands, sasi is an historic family of institutions and practices that has been used to regulate access to particular resources and territories under a variety of property regimes. Although a systematic study of sasi over a broad geographic range has never been undertaken, it is clear that these practices have been and, in many

80 4ALUKU ISLANDS 81 areas throughout the 1,029 or more Maluku islands (see Map 4.1), continue to be followed. Time and space are among the most important media through which many Moluccan communities regulate access to resources and territories that are im- portant from subsistence and commercial perspectives. When an area is ditutup (closed), entry prohibitions are in effect, and the area under sasi is marked by signs placed at strategic locations. Temporary signing of individual trees, groves, or entire territories is accomplished with striking ritual constructions of paired co- conut palm fronds jutting outward from a central sugarcane trunk crowned by a young coconut; sound-producing "instruments" at the apex of the assemblage make percussive sounds when struck by the wind. This is one of the fundamental ways in which the boundaries of resource areas under sasi are marked. In the case of Nolloth, only the chief kewang and his assistants, a group of ten anak-anak kewang charged with enforcing the closure of the area and implementing the pro- hibitions within the community-management territory, are permitted to enter freely during the closed period; a host of other prohibitions on gear, behavior, and identity of entrants also is put into play. When sasi is opened, entry prohibitions and other associated restrictions are lifted; the kewang circumambulates the terri- tory formerly under sasi, unties the sasi signs or bindings, and announces that the prohibitions governing the territory have been lifted. Although the practice of demarcating "cultural regions" is problematic because it suggests historically fixed and homogenous regions, it may be useful to consider the Maluku Islands as the westernmost edge of a cultural and institutional Wal- lace's line. Beginning with the Maluku, and extending east through New Guinea and the islands of the Southwest Pacific, marine tenure institutions, rules, and practices have played an important role in regulating access to reefs and inshore marine waters. To the west of the Maluku, beginning with Indonesia's orchid- shaped island, Sulawesi (formerly Celebes), open-access, freedom-of-the-seas ideologies and practices are mixed with a variety of marine tenure practices that only now are beginning to be understood and documented (see Bailey and Zerner 1992; Zerner 1992, 1994a, 1994b).

The Development of Sasi Institutions and Practices Historically, and in many contemporary Moluccan communities as well, sasi prac- tices are not just a body of changing rules but also a diverse set of institutional roles that have varying degrees of relationship to central governmental and reli- gious authority. Sasi role holders, including kewangs and mauwengs as well as other practitioners, combine religious, civil, martial, and commercial functions. Attempts to characterize sasi ahistorically, as though these practices, rules, and in- stitutional roles had a fixed essence, would be inadequate and inaccurate. Zerner (1994b) and von Benda-Beckmann, von Benda-Beckmann, and Brouwer Map 4.1

s Major Maluku Islands

Map 4.2 9 x

Scram Strait Nolloth 4j7

Ihamahu UA J_

a nipungniihu 1 ',--_ 1 \ T u h a h a ice'

'\\ a parua

Slri Sori Serani aperu SiriSori Islam Saparua \ Day Ulat

it Saparua Island

82 MALUKU ISLANDS 83 (1992) speculate that sari practices initially may have been directed toward an- cestral spirits believed to inhabit Moluccan marine and terrestrial landscapes. Such spirits were said to control the luck of the hunt, the fertility of crops, and in- dividual fate. Those who violated areas marked by sasi signs risked sickness or even death at the hands of these spirits, in addition to a variety of civil punish- ments and fines implemented by the kewang. During the late nineteenth century and the first three decades of the twentieth, Zerner and von Benda-Beckmann, von Benda-Beckmann, and Brouwer suggest, sasi practices were recrafted and deployed by local elites, assisted by Dutch colo- nial administrators who sought access to sources of income realized by the con- trol of the sale of valuable agricultural and, to lesser extent, marine commodities. Sixty-nine years before Mr. Metekohy chanted a few sasi verses for me, a vil- lager recalled the anxious atmosphere surrounding ritual closure of the coconut groves around Siri Sori Serani village in 1920:

In the village very ripe coconuts were in the trees, ready to be plucked. After consultation with the chief kewang, the village council decided to impose the sasi.... Following this, a sasi proclamation was made. One evening in the dark of the moon, the chief kewang, attended by his as- sistants, walked around the village.... [T]he kewang called out loudly ... [and] the assistant kewang blared his great shells to produce a long and somber sound. The chief kewang chanted a long-lasting "silo ... o ... o!" [a word that means sasi]. Whereupon his assistant answered "mese ... o ... o! [is laid down]". As they went around the village, I heard the sounds. I was but a child sitting in my house, and I began to tremble in my seat. (Volker 1925:298-299)

As a young boy, fears of spirits, as well as fears of being apprehended by the ke- wangs on their nightly patrols, prevented the narrator from entering the grove of coconuts on which the sasi prohibition had been laid:

From this evening the sasi was laid down and the work of the kewang began: they saw to the implementation of sasi. For no one can escape entrapment and punishment, if he plucked something out of a garden. Personally, it is enough to come up against a kewang on a dark evening, far away from the vicinity of your own house. In order to quell suspi- cion, men will reverse and criss-cross their paths to avoid the bush police; for if a man cannot give a clear and polished explanation for being out so late beyond his house, then he must bear the punishment anyway. (Volker 1925:299)

The teller of this tale cringed at the prospect of violating the coconut grove sasi be- cause he was terrified of the potential response of watchful spirits. These impal- pable witnesses might inflict a catastrophe upon him or the community: 84 CHAPTER 4 I still remember this very well, that we boys in the time of sasi played in the groves, and an old dry coconut fell down [we would not pick it up]. [Normally] we wouldn't think twice about grabbing it and whisking it away. This was because we felt a sort of holy awe, and we knew we would cause a great flood if we took the coconut. We had a secret terror of the ghosts. (Volker 1925:299)

Sasi and the Regulation of Access to Agricultural Commodities The effectiveness of sasi practices has waxed and waned throughout its recorded history, depending, in part, on the source(s) of authority and the political and eco- nomic interests these authorities and rules represented. Whether, and in what cir- cumstances, local villagers were interested in or willing to submit to sasi regimes is not clear. Religious sanctions and practitioners, Dutch colonial authority, and local Moluccan elites dependent upon the colonial enterprise for legitimization of their power and income were among the varying sources of authority behind im- position of sasi in different times and places. Recently, as we shall see, Indonesian village-level government officials as well as environmental nongovernmental or- ganizations have begun to restructure sasi rules, practices, and ideologies in at- tempts to reconstruct sasi in ways that further their own (very different) interests. In certain areas of the Maluku Islands, sasi developed as an instrument for fa- cilitating commercial regulation of access to valuable agricultural crops, especially coconuts, betel and kemiri nuts, and possibly cloves. Indeed, the pivotal historical role of the Central and North Maluku Islands (including Ambon, Saparua, Seram, Banda, Buru, Haruku, Ai, Hatta, and Ran, as well as Halmahera, Ternate and Tidore) as supply centers for the world spice trade suggests the strong possibility that, in certain areas, these practices may have been stimulated by markets for agricultural commodities. Prior to the arrival of Europeans in the sixteenth century, traders from Makassar, East Java, the Philippines, and Banten were sailing to Moluccan ports to seek agricultural commodities (Andaya 1993). Like their European successors, these traders needed to establish and secure access to prized crops grown in remote village groves. During the sixteenth century, Portuguese, British, and Dutch traders also arrived, eager to establish monopolies on the Moluccan spice trade for European markets. The development of institutional control mecha- nisms, whether superimposed on already existing indigenous practices or in- vented by local native elites, traders, and, later, Dutch government officials, may have been driven by commercial opportunities. If this speculation is correct, then the codification of sasi customary law in Dutch and Malay-Ambonese texts be- comes more easily understandable. These texts may have constituted attempts to control, codify, and commercially rationalize sasi customary law, practices, and in- stitutions. At least one account from the island of Ambon in 1920 reveals a sasi institu- tional structure in which the proceeds of harvest auctions were used to pay for MALUKU ISLANDS 85 surveillance of agricultural territories under sasi and to provide salaries for re- source monitors. This particular kind of sasi elaborated the arrangements for timing and manner of harvest and the rules for gathering, marking, and auctioning individual and clan-controlled crops that had been gathered into central ware- houses. Prices for the yearly crops were negotiated and allocations made to dis- tribute profits to individual and clan landowners. As in many villages practicing sasi today, the Siri Sori Serani rules of 1920 stipulated that the market price for the commodities sold from one sasi were negotiated at a meeting of local government officials, sasi and clan leaders, and elders. Apparently, portions of each year's profits also were distributed to village government and religious institutions. Money obtained from the sale of certain sasi harvests occasionally was donated to the mosque or used to buy oil for the mesjid (Volker 1925). The elite leaders of Siri Sori Serani village elaborated one such arrangement during the early twentieth century:

During the imposition of sasi, the rightful landlord is permitted to enter his own garden in order to clean up the grounds and to gather already ripened and fallen coconuts. He must take these to the village meeting place, where they are branded and stamped with the sign of his clan group and stockpiled. So I saw at the conclusion of the sasi-period almost the entire space under the village meeting place filled with piles of coconuts. (Volker 1925:299-300)

The imposition of sasi in this particular village seems to have been initiated by a rural elite working in collaboration with a Dutch colonial official. A letter signed by leaders of this village to a local regent implores him to issue a "strong promise" reinforcing and further codifying sasi rules because local villagers apparently did not want to obey them. According to the letter,

Because it is clear that we villagers pay little attention to sasi regulations which are usually implemented by us according to custom ... appar- ently villagers do not yet understand that sasi regulations implemented under each rule will improve the lives of the inhabitants themselves, we request that this matter will hopefully be taken care of by Sir [the con- trolleur] through issuing new orders using a strong promise to confirm that this sasi is for the good of the inhabitants. (Volker 1925:301)

In response to this collective request from local leaders and sasi enforcers for as- sistance, the regent issued a letter intended to strengthen villager compliance by regularizing and codifying sasi procedures, purposes, and penalties. His "strong promise" and new rules constituted an attempt to enforce sasi, to "fix this sasi for the lifetime of the inhabitants" (Volker 1925:307). In the past, as in the present, sasi presented problems of institutional design, compliance, and articulation with markets. In his letter, the regent stipulates that "in the event someone wants to buy or to contract the coconut sasi one year in advance" (Volker 1925:309), all the 86 CHAPTER 4 district officials, "old ones of the land" (Volker 1925:309), and young inhabitants are ordered to gather and meet on that issue. At one such meeting, the regent asked,

Whoever wants to enter all his fertile coconut trees into the prohibition, his fruits will be channeled within the sasi, his fruits will never be stolen because the kewang officers have purchased them, [and] the prohibi- tion will be guarded every day until the contract is finished. (Volker 1925:310)

The lands of farmers who declared their coconut groves within that year's sasi were monitored, patrolled, and protected by the kewang, aptly described by at least one Dutch official as the "bush police." If the kewang discovered any villager other than the rightful owners within the protected groves, they were arrested, tried, shamed publicly, and fined. Fines levied on outsiders, whether nonlocal traders or villagers, were higher than those levied on local villagers. It is not at all clear that the common farmers and fisherfolk of local villages sup- ported these sasi regulations. In Siri Sori Serani village, for example, not all co- conut growers wished to participate in this scheme of community-enforced pro- duction and market access. The colonial official's response to the letter sent by local leaders suggests that participation in market-oriented sasi in that one village may have been coerced:

Whoever does not want to follow these "sasie" [sic] rules and does not want to sign his name to this letter, then that person will be viewed as a person who does not want to be governed and order[ed] and wants to use his own desires in a free manner only. (Volker 1925:307-308)

The proceeds from agricultural lands entered within a particular yearly sasi were distributed according to various schemes. According to one scheme, the total amount of coconuts from an individual farmer's or clan's land was tallied and the coconuts marked. The proportional amount of profit due to individual farmers or clans, obtained from the proceeds of a collective auction of all crops sold to a par- ticular trader in that year, was returned to them. In alternative arrangements, a certain percentage of profits from a particular sasi was allocated for the public good (for example, the purchase of lamp oil and a donation of cash for the Siri Sori Serani mosque). Administrative costs for imposition and implementation of the sasi were provided by allocating a share of each farmer's coconut crop to the kewang, who thus had a direct incentive to guard the ripening coconuts under sasi, prevent theft, and convey the harvest to the village hall without loss or damage. In Sid Sori Serani village of 1920, for example, the chief kewang "receives one- tenth of all gathered fruits" (Volker 1925:310). Whether designed, invented, or fused with previous religious practices, agricul- tural sasi, as practiced in the late nineteenth and early twentieth centuries in the Central Maluku Islands, apparently provided an institutional and legal framework MALUKU ISLANDS 87 in which the security of agricultural crops could be guaranteed for individual groves or collectively held areas. Ripening groves of coconuts, betel, and candle- wood or kemiri nuts were protected from theft through the watchful eyes and mobile patrols of kewangs. Moreover, sasi institutions provided political and eco- nomic arrangements whereby individuals could coordinate their decisions-for example, to enter or not to enter into a particular sasi-through collective institu- tions. Although the equitable consequences of these relationships are not known, village-based sasi linked individual production to a larger market manifested by the yearly arrival of traders. Sasi arrangements allowed a forum for collective de- cisions on allocation of benefits to sasi officials, to individual cultivators, and for the collective good. T. Volker, a Dutch colonial legal observer, commented on the institutional ben- efits of sasi in 1921. His remarks anticipate the arguments and even the rhetor- ical strategies of contemporary common-property theorists and community- management enthusiasts who extol the virtues of direct economic incentives to local communities. Explaining the purported benefits of sasi as a collective insti- tution, Volker waxes enthusiastic:

The young fruits remain on the trees and are not plucked before their prime, for whatever purposes. Because the kewangs have a stake in the outcome, the production is increased. They see to it that the grove renters keep their groves clean, and guard against harmful wild animals such as boars, badjings, and crawfish. It is more advantageous that a significant amount of fruit should be harvested in one fell swoop and taken en masse to be processed as copra. This [collective harvest and processing] insures that it will be traded in good quantities, and so that the proprietor can conduct a lively business with honest neighbors. This is better than selling the coconuts in increments of four and five fruits, or drying them and selling them to a Chinese for a few cents that are quickly spent. Thus, as a result of sasi, an entirely different situation comes to pass. The right of sasi is still an undervalued source of educa- tion, for the little peoples are, in general, still self-willed and pig-headed. Sasi brings with it a discipline that has an important educative element. (Volker 1925:300-301)

The picture of sasi that emerges from Volker's own account, as well as from the letter from an indigenous elite linked to and dependent upon the colonial enter- prise, suggests that sasi, as practiced in the Central Maluku during the early twen- tieth century, was a set of institutional arrangements and rules based on state- sponsored policing of productive areas. These arrangements implied control over harvesting, market negotiations for sale of the product, and allocational decisions. The implication that local elites, with the benign or possibly smug connivance of the Dutch regent, were imposing "discipline" on "self-willed, pig-headed little peo- ples" and manipulating local practices in the service of particular economic 88 CHAPTER 4 interests is also apparent. By the early twentieth century, and probably for decades before written records were produced, sasi was a changing, hybrid institution that arose in a border zone of intercultural contact, crosscutting interests, and com- peting claims on resources.

Sasi on the Seas: A Marine Common-Property Institution in Flux

Sasi institutions and rules also regulated access to nearby marine environments in local coves, bays, and estuaries. Historical materials suggest that until the 1960s marine sasi regulations were far less differentiated and well developed than their agricultural counterparts. In the Central Maluku, markets for marine commodities apparently did not gain significance before the late 1950s or later. Moluccan coastal communities historically possessed well-defined marine ter- ritories under the control of particular villages. In the Central Maluku, these areas are known as petuanan taut, from the Malay tuan, or "owner," and taut, which means "sea." Lateral boundaries frequently are associated with natural features in the landscape, such as promontories or points. In Teluk Kau on Halmahera, ac- cording to Shalan, a young student who grew up in the area, an imaginary line be- tween two promontories located on opposite sides of a cove 15 km deep and 40 wide constituted the boundary of a single village's petuanan (Shalan 1991). While promontories often are regarded as the lateral boundaries of petuanan, the outermost edge of community-controlled waters is usually located at the junc- ture of coral reef and drop-off. The inshore waters under community control, from the low-tide line to the farthest edge of the reef (sometimes called "white water"), are known as meti in the Central Maluku. This area is scrupulously monitored during the imposition of sasi by the kewang, his assistants, and local villagers. Sev- eral island communities also claim and control rights over submerged atolls and underwater reefs, known locally as negeri tengelam (literally "submerged commu- nities"), which may be several miles from the island on which the community is located. During the performance of sasi closure rituals, the boundaries of a marine pet- uanan are made visible and communicated. At the inception of the closing cere- monies, ritual practitioners carry incense and offerings of food to sacred places. These places may include the tips of promontories and points along the shoreline, as well as special places in the mountains believed to be inhabited by ancestral spirits. During the imposition of sasi on Ai Island in the Central Maluku, semiper- manent ritual "trees" (lengths of wood crowned with a fringe of banana leaves and topped with a young coconut) are anchored offshore. With their fringe of banana leaves and coconut crown visible on the surface of the sea, these constructions function as a ritual sign of the sasi prohibition and a boundary marker signifying the outermost edge of the marine petuanan. Marine sasi has differed from land-based sasi in several ways. Unlike the variety of property regimes and rights-holding bodies that seem to have characterized MALUKU ISLANDS 89 agricultural sasi, local communities, rather than individuals, most commonly pos- sessed ownership or control rights. In some coves, however, individuals were able to acquire rights to operate lift-net platforms or other gear in particular places. Also unlike agricultural sasi, which was deployed in the Central Maluku to regulate access to productive land and forest areas for subsistence and for commercial pur- poses, marine sasi, until very recently, only regulated access to the petuanan for subsistence purposes. With certain exceptions, marine petuanans were closed only during the arrival of schools of pelagic fish such as tuna and the thousands of lomba fish (Thrisina baelama) that yearly returned to the mouths of certain rivers for days at a time:

The sea sasi and on the reef drop-off is implemented during that time of year when many fish migrate to the bays or into river mouths. If a fish- erman discovers that the influx is beginning, he warns the village council as quickly as possible.... The kewang ... sets a branch as a sasi symbol on the edge of the reef drop-off just in front of the place where the fish swim in.... After a couple of days, if there are enough fish in sufficient quantities in the signed places, the entire population ventures out in boats with nets, ropes, and driftnets. They hang out large sheets to drive the fish into the nets. The men break up into communal fishing parties. (Volker 1925:302)

Prior to the 1950s or 1960s, marine sasi regulations focused on increasing hunting success and coordinating collective drives of fish that were temporary, possibly migratory residents of local bays and river mouths. When a school of fish entered a marine petuanan, the ocean kewang dived into the water and observed the movements of the school. Until the school settled down, he declared sasi closed, and no one, whether a local or an outside villager, was permitted to fish or enter the area. In Porto village, for example, only after this happened was sasi opened and a collective fish drive begun. In this regulated free-for-all, villagers' access rights were equal, and each fisher's share depended only on personal skill, luck, and position in the hunt. The few written rules concerning early marine sasi practices all relate to timing, equipment, and coordination of fish drives. They focus on maximizing the catch and minimizing the possibility of startling the fish and scattering them through dis- ruptive behaviors (such as standing up in a boat) or the use of improper gear. As long ago as 1870, rules concerning marine sasi in Porto village stated that [w]hen schooling fish have just entered our harbor, no one may toss his net while standing in the water, and from time to time [it is forbidden] to use a redi, because these actions will panic the fish, they will not stay close to land and be accessible to people [trying to] net them. Whoever violates this rule, the person that stands up in his boat, he will be fined three rupiah and the [user of] the redi will be fined ten rupiah that will be placed in the village treasury. (Aponno 1977:20) 90 CHAPTER 4 Gear restrictions continue to play an important role in the Central Maluku today. In the spring of 1991, Shalan, a high school student, recalled that only certain kinds of scoop nets were permitted during fish drives within a bay along the jagged coast of Kau Bay on Halmahera Island:

Everyone within the village is entitled to fish for any kind of fish within our bay. It is only when school fish enter the bay that strict prohibitions on timing of the drive, gear and behavior go into effect. If schooling fish are sighted by a local, they are reported to the saniri negeri [a kewang- like resource monitor] who goes out to observe them. If the saniri opens up sasi, then our fish drive begins. About forty boats from our village, each with a crew of four, paddle toward the school and surround them. They may not use motors and they may not use any other gear than scoop nets. As they surround the school, the fish are driven wild and into a panic. They try to escape from our circle of boats and in the process, they jump right into our scoop nets. (Shalan 1991)

In the 1990s, on Saparua Island (see Map 4.2, page 82) as schooling fish swim to sea from an estuary they had entered, the kewangs of Porto village signal the re- laxation of gear restrictions and the opening of sasi. In 1870, the marine kewang of Porto village (known as the laukewanno) also en- forced regulations concerning the use of large nets that could close off an estuary or an entire cove. Fees levied on users of large gear provided the kewang with a salary and an incentive to perform his duties assiduously. In return, the kewang guarded the nets, prevented theft, and reduced conflict:

Whoever uses a closed net (djarin tutop), when he wants to close [off] the cove in our land, must pay fifty cents each time to the community treasury so that the kewang can guard against other people entering the net and taking fish, at their whim, which causes conflict and curses with dirty words with the result that people make the Rajapatti dizzy and there is no tranquillity within the community. (Aponno 1977:20-22)

People who were foolish or avaricious enough to enter the area near,these big nets during the temporarily closed-off time were fined: "Whoever dares to enter within [the area of the] closed net mentioned in article 86 without permission from the owner will be fined twenty five cents [that is deposited in] the community trea- sury" (Aponno 1977:20-21). Other rules from 1870 Porto prohibit the use of fish poison and prescribe sanctions for violators.

A Shift in Punishment Standards: From Shame to Money In the 1920s, if a man stole fruit from his own grove before sasi was opened, he might have been fined a block of white cloth, a bottle of palm wine, a gong, or a quantity of gold. If he stole from another person's garden, however, he was not only fined but shamed: MALUKU ISLANDS 91 If a sasi transgressor in the community was found to have stolen some- thing from another's grove, then a completely different punishment is levied. Aside from the fine, he is punished as follows. If someone is caught red-handed who is clearly the thief, he is nabbed, stripped, and his body is wound round with young coconut leaves.... [He] is followed around by the drum beater, gong bangers, [gangs] of howling children and dragged through the community. He must then call out in the local language or [in] Malay: "Don't steal like me, don't steal like me." In gen- eral people believed that the man wouldn't live much longer. What was certain, even if he lived, was that life for him in his community or in neighboring communities was no longer possible or desirable. All honor was stripped from him. He was shamed in every sense of the word. (Volker 1925:301-302)

While fishermen in 1991 recalled stories of violators being bound and shamed with the sasi sign who were forced to walk about the community shouting, "I have violated sasi!", most punishments implemented today involve money. "People would no longer submit to being bound today," explained a head of Nolloth vil- lage on Saparua Island. "They would find such a punishment intolerable" (Matatula 1991).

Changes in Marine Sasi, 1953-199 1: A Market in Mollusks

Prior to the 1960s, although reefs were within the boundaries of marine petua- nans, few rules regulated access to reef fish or mollusks. Trochus niloticus, a reef- resident mollusk, was gleaned freely from the shallow waters and reefs. The animal inhabiting the shell was extracted and eaten and the shells thrown away. During the 1960s, a commercial market developed for trochus shell, which is used to make paint pigments and a variety of ornamental items, including buttons, for East Asian markets in Taiwan and Japan as well as for European consumers (Reid 1992). From this beginning, and continuing through the 1970s and 1980s, trochus extracted from Central Moluccan reefs became the object of an interna- tional system of commerce linking local fishermen and fisherwomen, sasi institu- tions and laws, local-level government officials, marine commodities traders, and international exporters. The market for troches has driven changes in the focus, structure, and operation of marine sasi institutions and practices for the last three decades. Historically, coastal fishing communities have been among Indonesia's poorest peoples. Fishing, even when combined with agriculture in small gardens and swidden plots, is a notoriously risky way of making a living. In the Central Maluku, the rise of a commercial market in trochus created a window of economic opportunity for local traders on the islands of Ambon and Banda, village-level gov- ernment officers searching for funds to supplement their cash-starved routine bud- gets, and local fishermen. During the early 1950s, merchants based on Banda 92 CHAPTER 4 Neira Island began buying trochus and other shell products from fishing commu- nities on nearby Hatta, Al, and Ran islands. At roughly the same time, merchants based on Ambon Island journeyed to Saparua, Haruku, and Ceram islands and began offering cash for trochus. Government officials soon realized that the sasi system offered an institutional and legal means to control the trochus harvest and its profits. In 1968, for ex- ample, Mr. Matatula, the village-level government head of Nolloth village (Saparua Island), issued a proclamation declaring the existence of a trochus sasi within community waters and asserted control of the sasi on behalf of local gov- ernment (Matatula 1990; Matatula et al. 1990). The formerly "free" access of in- dividual local families to the community-managed marine petuanan, he declared, was forevermore prohibited. Matatula's unusually candid "History of the Trochus Sasi in Nolloth Village" states, Until 1968 ... the trochus gathered even more attention, because the market became better and better by day. It was [in 19681 that the com- munity was never again permitted to take trochus in a free manner... [B]ased on the decision of the Government Body, all the income from the trochus must be allocated to all community needs and the Village Government generally and [this income] will not be permitted [to be used] for private [needs]. (1990:1)

Indeed, by 1988 trochus exports were big business throughout Maluku Province. In a single year, the province's total volume of shell exports tripled, from less than 80,000 kg in 1987 to more than 256,000 kg in 1988. From the shallow shelf and reef of Kei Besar, an island in southeastern Maluku Province, a total of 7.5 tons of trochus shell with an approximate value of US$65,000 was harvested in 1989 (Abrahamsz 1991). By 1991, trochus shell was selling for Rp16,750 (more than US$8) per kg. Governmental responses to trochus depletion throughout the Maluku, how- ever, were far from uniform. At the same time that village-level governments began revising sasi in their own interest and asserting their control over community ter- ritories and revenues, higher-level government actors in Jakarta moved in different directions. In 1972, Decisional Letter of the Minister of Agriculture No. 327 al- ready had declared Trochus niloticus a creature that "cannot be found" or is "rare" throughout Maluku Province (Soelaiman 1978:4). In 1987, as trochus extraction soared, the minister of forestry declared T. niloticus a protected species (MOF 1987). His decision meant that trochus harvests had become illegal. Ideas about conservation were emerging from the political center, in government ministries far from Moluccan shores. As prices and the pace of trochus extraction increased, the geographical extent of government-driven revisions of sasi, including reallocation of benefit streams and rights, spread throughout the Central Maluku. On Hatta Island, for example, rights to harvest reef-resident creatures, including trochus, remained open to all MALUKU ISLANDS 93 local residents before and after the late 1960s. In 1965, however, a village-level governmental regulation (MOF 1987) established that the village government was entitled to 6 percent of the total trochus harvest. The motives of the impoverished village head, who explained "that village government income in Hatta is derived solely from the trochus profits" (Saidjan 1991) were clear. By 1991, on Ai Island, several kilometers northwest of Hatta, the local government share of the trochus sasi was one shell out of every three harvested, or 33 percent of the entire reported harvest. In that same year, on Saparua Island, village governments in Porto and Paperu asserted or considered asserting government entitlement to 100 percent of the total trochus harvest; Nolloth's government had staked its 100 percent claim as early as 1969. While the government share of the profits from trochus sasi varied widely from island to island, a relatively rapid, market-driven restructuring of the legal, eco- nomic, and institutional aspects of sasi clearly had occurred between 1970 and 1990. This restructuring was in the direction of centralization and control of access to reef areas and commodities in government hands. Developments in several communities on Saparua Island exemplify these changes. In 1990, for instance, a Nolloth village head issued another version of sasi regu- lations (Matatula et al. 1990). This entailed progressive seizure and tightening of governmental control over the marine sasi area through narrowly drawn regulations controlling the space and time regulations, as well as others specifying the be- havior, identity, and gear of those permitted to enter the sasi area. These regula- tions, signed by four customary officials (the pakter, the kewang besar, the ukulima, and the ukulua) as well as by the government head (a former police officer), demon- strate how formal government has effected rapprochement and operational collab- oration with customary authorities. Through government proclamation, sasi rules, roles, and institutions have been seized and squeezed in a progressively more cen- tralized governmental embrace. Moreover, the language of the Nolloth regula- tions-legalistic, scientific, precise and bureaucratic, peppered with spatial coordi- nates and fixed on maps, numbers and alphabetized remarks-presents a striking contrast to the ritual speech, incense, and silent offerings of sasi ritual.

Consequences of a Revised Sasi: The Distributive Justice Question The social and economic consequences of governmental codifications and cre- ative manipulations of the sasi system such as those implemented on Saparua Island in Nolloth, Porto, and Paperu villages are not yet clear. How do local vil- lagers, sasi office holders, and government officers evaluate these dramatic changes? The question of distributive justice inevitably arises in assessing the con- sequences of increasing governmental control of sasi. In 1990, the Nolloth village head painted a positive image of these changes:

Beginning in 1969 and continuing to the present day, management of trochus has developed progressively, in accordance with the growth of 94 CHAPTER 4 the world and our country, with the result that we, as Village Leaders, are always working and making strenuous efforts to provide guidance and explanations concerning village-level economic development gen- erally, and trochus particularly, as well as various kinds of shells, coral, rocks, sand, gravel, decorative fish, and sea cucumber. (Matatula 1990:1-2)

Nolloth's village head points with pride to the material improvements in village life he has implemented with the proceeds from trochus sales. Since 1986, these have included the construction of a drainage ditch for expelling excess rainwater, a sea- wall, streetlamps, two wells, a village coconut grove, and two police posts. A vil- lage museum is now being built with sasi proceeds. In contrast, a fisherman from Paperu village, commenting on the rumored im- position of a 100 percent government monopolization of the trochus sasi in Porto village in 1991, strongly rejected the idea of government control of profits and key decisions on labor, timing, and gear:

The sasi on schooling fish is an heirloom [pusaka] handed down from the grandfathers, while the sasi on sea cucumber and trochus is com- pletely new and only began in 1989. I do not agree with what is being proposed in Porto village. The idea of a corporate village right [hak negeri], separate from the rights of individual families to fish and harvest in community waters, was never asserted before. In the old days, people made contributions, on a voluntary basis, to the village government or they gave privately, as families, to the Raja. (Kewang Paperu 1991)

The same man, a respected marine kewang, focused on the potential for corrup- tion and diversion of profits if village government succeeded in centralizing sasi administration and control of the economic flow it generates:

If the Porto government controls all trochus rights, who knows how many trochus or sea cucumber will be gathered? Who knows who will make profits from those operations? People could get all kinds of hidden profits from this procedure, and steal our property for their private gain. We have an expression, "Small fish are eaten by the big ones." It means that small people, the common people, who are like small fish, are eaten by the big ones, like the silvery Bobara' fish. The common people may die making the big people rich. (Kewang Paperu 1991)

Government assertion of control over sasi often means control of the identity of laborers (villagers or workers hired from outside), the bidding process through which marine commodity traders are selected as purchasers, the minimum price accepted per kg of shell, and the technologies and manner in which trochus is ex- tracted. All of these decisions have allocational as well as environmental conse- quences that as yet are not clear. Private exporters as well as local governments also vie for control of inshore MALUKU ISLANDS 95 fisheries resources. In the southeastern Maluku islands Aru and Kei, Indonesian entrepreneurs based as far away as Bali have succeeded in acquiring rights to community-owned fishing grounds by advancing loans-for which the petuanan, or the right to harvest it, is given as security-to individual families during the monsoon season. Once indebted, many families transfer their rights to these en- trepreneurs. In some communities, outsiders have acquired sole rights to harvest local marine resources, resulting in loss of local control of the community's re- source base. Recent studies from Aru and Kei document significant loss of income by local residents as a result of these transfers (Abrahamsz 1991; Khouw and Simatauw 1991).

A Crisis in Environmental Management?

The rise of a market for trochus, in conjunction with widespread rural poverty in coastal communities and increased motorization of small craft, also may have gen- erated a crisis in reef-resource management throughout the Maluku Islands (Reid 1992). In the inshore waters of many islands, trochus, mother-of-pearl, and other reef-resident species are being overexploited and depleted (Abrahamsz 1991; Reid 1992; Zerner 1991a, 1991b). Beginning in the 1960s, sasi prohibitions against har- vest of trochus lasted from three to five years, a period sufficient to allow popula- tions to mature and reproduce at least once (Kastoro 1991:22-23). As trochus prices have risen and local needs have increased, including those of fishing fami- lies and government offices, the interval between harvests has declined progres- sively on many islands. On some islands, divers are permitted to use scuba tanks to conduct the trochus harvest. Rapid motorization of small-scale Sulawesian craft has made it possible for local reefs to be depleted by roving crews of opportunistic fishermen from other islands (Reid 1992). The result of this constellation of devel- opments is that Moluccan trochus stocks have been seriously depleted. On Saparua Island, for example, annual trochus harvests are yielding only 800 kg, whereas previous harvests on a three-year cycle were 3,000 to 4,000 kg (see Figure 4.1). During the Japanese occupation of Hatta Island in the 1940s and the immediate postwar confusion, trochus was not harvested for eight years; when harvesting resumed, in 1950, the yield was 30 tons. From 1950 through 1984, sasi was opened at two-year intervals on Hatta, with yearly yields of approximately 2 tons. Since 1984, sasi has been opened every year on Hatta, and yields are down

to about 1 ton.

Marine Environmental Problems and the Cultural Construction of the Idea of Nature The causes of overexploitation are multiple and converging. In addition to local governmental needs for routine budget funds, the national government also has Figure 4.1

YEAR Kilograms 1969 1972 1975 1978 1979 1980 1984 1986 1987 1988 1989 1990

4,250 4,100 3,950 3,800 3,650 3,500 3.350 3,200 3,050 2,900 2,750 2,600 2,450 2,300 2,150 2,000 1,850 1,700 1,550 1,400 1,250 1,100 950 800 650 500 350 200 50 J

Trochus shell yield management, Nolloth Village, 1969-1990.

96 MALUKU ISLANDS 97 stimulated village-level competitions that exhaust local coffers and place addi- tional pressure on extractive industries. But local governments are not the only nor invariably the primary source of pressure for trochus extraction. Rapidly rising consumer desires, stimulated by television images of a growing Indonesian middle class and its consumption patterns, are pushing local govern- ment and customary officials to shorten the interval between sasi harvests. In- creased population densities on isolated Moluccan islands, in addition, are leading to increasing needs for alternative sources of income. Despite evidence that shortened intervals result in drastically decreased stocks, local officials claim that villagers' need for income-to perform religious rituals, to pay school fees, and to acquire consumer goods-are forcing them to extend the harvest period. In response to a question about the decreasing interval between harvests, the village chief of Hatta replied,

We open sasi so frequently now because there are more people on this island and making a living is very difficult. It is hard to make a living on this island. In order to meet the operational needs of the village and its inhabitants we must use village-wide auctions of the trochus harvest as well as sea cucumber. (Saidjan 1991)

Although conventional community-based development narratives often demo- nize private-sector agents and government officials in analyzing the depletion of local environments, in the Maluku, and probably elsewhere, the story is more com- plex. Both government and private-sector actors may be exerting positive restrain- ing influences on pressures toward overexploitation driven by local community de- sires. Private-sector trading companies that establish strict minimal size limits for acceptable trochus may discourage extraction of immature trochus indirectly. In some areas, local officials claim they have attempted to resist local villagers' desires to shorten the interval between harvests:

Local people do not understand ideas of sustaining the environment. They just want new things, good things, beautiful things. They are looking for their own happiness. From their perspective, the best way of getting a satisfactory life cheaply and easily is from a yearly auction of the trochus. I am convinced that we must lengthen the interval between trochus harvests to at least three or four years, but my community does not grasp why we need to limit harvests. (Matatula 1991)

Local communities, moreover, may not grasp the need to limit harvests because their conceptions of the nature of the marine environment differ radically from contemporary scientific understanding of marine population dynamics and the ef- fects of extraction pressures on stocks. The complexity of Moluccan cultural conceptions of the marine environment 98 CHAPTER 4 and societal relationships is apparent in the following interchange with Mr. Matatula, the government head of Nolloth village:

Charles Zerner (CZ): Why have trochus yields decreased? Why didn't you keep sasi closed for a few more years? Government head (GH): There is a difference of opinion between my own views and what many people feel about this question. Many people think that the amount of trochus found by divers depends on the proper performance of sasi rituals. CZ: What is the relationship between ritual performance and the amount of trochus found on the reef ? GH: I'll give you an example. Before I permitted trochus sasi to be opened in 1991, we sent out some men to dive and investigate the condition of the trochus. On a single dive each of them found four large shells within minutes. We took this as an indication that there were plenty of mature trochus, so we opened the sasi. On the first and second days of the season, trochus were plentiful. But by the third day it became more difficult. In the following weeks it was almost impos- sible to find a single trochus. CZ: Perhaps there weren't many trochus to find? GH: People here suspect that the trochus were made to vanish. Disappeared. They reason that the ritual opening of sasi must have been incomplete. The ritual must have not been performed in accordance with customary rules [peraturan adat]. (Matatula 1991)

This interchange, which occurred early in my exploration of sasi, suggests that Moluccan societal relationships and conceptions of causation in the natural world are culturally distinct and laden with implications for natural resources management. Further conversations with Moluccan fishermen revealed that many imagine the marine world to be populated by a highly responsive community of invisible spirits. These spirits are believed to inhabit particular places within the Moluccan terrestrial and marine landscape, including promontories, graves, knolls, and mountaintops, as well as submerged places within the marine petuanan. Watchful spirits listen, see, and respond to the everyday practices as well as the ceremonial performances of the community. A fisherman's fate, as well as his luck in fishing- whether fish cluster about his net or disappear from sight-depends upon his re- lationship to these fractious spirits of the place. From this Moluccan perspective, the effects of shortened harvest intervals on trochus stock replenishment or the consequences of increased extraction pressure through the use of scuba gear may not be relevant questions. For many Moluccan fishermen, the key question concerns the status of their personal relationships to a local community of ancestral and environmental spirits. Spirit polities can draw MALUKU ISLANDS 99 a school of fish to the scoop nets and can just as easily "disappear" them and the fishermen's nets in an instant. Fishermen also can be disappeared. Late at night, on the island of Halmahera in North Maluku, my research assis- tant, Hasmi Bandjar, and I learned more about how Moluccan fishermen con- struct images of the marine environment and about the ways they conceptualize relationships between marine rituals and the presence of fish. Shalan, who grew up in an isolated cove in Kau Bay, told us many stories of growing up on a bay pro- tected by unique practices and a family of local guardian spirits. In response to my repeated question, "Why are only scoop nets permitted and other gear prohibited during fish drives?", he quietly revealed a landscape in which fish, spirits of the cove, gear restrictions, and ritual performances are linked in a dynamic, morally sensitive relationship:

Shalan: If you actually use a net, the net will disappear and the fish will not be catchable. They can disappear. CZ: Why? Shalan: Because there are guardians there. They will only permit us to capture fish in scoop nets. For example, if a school of fish came to a place and they disap- peared, it means that people will be compelled to say prayers and scatter offer- ings in the spot. The offerings are given symbolically, to suggest that there is food. Yellow rice, eggs, and a chicken are offered in a little scoop net. Offerings are not only performed when a school of fish disappears. Each year a group of village elders sails to both promontories flanking the bay and makes offerings of incense and yellow rice in a ritual known as cakalele. In the center of the bay, they sacrifice a white female chicken, place it in a small, floating fish scoop, and set it adrift. The bay is believed to be the site of a former village whose inhabi- tants fled when the waters rose. The places where offerings are now made are believed to be inhabited by a spirit family descended from the inhabitants of the inundated former village. Before a fish drive is commenced, village elders sail to the center of the bay and inform the female spirit believed to inhabit that spot that they wish to begin the drive. (Shalan 1991)

Shalan's and other stories provide a window onto local ideas of the marine world and fishermen's moral relationship to it. At the same time, the cleavage- indeed, the chasm-between these conceptions of the natural world and the so- cietal relationships they entail and contemporary scientific notions of the inter- actions between natural dynamics and social practices is striking. While the scientific value of local fishermen's environmental knowledge increasingly has been recognized, the cultural, aesthetic, and ethical values of indigenous fishers' models and metaphors of nature and society often have been ignored. From the perspective of conservation or environmental management, however, the slip- page between these indigenous conceptions of nature and societal relationships 100 CHAPTER 4 to it, and Western scientific models of environmental dynamics, also must be addressed.

Inventing a Green Sasi, 1968-1991

The historical trajectory of a community-management structure, like that of any political and legal institution, is in part the product of the way interested parties and institutions have imagined its purposes, narrated its history, and attempted to shape it to suit their own purposes (Bowen 1989; Hobsbawm 1985). The history of sasi in remote coastal hamlets of the Maluku Islands is largely the product of the ways various powerful and differently positioned interests have interpreted, and continue to interpret, this institution's administrative structure, rules, and most importantly, purposes. In Indonesia during the early 1980s and continuing into the 1990s, new kinds of talk about the environment, economic development, and, later, social equity, began to inform central government policy, institutions, and the discourse of po- litical leaders and policy makers. Indonesia's Basic Law of the Environment was passed in 1982 (GOI 1982), and the influential Department of Population and Environment was formed in 1978. WALHI, the Indonesian Environmental Forum, an umbrella agency for more than one hundred nongovernmental envi- ronmental- and social-activist groups scattered across the archipelago, was formed in 1980. Between the remote villages of Maluku Province and the political center in Jakarta, people, ideas, and new ways of talking about the environment and com- munities moved across the seas. Since the 1980s, environmental and govern- mental representatives have begun to interact. Discourse among Java-based gov- ernmental and nongovernmental institutions; academic and scientific institutions in Ambon, the Moluccan capital; and formal and customary village leaders in remote Moluccan hamlets has resulted in changes in the interpretation and, pos- sibly, in the implementation of sasi. The story of Haruku village on the Central Maluku island of the same name is representative of these changes and the emerging new institutional actors. In Haruku, Mr. Ririmasse, a charismatic government head influenced by contempo- rary ideas about conservation and sustainable development disseminated by a national-level institution, decided to revive sasi and, in the process, make it envi- ronmentally acceptable. In 1992, Mr. Ririmasse recalled, "The situation of sasi was weak in the 1980s. No one on the island knew about or obeyed the rules. The local elite had no idea of resource conservation or sustainable management. The local youth were not interested in sasi. They were leaving the village and moving to the city" (Tjitradjaja 1992). The Haruku head, however, was a close friend of an official representative of the Indonesian Department of Population and Environment, Pak Daisy, whose influ- MALUKU ISLANDS 101 ential ideas led to the reshaping of sasi. According to the head, he "got ideas from Pak Daisy and ... reformulated sasi as an instrument of conservation. I saw the youth rejecting the sasi, so I typed them up and added a few rules" (Tjitradjaja 1992). More than "a few rules" were added when Haruku's sasi law was codified in 1985 (Ririmasse, Kissya, and Ririmasse 1985). In the process of reviving sasi, the institution was actually being reinvented. As represented in the new rules, sasi's purposes were articulated as "sustainable management and protection of living stocks as well as the equitable distribution of economic benefits." Community practices laden with a surplus of meanings and functions were being recast as community-based resource-management institutions focused on equity and pro- tection of stocks. Government-generated shifts in the functions of sasi, moreover, were paralleled in the political and aesthetic spheres. The "voice" and the appearance of sasi were changing. Rather than make pronouncements with a conch shell trumpet and the chanting of a ritual poem, sasi's new bureaucratic voice announced, in tightly con- structed sentences, that

sasi is a prohibition against the taking of various kinds of specific nat- ural resources within a specified time period in order to conserve while guarding [its] quality and population [level] as well as conserving the social rules of local society and the equitable distribution of economic returns for the whole society. (Ririmasse 1984:4)

Sasi not only was being "greened," it was being married to the government. In one village in the Central Maluku, a government head proudly announced that ke- wangs' assistants had been made to wear uniforms, suggesting the development of an increasingly rationalized and bureaucratized structure. In a section entitled "The Legal Basis Supporting Sasi," this newly written constitution acknowledges the dual sources of sasi authority: "Sasi contains rules that are ordered through vil- lage legal decisions and the Kewang Institution that is a customary institution which is delegated [the responsibility] to monitor implementation of the afore- mentioned sasi rules" (Ririmasse, Kissya, and Ririmasse 1985:4). Rather than use the imprecise but familiar, even familial, toponyms (named topographical features of coastal promontories, rocks, and marine drop-offs that are the landscape power spots and naturally visible cleavages in the marine land- scape), the Haruku sasi constitution specifies boundaries of the marine manage- ment area in meters and topographically fixed coordinates located on a map. As local perceptions and memories about particular marine landscapes were being transformed, mapping reefs and sasi boundaries and positioning them within a grid of mathematical coordinates constituted a means through which govern- mental definition of and dominion over community-held sea space is advanced (Vandergeest and Peluso 1993). 102 CHAPTER 4 New Rules and New Purposes The codification of Haruku's sasi rules in 1984 was not a rote recapitulation of rules already transcribed and manipulated by the Dutch. Mr. Ririmasse, the Haruku government head, inserted several new rules directed toward conserva- tion and sustainable management of resources. Rule B.8 is directed toward pre- vention of erosion and stabilization of riverbanks: "People are prohibited from cut- ting down trees near the river bank throughout the area under coastal sasi with the exception of sago trees" (Ririmasse, Kissya, and Ririmasse 1985:1). Under a divi- sion titled "For Men," Rule 6 mandates: "People are prohibited from entering [the estuary and river] with a motorized boat or any kind of a speed boat while the motor is on" (Ririmasse, Kissya, and Ririmasse 1985:1). Prior to his term in office, Mr. Ririmasse explained, "there were only two basic kinds of sasi, a sasi on clove production and on the ocean." By the time he came into office, motorboats had begun to travel into the estuary and up the river, dis- turbing the migration of marine fishes and cutting them up with their blades. "Local people opposed the entry of the motor boats," he explained, "and I added a rule prohibiting entry" (Tjitradjaja 1992). Environmental problems are not static but change over the course of time. The role of contingency in shaping legal institutions, moreover, can not be underesti- mated. After Mr. Ririmasse's retirement, a new problem emerged: the entry of bagan, or fine-mesh lift-net platforms, into the estuary area. The new government head claimed that he was "unable to act to prohibit the entry of bagan because `There was no sasi rule prohibiting the entry of bagan"' (Tjitradjaja 1992). In the early 1990s, frustrated local fishermen took matters into their own hands, working with a courageous kewang and a vibrant nongovernmental organization known as HUALOPU. In a burst of creative advocacy, the kewang, nominally the traditional sasi authority, relied on national fisheries laws prohibiting the use of fine-mesh nets (which indiscriminately capture fish of all kinds and at all life-cycle stages) and pressed his case for expulsion of the bagan with local police. HUALOPU members successfully supported the kewang in his attempt to goad the police into action by taking photographs documenting the illegal entry of the bagan into the community-controlled rivermouth. The alliance of a kewang with a nongovernmental organization and the novel use of national fisheries statutes and documentary evidence to expel the bagan would have been unimaginable two decades ago. Indeed, the alliance parallels the productive diffusion of leading-edge conservation and economic-development strategies from the political center to the political periphery that led to the revision and rescripting of the purposes and positive regulations of sasi. In the 1990s, the development of sasi purposes, institutions, and rules con- tinues to be fueled by a volatile mixture of environmental and social values, in- novative conservation planning, and economic-development schemes brought into productive conjunction with local institutions and traditional resource-man- agement practices. Sasi, initially conceived as a traditional management institu- MALUKU ISLANDS 103 tion, is now being strategically edited, revised, and integrated into current envi- ronmental ideas and plans. Far from having germinated in an indigenous institutional or political vacuum, these ideas are being nurtured and disseminated by a variety of sophisticated in- stitutional bodies with overlapping and often intersecting interests. Among these organizations are the Department of Population and Environment and the In- donesian Environmental Forum (Jakarta); the Environmental Studies Center, Fac- ulty of Law, and Fisheries Faculty of the University of Pattimurra, Ambon; and the innovative, energetic HUALOPU, which focuses on improving the lives and liveli- hoods of coastal peoples throughout the Maluku while seeking to integrate con- servation with sustainable development. The role of the Department of Population and Environment has been particu- larly important. As the governmental bully pulpit, this department and its elo- quent, charismatic former minister, Dr. Emil Salim, constituted a national voice for environmental affairs. Under Salim's direction, the department addressed a wide variety of constituencies within and beyond Indonesia's national boundaries. The department's annual award of the Kalpataru, a kind of national seal of approval initiated in 1981 for good local environmental housekeeping that is awarded to communities throughout Indonesia, authorizes and honors these constituencies. Awarding the Kalpataru engenders further developments. Indeed, an assistant minister avidly explained that "the Kalpataru's purpose is to drive conservation, preservation, and restoration of the environment, and sustainable development. Our office, although a State department, can not implement policies. It can only promote policies, disseminate articles, issue guidelines, and advise the president." The award--one that generates considerable prospective pressure-"is an incen- tive," he continues. "But the Kalpataru is not just given to those who do things well. During the Kalpataru presentation session, the minister also points to places or problem areas in the district which still need attention. Public discussion of these areas forces local district heads to act" (Tadjoedin 1992). During the 1980s, the Kalpataru was awarded to several Central Moluccan communities practicing sasi. In 1982, for example, based on the positive recom- mendation of the Environmental Studies Center of the University of Pattimurra, Ihamahu village on Saparua Island was awarded the Kalpataru. Recognition of the village also generated the impetus for further development. In acknowledging receipt of the award, the Ihamahu village leader looked backward in history and, at the same moment, forward, retrospectively reconstructing the historic inten- tions of former sasi practitioners:

When the regulations that were implemented along with the sasi system that was performed since the old days are viewed historically, it may be stated that Moluccan society, particularly Ihamahu society, through the kewang, were already conscious of the importance of their environment in terms of its importance to them as individuals. All of this was put into 104 CHAPTER 4 effect through custom [adat], without knowing that the modern world needs acts of conservation toward living resources. Whatever was done by the Ihamahu kewangs since the times of the old rulers-perhaps they understood conservation issues, even though [this understanding] was only based on realities and signs from their own observations which were still rather old-fashioned. (KKI n.d.:9)

Although from a narrow, positivist perspective the retroactive assessment of sasi as an intentionally conservationist institution from its "beginnings" is probably in- accurate, these interpretive fictions play a positive role as enabling myths (Bowen 1989; Cole 1986; Hobsbawm and Ranger 1983). They are used to ground and guide future action at the governmental center and in remote provinces, and to ad- vocate for historically marginalized and disenfranchised communities. Historic sasi practices in Ihamahu village, for example, have been reconfigured retroac- tively as directed by current institutional goals. In these reconfigurations, historic sasi is imagined as a "good conservation system" executed by environmentally conscious practitioners. The certification of sasi practices and institutions through awards from the In- donesian political center also compels future institutional changes. An Ihamahu official emphasized that "after receiving the Kalpataru, it [became] clear that the responsibilities of the Ihamahu Kewang became heavier in the sense that the Kalpataru goblet was a reminder to the Kewang to always protect the living envi- ronment.... Because this group [the kewang] is conscious, [they] provide an En- vironmental Support Group with the consequence that several activities have been added in addition to the routine performed since the beginning [of sasi]" (KKNI 1987:9-10). Since award of the Kalpaturu, the "Ihamahu Kewang [now] perform in a regular manner" (KKI n.d.:9) a variety of new activities, including planting vegetation near the sites of key water sources; building fences; and planting mangroves for coastal stability, water purification, and protection of the marine and coastal fisheries. Other rules prohibit commercial extraction of sea cucumber, coral, trochus, and garu wood (used for incense) from the reefs and forests under community control. Moreover, the Ihamahu kewang make periodic reports on their activities to the En- vironmental Studies Center and the Department of Population and Environment, communicating personally with Dr. Salim. Projecting a conservationist past onto current sasi practices, these interpreta- tions authorize supportive actions by important central governmental and NGO actors and, at the same time, guide or goad local communities toward conserva- tion-promoting behavior. By directing the attention of those at the political center, especially the respected Dr. Salim, toward the political periphery, these enabling fictions link the energies, knowledge elites, and, potentially, the financial flows of powerful Indonesian and transnational cosmopolitan centers to singularly remote locations throughout the archipelago. Indeed, Salim has informally remarked that MALUKU ISLANDS 105 the "purpose of these awards is to recognize the achievements of these communi- ties and to stimulate them to go even further" (Salim 1992). By the early 1990s, sasi in the Maluku was marked by intense interaction be- tween diverse kinds of organizations. The Department of Population and Envi- ronment, HUALOPU, and the University of Pattimurra's Environmental Studies Center continue to be key institutional players interested in the social and envi- ronmental potential of sasi. Both of the latter two groups are interested in practical attempts to shape and develop sasi. Both are intensely committed to exploring the possibilities they envision: sasi as the institution with the greatest potential for supporting sustainable environmental management and biological diversity in reefs, rivers, and inshore waters and, most importantly, as an instrument for pro- moting distributive equity and democratization. By 1992, a representative of the World Bank had visited the HUALOPU office in Ambon and traveled to a neigh- boring island in search of new "institutional alternatives and opportunities for marine resource development in Indonesia" (Leibenthal 1992).

Conclusion In the past sasi was constituted as rules, practices, and sanctions that must be im- plemented to avoid the punishments of spirits and sasi enforcers. Sasi also has been imagined and manipulated as a secular social-organizational armature facil- itating commerce between traders from distant Asian and European lands, local rulers, and small Moluccan communities. In an era when concern for equity in de- velopment, the political and cultural rights of indigenous communities, biodiver- sity, and sustainable growth have become global as well as national issues, sasi is now being imagined and discoursively constructed as a changing embodiment of these values and aspirations. What is not at all clear is whether these discoursive changes will result in significant or lasting changes in the purposes or practices now known as sasi and what the environmental and socioeconomic consequences of these changes will be (Zerner 1994a, 1994b). Imagine sasi as a palimpsest-a parchment or writing tablet on which, at pre- vious moments in its history, several kinds of inscriptions have been superim- posed. Like a palimpsest, sasi contains traces and fragments of its previous incar- nations. In its totality, sasi contains regions of translucence as well as areas of opacity, archaic residues as well as recent overlays that may never be understood. Obsession with the original intentions or motives for sasi, like similar preoccupa- tions in constitutional law, are beside the point. For everyone except interpretative fundamentalists attempting to prove a singular and simple original motive for sasi (conservation, for example), understanding why and how this institution has changed, and how to generate positive changes in the future, is far more important. Above all, the sasi emerging in the 1990s is a hybrid creature, shaped in a cultural border zone where traders, fishers, and markets interact with global 106 CHAPTER 4 conservationists and social nongovernmental organizations. Sasi is, and was, as much a product of local religious fears and practical purposes as it is the product of continuous cosmopolitan influences such as the desires of Javanese and Portuguese traders sailing to Moluccan shores on monsoon winds and, more re- cently, the intentions of contemporary trochus merchants cruising from cove to cove in speedboats and calibrating world market prices electronically. Given the changing cross-currents and competing interests in this complex interaction, how can coexistence of Moluccan reef and human communities be improved? The most recent interpretations of sasi, particularly those created by inspired ministers and aspiring NGOs, constitute a social mandate and environmental constitution for change. Like all constitutions, the broad-ranging collection of sasi pronouncements needs to be interpreted and operationalized through specific and concrete actions. Sasi can and should be reread as legal mandate and institutional armature deployed in the service of reef and inshore marine fisheries conserva- tion, sustainable development, and community equity. Many foundational elements for such a contemporary reading, including access imitations to fisheries in terms of territory, season, gear, and people, already are )resent. A second and key element is that of the "kewangate," the implementatory, udicial, and enforcement institution that plays a central role in putting sasi into effect. Third, but not last, sasi is an ongoing practice or habit with which local people are familiar. These three pillars of sasi-its institutional, legal, and cultural core-can and should be used in the attempt to create new ways to conserve reefs and assist local-community development. To effect such a transformation, a host of other actions is necessary. Among these steps should be a series of major research initiatives, community-based en- vironmental education and alternative economic development programs, and, possibly, political and legal reform. Baseline investigations into sustainability of current practices and their relationship to sasi are needed. For example, how does sasi affect resource bases in communities under varying conditions of sociopolit- ical control, population pressure, gear use, reef conditions, and harvesting prac- tices. More quantitative information is needed on site-specific catch levels, sasi re- strictions, and resource availability, not only for trochus but for a variety of reef species. Such studies, conducted in a variety of sites, could reveal the dynamic links between social and legal organization, markets, and population pressure, and their effects on habitat as well as on species conservation and management. Much more information needs to be gathered about the socioeconomic effects of marine sasi regimes under different conditions of property rights and market penetration. For example, we need to know how recent changes in sasi codifica- tion affect the differential access of poor families, women, and nonindigenous communities to marine resources for subsistence as well as economic needs. Poor families or individuals may be made even poorer by these changes, decreasing any economic incentives to sustainably manage reefs. MALUKU ISLANDS 107

The extent to which sasi is practiced throughout the Maluku, or where and how, is not clear. Several sources suggest that sasi practices are "on the wane" or "wilting" (Lokollo 1988; von Benda-Beckmann, von Benda-Beckmann, and Brouwer 1992). If we do not yet know how extensive these practices are or why they are described as "wilting," then expending considerable financial resources on revitalizing or restructuring these practices for socioeconomic and conserva- tion purposes may be unwarranted. Several long-term ethnographic accounts of sasi, conducted in sites that vary in degree of market interaction, governmental control, and NGO influence, would be extremely useful in tracking the institution in action. A comparative, Malukuwide survey of sasi that assesses the relative strengths of the factors driving change in sasi and regional variation in the processes of social, political, and economic con- trol should complement long-term ethnographic research. Team research pairing marine biologists with ethnographic researchers would yield the most useful kinds of results. If sasi is to become an institution that embodies the emerging global values of biodiversity and habitat conservation, social equity, and economic development, a variety of research probes are not the only kinds of action needed. To assist in the creation of a sustainable sasi, a program in environmental education at the vil- lage level, effectively implemented by local NGOs and village youth, is much needed. At the technical and economic level, exploration of the potential of other forms of marine cultivation or management, including cultivation of giant clams, trochus, and pearl oysters, would be very useful. The Philippines, Japan, and other places in the South Pacific whose communities have experimented along these lines have lessons to offer. The further development of sasi also may require administrative and legal re- forms that recognize community-based jurisdictions for management. Indeed, rep- resentatives of HUALOPU, the Environmental Studies Center, and the legal bureau of the Moluccan governor's office are collaborating to draft regulations along these lines, although it is too early to tell whether these reforms will succeed or what their effects will be. Attempts to recognize marine tenurial areas vested in local communities, how- ever, must avoid the pitfalls of romanticism or naive optimism. Proponents of legal recognition for community-based marine-management areas must deal with evi- dence of long-term, historical community conflict over the issue of boundaries. As the government head of Porto village on Saparua Island explained, "there hasn't been a time when Porto has not been at war with neighboring communities over the size of its own area. The history of our boundaries is the history of conflicts and burnings, right through the last few decades" (Aponno 1991). In the marine sector, Abrahamsz (1991) notes repeated instances of intercommunity theft from mother- of-pearl beds. Marine sasi, like their agricultural counterparts, are apparently discrete and not overlapping. Rather than move for unilateral decentralization of 108 CHAPTER 4 management authority to particular villages or clusters of villages, imposition of areas of comanagement (such as those proposed for forests and marine bays in the Philippines, in which the various strands of rights in the bundle of authority are ap- portioned among local villages, provinces, and the provincial government) may be more realistic. Ultimately, a series of pilot projects could implement a "green" form of sasi around several Moluccan reefs. Among the most notable examples of innovative reef management and conservation projects are several in the Philippines and other areas of the South Pacific that seek to integrate the idea of a marine reserve, or totally restricted area, with a marine conservation area that allows for fishing under sustainable local controls. Perhaps a Moluccan form of reef reserve and fishing area that borrows from Philippine examples as well as transnational con- servation theory and builds upon already existing forms of Moluccan sasi may one day be implemented. Whether-and how-particular Moluccan fishing communities are able or eager to rejuvenate and redirect the multitude of practices and institutions clus- tered under the sasi umbrella is one of the key questions to be answered in the next chapter in the history of this remarkable and changing set of social ideas, laws, and institutions. Whether the sasi that develops actually will promote social equity, environmental sustainability, or biological diversity is uncertain. , From this brief history, however, it is clear that a new constellation of govern- ment policy makers, nongovernmental social and environmental activists, and scholars is seeking to invent new forms and new purposes for sasi. Local notions of institutional structure and law, nationalist ideology and pride, and global, transnational conservation ideas are being mixed, not melted, into these institu- tional forms and narratives in ways that are unpredictable, yet characteristic of the processes of global change everywhere (Appadurai 1990; Bremen 1988; Rouse 1991). Although the site of sasi is specifiable as a series of coordinates and con- tours on a map of the Maluku, the actions and values injected into and through this "structure" are inevitably and quintessentially transnational, relying on infor- mation and ideologies that flow between Washington, D.C., Jakarta, and remote Moluccan coves. Transforming sasi these days is an act of global imagination and improvisation.

ACKNOWLEDGMENTS Due to limited opportunity for field work, this chapter should be regarded as a speculative and preliminary attempt to deal with sasi's complex history. It is based on two months' field research in the Maluku Islands during 1991, sponsored by the Fisheries Research and Development Project (FRDP) based at the Central Fisheries Research Institute, Agency for Agricultural Research and Development, Department of Agriculture, Indonesia; a brief visit in 1992; and the analysis of archival materials. The project was funded by the U.S. Agency for International Development. Support for preparation of this paper was provided by the Woodrow MALUKU ISLANDS 109 Wilson International Center for Scholars (Asia Program) and the Liz Claiborne Art Ortenberg Foundation. Archival research and translations were completed during my tenure as a Woodrow Wilson Fellow. Revisions were completed with the sup- port of the Rockefeller Brothers Foundation and the Rainforest Alliance, Natural Resources and Rights Program. Many people and organizations have contributed to my understanding of sasi. I wish to acknowledge the assistance of Carla Sapsford, whose translations of a Dutch text made accessible an important historical window on sasi; Hasmi Bandjar, my research assistant and colleague during field research; Frances Gouda, whose passion for history and the Dutch colonial project in Indonesia has been an inspiration; and Toby Alice Volkman, who has contributed her in- sights and clarity of perception to this paper. Productive discussions with Iwan Tjitradjaja (University of Indonesia); Sandra Moniaga and Suraya Affif (Indone- sian Environmental Forum); Cliff Marlessy (HUALOPU); Meentze Simatauw, Kalvein Khuouw, and many other HUALOPU staff members; as well as the staff of the Environmental Studies Center (PSL, University of Pattimurra) have con- tributed to my understanding of recent developments in sasi and coastal com- munity management. The fishermen, local government leaders, and kewangs of Saparua, Halmahera, Hatta, Ai, and Ran islands made this intellectual journey possible. An earlier version of this chapter was presented at the Second Annual Meeting of the International Association for the Study of Common Property (Winnipeg, Canada, 1991). Michael Wright, David Western, and John Cordell provided in- sightful comments on a draft, encouraging the chapter's further development. The chapter was substantially completed in September 1992. During final revisions in 1994, the author had the opportunity to read von Benda-Beckmann et al.'s man- uscript, which provides a parallel view of sasi. Unfortunately, this manuscript came to my attention only as this chapter was going to press, so I have not been able to include ideas presented in it. I have cited it, however, for the benefit of the reader. All the usual disclaimers apply.

SOURCES Abrahamsz, J. 1991. "Nasib Lola dan Lembaga Sasi di Kepulauan Kei, Maluku Tenggara" (Fate of Trochus and Sasi Institutions in the Kei Archipelago, South- east Maluku). Kabar Dari Kampung (News From the Village) 48(IX):52-61. Andaya, L. 1993. The World of Maluku. Honolulu: University of Hawaii Press. Appadurai, A. 1990. "Disjuncture and Difference in the Global Cultural Economy." Public Culture 2(2):1-24. Aponno, J. 1977. "Peraturan Kewan Negeri Porto Tahun 1870" (Kewan Regula- tions from Porto Village in 1870). Typescript. 1991. Interview with the author and H. Bandjar, Porto Village, Saparua Island, Indonesia, March 29. Bailey, C., and C. Zerner. 1992. "Community-based Management of Fisheries Re- sources in Indonesia." Maritime Anthropological Studies 5(1):1-17. 110 CHAPTER 4 Bowen, J. R. 1989. "Narrative Form and Political Incorporation: The Uses of His- tory in Aceh, Indonesia." Comparative Studies in Society and History 31(3): 627-693. Bremen, J. 1988. The Shattered Image: Construction and Deconstruction of the Vil- lage in Colonial Asia. Dordrecht, Holland: Center for Asian Studies/Foris Publi- cations. Cole, David. 1986. "Agon at Agora: Creative Misreadings in the First Amendment Tradition." Yale Law Journal 95 (5):857-905. Government of Indonesia (GOI). 1982. Undang Undang Nomor 4 1982 (Law Number 4 Concerning Basic Provisions for the Management of the Environ- ment). Hobsbawm, E. 1985. "Introduction: Inventing Traditions." In The Invention of Tra- dition, ed. E. Hobsbawm and T. Ranger, 1-14. Cambridge, England: Cambridge University Press. and T. Ranger, eds. 1983. The Invention of Tradition. Cambridge, England: Cambridge University Press. Kastoro, W. W. 1991. "T\vo Commercial Mollusks: Giant Clams and Trochus Niloticus." Voice of Nature 23:22-23. Kewang Paperu. 1991. Interview with the author and H. Bandjar, Paperu Village, Saparua Island, Indonesia, March 28. Khouw, K., and M. Simatauw. 1991. "Hasil Survei Sosial Ekonomi Masyarakat Pulau-Pulau Aru, Maluku Tenggara" (Results of Socioeconomic Survey of the People of the Aru Islands). Ambon: Yayasan HUALOPU (HUALOPU Founda- tion). Typescript. Korps Kewang Ihamahu (KKI). n.d. Korps Kewang Ihamahu Dan Peraturan-Per- aturan Sasi (Kewang Corps and Sasi Regulations of Ihamahu). Yayasan HUALOPU (HUALOPU Foundation), Pusat Dokumentasi Dan Informasi HUALOPU (HUALOPU Documentation and Information Center), Ambon, In- donesia. Mimeograph. Korps Kewang Negeri Ihamahu (KKNI). 1987. "Korps Kewang Negeri Ihamahu Dan Lingkungan Hidup" (Kewang Corps of Ihamahu Community and the Living Environment). Paper presented at the Nature Conservation Seminar (Marine), July 2-9, TOISAPU, Ambon, Indonesia. Ambon: Yayasan HUALOPU, Center for Information and Documentaton. Mimeograph. Leibenthal, A. 1992. Interview with the author at the World Bank, Washington, D.C., September 4. Lokollo, J. E. 1988. "Hukum Sasi di Maluku: Satu Portret Bina Mulia Lingkungan Pedesaan Yang Dicari Oleh Pemerintah" (Sasi Law in the Maluku: A Noble Por- trait of the Village Environment Sought By the Government). Paper presented on the twenty-fifth anniversary of the University of Pattimura, May 11, Ambon, Indonesia. Matatula, A. 1990. "Sejarah Bia Lola (Trochus niloticus) di Desa Nolloth (Keca- matan Saparua)" (History of the Lola [Trochus Niloticus] in Nolloth Village [Saparua District]). Nolloth Village Office, Saparua Island, Indonesia. Mimeo- graph. MALUKU ISLANDS 111

. 1991. Interview with the author and H. Bandjar, Nolloth Village, Saparua Island, Indonesia, March 28. , U. Manuputty, J. Huliselan, and P. Metekohy. 1990. "Peraturan Sasi Desa Nolloth" (Sasi Rules of Nolloth Village). Nolloth Village Office, Saparua Island, Indonesia. Mimeograph. Ministry of Agriculture (MOA). 1972. Keputusan Menteri Pertanian No. 327 (De- cisional Letter of the Minister of Agriculture No. 327). Government of In- donesia, Jakarta. Ministry of Forestry (MOF). 1987. Keputusan Menteri Kehutanan Nomor 12/Kpts-II/1987 tentang Penetapan Tambahan Jenis-Jensi Binatang Liar Yang Dilindungi Di Samping Jenis Jenis Liar Yang Telah Dilindungi (Ministry of Forestry Decision No. 12 Concerning Determination of Additional Kinds of Wild Animals Which Are Protected Along with Wild Animals That Already Are Protected). Government of Indonesia, Jakarta. Reid, A. 1992. "Indonesian Fishermen Detained in Broome: A Report on the Social and Economic Background." In Illegal Entry!, 1-11. Occasional Paper Series, No. 1/1992. Casuarina, Australia: Center for Southeast Asian Studies, Northern Territory University. Ririmasse, B. 1984. "Peraturan Sasi Desa Haruku" (Sasi Rules of Haruku Village). Yayasan HUALOPU (HUALOPU Foundation), Pusat Dokumentasi Dan Infor- masi (Center for Documentation and Information), Ambon, Indonesia. Photo- copy.

, E. Kissya, and E. Ririmasse. 1985. Peraturan Sasi Desa Haruku 1984 (Sasi Rules of Haruku Village, 10 July 1985). Yayasan HUALOPU (HUALOPU Foun- dation), Pusat Dokumentasi Dan Informasi (Center for Documentation and In- formation), Ambon, Indonesia. Photocopy. Rouse, R. 1991. "Mexican Migration and the Social Space of Postmodernism." Di- aspora 1(1):8-23. Saidjan, A. F. 1991. Interview with the author, Hatta village, Hatta Island, In- donesia. April 1. Salim, E. 1992. Interview with the author, Jakarta, Indonesia, October 8. Shalan. 1991. Interview with the author and H. Bandjar, To Belo, Halmahera Island, North Maluku Province, Indonesia, April 6. Soelaiman. 1978. "Daftar Jenis Margasatwa Yang Dilindungi Dan Yang Terdapat Di Maluku" (List of Protected Creatures Which Inhabit the Maluku [through 1978]). Department of Fisheries, Nature Conservation and Protection Section, Ambon, Indonesia. Circular. Tadjoedin. 1992. Interview with the author, Jakarta, Indonesia, October 5. Tjitradjaja, I. 1992. Interview with the author, Ambon Island, Indonesia, May 17. Vandergeest, P., and N. L. Peluso. 1993. "Fixing Property in National Space: Ter- ritoriality of the State in Siam/Thailand." Paper presented at Yale University Agrarian Societies Seminar, February 23, New Haven, Connecticut. Volker, T. 1925. "Het recht van sasi in de Molukken (1921)" (The Sasi Right in Maluku [1925]). Adatrechtbundels 24:296-313. von Benda-Beckmann, F., K. von Benda-Beckmann, and A. Brouwer. 1992. "Changing 'Indigenous Environmental Law' in the Central Moluccas: Com- 112 CHAPTER 4 munal Regulation and Privatization of Sasi." Revised version of paper pre- sented to the Congress of the Commission on Folk Law and Legal Pluralism on Resouces, Rules and Identity in State and Unofficial Laws, August 27-30, Vic- toria University, Wellington, New Zealand. Zerner, C. 1991a. Community Management of Marine Resources in the Maluku Is- lands, Indonesia. Report prepared for the Center for Fisheries Research and De- velopment, Jakarta, Indonesia.

. 1991b. Key Issues in Indonesian Fisheries Law and Institutional Develop- ment: Implementation, Environmental Management, and the Rights of Small- Scale Fishers. Report prepared for the Center for Fisheries Research and Devel- opment, Jakarta, Indonesia.

. 1992. "Sea Change: Toward Community Management of Coastal Re- sources in Insular Southeast Asia." World Resources Institute, Washington, D.C. Typescript.

. 1994a. "Tracking Sasi: Transformation of Moluccan Reef Management Practices." In Community-based Coral Reef Management: Lessons from Experi- ence, eds. A. White, L. Hale, and L. Cortese, 19-32. West Hartford, Connecticut: Kumarian Press. . 1994b. "Through a Green Lens: Constructing Customary Environmental Law and Community in Indonesia's Maluku Islands." Paper presented at the Annual Meeting of the Association of Asian Studies, March 25, Boston. Institutions and Externalities in the Inland Fisheries of Bangladesh

Kazi Ali Toufique

ABSTRACT. Establishment of property rights maintained and preserved by the agents hav- over inland fisheries of Bangladesh has been able ing property rights over the resource system. to internalize appropriation externalities. How- Such a collaboration has not been ruled out ever, the agents have failed to internalize provi- in the literature (van den Brink, Bromley, sion problems due to the physical characteristics and Chavas 1995; Runge 1986), but does not of the resource systems. We have observed that analysis the rules traditionally followed by the fishers are follow from the of those who think maintained and preserved by agents having ex- that the creation of private property is the clusive property rights over the resource systems. only solution to common-pool problems. The relationship between the agents and the fish- We have drawn these conclusions mainly ers is not anonymous but of a patron-client type. from a six-month-long field-level investiga- Such outcome does not follow from the analysis tion2 carried out in the district of Sirajganj,3 of those who think that private property rights are Bangladesh, during the year 1992. Secondary the only solution to conunon-pool problems. (JEL evidence that we will be presenting in this Q22) paper also do not contradict our arguments. The resource system is defined and described 1. INTRODUCTION in Section II. The nature of property rights is discussed in Section III. We will analyze the Does establishment of property rights over nature of externalities in the inland fisheries common-pool resources (CPR) prevent com- and the mechanism devised by the appropria- mon-pool losses? Opinion is divided on this tors towards its internalization in Section IV. issue, so is empirical evidence. The main Important public policy issues are discussed purpose of this paper is to present a case where establishment of property rights over inland fisheries of Bangladesh has been gen- The author is a Research Fellow, Bangladesh Insti- erally successful in containing common-pool tute of Development Studies, Dhaka. This paper is partly based on my Ph.D. dissertation, submitted to the losses. The agents have been more successful University of Cambridge, U.K. A financial grant from in internalizing appropriation externalities the Cambridge Political Economy Society is gratefully such as rent dissipation, assignment prob- acknowledged during my research in Cambridge. I wish lems, and technological externalities. They to express my gratitude to an anonymous referee of this journal and to Mushtaq Kahn for helpful comments and have, however, failed to internalize provision suggestions. The usual disclaimer, however, applies. problems mainly due to the physical charac- ' It is important to distinguish between resource sys- teristics of the entire resource system.' This tems and resource units. Resource systems are stock supports the hypothesis developed by variables capable of producing a maximum quantity Schlager, Blomquist, and Tang (1994) that (under ideal conditions, of course) of a flow variable without harming the stock or the resource system itself. appropriators cannot address provision prob- On the other hand resource units are what individuals lems in dealing with mobile resource units appropriate or use from a resource system. Examples that do not have the attribute of storage. of resource systems are fishing grounds, groundwater Though the nature of this property rights re- basins, grazing areas, and so on. Resource units are tons of fish harvested from a fishing ground or gallons of is gime predominantly private, the relation- water withdrawn from a groundwater basin and so on ship between the holder of exclusive prop- (Ostrom 1990, 30). erty rights and the users of the resource 2 For a description of the fieldwork method, see system is not as anonymous as one would ex- Toufique (1996). pect in private transactions. More impor- Bangladesh is administratively divided into 5 divi- sions of 64 districts, which in turn is divided into 489 tantly, the rules traditionally followed by the thanas-the smallest administrative unit of govern- users to minimize common-pool losses are ment.

Land Economics August 1998 74 (3): 409-21 410 Land Economics August 1998 in Section V. Finally, Section VI summarizes value added to the agricultural product, and and concludes this paper. 9 percent of national exports (GOB 1995b). About 8 percent of the total population de- H. THE RESOURCE SYSTEM pends on this sector for its livelihood (Plan- ning Commission 1978) and about 73 per- Let us first define the domain of the re- cent of households were involved in source system under our consideration. Fish subsistence fishing in the floodplains in in Bangladesh comes from two sources, in- 1987/88 (DOF 1990). land (mainly freshwater) and marine. The marine sector is characterized by open ac- III. THE NATURE OF PROPERTY cess. In this paper we are not concerned with RIGHTS OVER THE RESOURCE the marine sector.' The inland fisheries sector SYSTEM is classified into two categories, capture and culture. By Regulation XXVII of the Permanent Capture fisheries refer to the harvesting of Settlement of 1793 in Bengal, the rights over fish and prawn populations in the open inland fisheries were included as a part of the rights water systems that are self-reproducing and over estates held by the zamindars or the self-sustaining under normal conditions. The landlords. The zamindars leased out these aquatic components of the capture fisheries rights over fisheries to farmers, moneylend- are the rivers and estuaries, the floodlands, ers, and occasionally to the fishers (Pokrant, the beels, and the Kaptai Lake. A heel is a Reeves, and McGuire 1997). The fishers had small lake or a deeper portion in a low-lying to pay tolls to the ijaradars or the leasehold- natural depression area that may or may not ers for access rights to the fisheries. Thus dry up in the dry season. most of the colonial period in Bengal is char- On the other hand, the growing of fish in acterized by de jure as well as de facto pri- confined bodies of water such as the ponds vate property rights over the inland fisheries and the oxbow lakes through aquaculture op- resource system. erations is termed culture fishery. The A significant part of the inland fisheries is aquatic components of the culture fisheries now divided into 13,003 water bodies called are ponds, baors, and coastal aquaculture jalkars or jabnahals (World Bank 1991). (shrimp-farms). Baors are oxbow lakes or Following the abolition of the zanzindari sys- dead rivers isolated from the rivers by a tem in 1950, these water bodies are now de change in their courses. Private property jure-owned by the state but property rights rights are predominant in the ponds and in over them are temporarily' transferred to the the shrimp-farms. They are also beyond the fishers mainly through a leasing process. purview of this paper. In this paper we will However, it has been observed that the fish- be mainly concerned with the rivers and estu- ers in general have failed to establish effec- aries, floodlands, heels, and baors. Together tive property rights over the water bodies. they covered 92 percent of the area under the Property rights over the water bodies are inland fisheries and contributed to 69 percent eventually transferred to a class of people of total inland fish caught in 1993 (DOF/ coming from outside the fishing community.' BFRSS 1995). Fish is a major source of protein from ani- mal source, particularly for the rural poor " Consumer preference for freshwater fish is very (GOB 1995a). The inland fisheries sector is strong. According to World Bank (1991), 75 percent of the total fish consumed in a year in Bangladesh comes endowed with more than 260 species of fish from freshwater sources. (A. Rahman 1989) and about 56 species of ' FAP 17 (1994), Toufique (1996), and Ullah (1985) prawn (Kibria 1983). It ranks fourth in the have found that the effective period of tenure is quite world, after China, India, and the former So- long due to various imperfections in the leasing mecha- nism. imperfections are described in Toufique viet Union (Agiiero 1989) in fish production. Such (1997a). a The fishery sector, as whole, accounts for 6 See Toufique (1997a) for a description of the trans- 3.5 percent of GDP, 10.4 percent of gross fer process. 74(3) Toufique: Inland Fisheries 411

This class of people is usually referred to as problem is how to allocate the yield. Thus, the lessees. A. Rahman (1987) calls them appropriation problem involves "agreement "white collared middlemen," or "water- on the level of appropriation, the method of lords." A field worker identified them as appropriation and the allocation of output" "local propertied class" (Anonymous 1991); (Ostrom, Gardner, and Walker 1994, 9). The Ullah (1985) and Chowdhury (1987) identi- focus is mainly on the flow aspect of the re- fied them as fish-merchants and moneylend- source system. Provision problems "concern ers. Following their colonial counterparts, the effects of various ways of assigning re- the lessees charge tolls on fishers for rights sponsibility for building, restoring, or main- to access to the water bodies under their con- taining the resource system over time" (Os- trol. However, we have observed in Sirajganj trom 1990, 47). The focus is mainly on the that the fishers were able to hold rights over stock aspect of the resource system. some water bodies which are of very poor Note that the appropriation and the provi- quality. Property rights in these water bodies sion problems are not so neatly divisible in are held by a class of wealthy fishers who every practical case.' The information (such own a substantial amount of fishing assets. as knowing the relationship between effort Like the lessees they also charge tolls on and yield) and the coordination requirements fishers. Following Toufique (1996) we will (such as negotiating and enforcing an agree- refer to them as the fishers-lessees.' ment among the appropriators to share the subtractable yield) for making such a divi- IV. THE NATURE OF sion are very demanding. The extent to EXTERNALITIES AND THE which such a clear division can be made de- INTERNALIZATION MECHANISM IN pends also, to a certain degree, on the nature THE INLAND FISHERIES OF of the resource system under investigation BANGLADESH and the associated property rights structure. In many situations, it is difficult to make this Common-pool resource (CPR) refers to sharp division because an action can well be "a natural or man-made resource system that classified as a provision as well as an appro- is sufficiently large as to make it costly (but priation problem. However, we will maintain not impossible) to exclude potential benefi- the distinction mainly on the ground that it ciaries from obtaining benefits from its use" provides a helpful guideline for separating (Ostrom 1990, 30). According to Gardner, the two important problems: that is, the ac- Ostrom, and Walker (1990, 335-36) com- tions that do not directly affect the stock (the mon-pool losses or dilemma describe a situa- appropriation externalities) and the actions tion where "individuals jointly using a com- that do (the provision externalities). We will mon-pool resource (CPR) are presumed to shortly take up this issue in the context of in- face a tragic situation in which their individ- land fisheries of Bangladesh. ual rationality leads to an outcome that is not There are three types of appropriation rational from the perspective of the group." problems: rent dissipation, assignment prob- Thus, as it will become clear in the follow- lems, and technological externalities. Rent ing, in the present context common-pool dissipation relates to the problem of "de- losses refer to the losses associated with a termining the efficient level and mix of input complete or partial failure of the agents to in- resources necessary for obtaining (a given) ternalize the relevant externalities associated flow" (Gardner, Ostrom, and Walker 1990, with the exploitation of a resource system. 341). Assignment problems deal with the According to Gardner, Ostrom, and timing and location of the appropriation pro- Walker (1990), appropriators working on a resource system face two types of problems: ' Such cases existed in colonial Bengal (Pokrant, appropriation problems and provision prob- Reeves, and McGuire 1997). Fishers' rights over some water bodies in the floodlands are also reported by lems. In appropriation problems the produc- Kremer (1994, 83). tion relationship between yield and the level 8 I am grateful to the anonymous referee for pointing of inputs is assumed to be given and the out this to me. 412 Land Economics August 1998 cess. Technological externalities explain the can be drawn from our fieldwork experience situation where the use of one technology in- and also from some secondary investigations. creases the costs for the users of other tech- In the context of the inland fisheries of nologies. Bangladesh the resource units can more ap- Provision problems have a supply and a propriately be viewed as less mobile and as demand aspect. The supply aspect focuses on more mobile. We will interpret mobility in the incentives of the appropriators to contrib- terms of the movement of fish and the conse- ute resources for the maintenance of a re- quent ability of the agents to control the bio- source system. The demand aspect of the mass of fish. More mobile therefore refers to provision problem involves regulating with- the situation where the agents have less con- drawal rates so that it does not adversely af- trol either on the movement of fish or on its fect the resource itself (Ostrom 1990, 49). biomass. Similarly, less mobile refers to the Recently, Schlager, Blomquist, and Tang situation where the agents have more control (1994) have argued that the attributes of mo- on the mobility of fish and hence on its bio- bile flows and storage should be added to the mass. already accepted characteristics of a CPR Note that Gardner, Ostrom, and Walker such as the difficulty of exclusion and the (1990) defined technological externalities as subtractability of yield. Mobility is "the spa- a space conflict between two types of gear tial movement of the units" and storage ca- (usually between a fixed and a mobile gear). pacity "enables the user to capture and retain But we have observed in Sirajganj that this unharvested units" (Schlager, Blomquist, problem occurs also for gear of the same and Tang 1994, 295). Examples of CPR with type. We will refer to the former as inter-gear mobile flows are demersal and pelagic fish- technological externalities and the latter as eries and an example of a CPR with storage intra-gear technological externalities. The attributes is a groundwater basin. Appropria- only way the lessee can stop common-pool tors of mobile resource units that do not have losses in a more mobile state of the resource the attribute of storage are likely to address system is through providing some sort of en- mainly the appropriation problems. On the forcement services. For example, in Kakian other hand, the appropriators of stationary re- and Baghabari water bodies in Shahjadpur source units that do have the attribute of stor- thana, fishing teams9 are allocated a certain age are likely to address both appropriation part of the river for a specified period. This and provision problems. This is so because is the only way of internalizing intra-gear the appropriators of the former are not sure technological externalities that has.been tra- they will reap the benefit of their effort ap- ditionally practiced by the fishers and strictly plied for the improvement of the resource enforced by the lessees. In the case of intra- system because the resource units are mobile. gear technical externalities, any conflict over It is costly to know the population dynamics space is resolved through the intervention of of mobile resource units. Also, the popula- the lessee as an arbitrator or as the ultimate tion dynamics of the resource units may be decider of who fishes where. The allocation affected by many factors and the appropria- of space is crucial because a seine net, the tors may not agree or even know the factors predominant gear used in the water bodies of that are most or least harmful to the long-run Sirajganj during the time of the fieldwork, sustainability of the resource system itself. takes a relatively large space for its opera- tion. Employment of a few seine nets can easily crowd a water body. More fishing Nature of Externalities in the Inland Fisheries teams operate in the richer water bodies and of Bangladesh as such the problem of congestion is acute there. In poorer water bodies, where the Obviously, the externalities in the inland fisheries of Bangladesh are extremely diverse 9 Commercial capture of fish is done by fishing- and complex. However, some useful insights teams guided by a team-leader. 74(3) Toufique: Inland Fisheries 413 fishers-lessees have control over the property traditionally paid in kind (fish), impounding of rights structure, it is mainly the fishers who the gear to the temporary or permanent disposses- enforce the rules of access among them- sion of fishing rights. (BCAS 1988, 24-25) selves. The task is relatively easy since the number of teams fishing in these water bod- In Sirajganj there are deeper segments of ies is usually small and stable. the river bed that are locally known as dohas. In the case of inter-gear technological ex- At the advent of the dry season, the water in ternalities a typical problem arises when a the elevated parts of the water bodies dry up mobile gear and a fixed gear target for the early. Fish aggregate in the dohas where same spot for fishing. The lessees ensure that branches of trees are initially placed as a such conflicts do not take place. In Bagha- sheltering device. This is commonly known bari and Kakian water bodies we have ob- as katha or brush-shelter fishing. During this served that mobile fishing gears are barred period the resource units are less mobile and from entering the area earmarked for fixed the owner of the kasha can affect the average gear installed for hilsa10 fishing. We have biomass of fish. Bamboo poles are erected also witnessed such arrangements in the wa- around the neighborhood of the katha to ter bodies such as Dhangara, Royhati-Bhuyi- mark the boundary. Guards are employed angathi, Ichamoti-Nalka, etc., in the thana of round the clock to control theft. No fishing is Raiganj with respect to the fixed gear called allowed for one to three months depending chatka. Property rights over these water bod- on the quality of the dohas and the time pref- ies are held by the fishers-lessees. erences of the contracting parties. When The lessees handle the problem of allocat- fishing in katha starts, it is done in a very ing fishing spots of heterogeneous quality simple way. The area of the katha is sur- among the owners of the fixed gears by rounded by a large fishing net. The branches charging differential tolls depending on the of trees are withdrawn and fish caught by us- quality of the spot. The higher the quality of ing various types of gears. When this is over, the spot, the higher is the toll. In the poor wa- the branches of trees are placed back and the ter bodies, sites for fixed lift nets (known as katha is again kept reserved for a certain pe- khora) are literally auctioned out to the inter- riod. Afterwards the same technique is used ested parties. Chatka sites are usually allo- again to catch fish. This is an example of cated through negotiation between the fishers control of effort during the period when the and the lessees. owner of katha has relatively greater influ- Fishers have reported to us that when an ence on the level of fish biomass. attractive fishing spot is identified, a rotation In the river Jamuna the only situation system is followed by them which is strictly where the lessees have some control over fish enforced by the lessees. Allocation of space biomass is when fish get trapped in what is by the lessees is reported in other studies as locally known as a kol. A kol is formed by well. In a fisheries report from the Bangla- river erosions. The quality of a kol depends desh Centre for Advanced Studies, it is ob- on its size, depth of water, and fish habitat. served that, Water in a kol is relatively clean and the river current minimum. It is ideal refuge for fish Fishing units must be authorised by the Ijaradari moving against the river current. Once such bureaucracy to fish in the fishery and must not use an area is discovered by fish, they start to ac- fishing gear other than those that have been au- cumulate and show no signs of leaving. Fish thorised. Furthermore, they must limit their fish- are allowed to stay to gain biomass. The les- ing to designated areas within the fishery. In the see employs guards to protect the kol. When event that the fishing unit is found to operate a the fish gains enough biomass or when it gear which differs from the gear for which per- mission was given in the first place, the particular unit is instantaneously penalised. The retribution 10 Hilsa, an anadromous fish, migrates from the sea is equally swift in case of violation of any of other in the south of Bangladesh to the rivers upwards to the regulations. The penalty could range from a fine north for spawning. 414 Land Economics August 1998

shows signs of disappearing, fishing is Kremer (1994, 37) observes that fishery started. Fishing in kols is as straightforward access is either territorial or open access as those in kathas. Either a massive seine net when it is'monsoon (May-September)," the is dragged from one end of the kol to the water level is high (June-September), and other or the side of the kol attached to the the large carps" are dispersed (June-Septem- main river is closed by a series of strong ber). The fishery access is under the control fishing nets and the fish inside are caught by of the lessee when monsoon starts to recede any method available. A kol is like a natural (October-April), water level is falling and trap for fish in Jamuna. becomes lowest (October-April), and the Thus the agents do have some decision- large carps are in the depressions (Novem- making influence on the size of fish biomass ber-March). In both these periods small fish and the movement of fish. During the seine- are dispersed. Thus the former state of the re- fishing in the monsoon (more mobile state) source system comes closer to what we have the lessees can reduce common-pool losses described as the "more mobile" state and by enforcing the rules of any rotation system the latter state of the resource system resem- or through assignment of fishing spots or bles the "less mobile" state. The deepest through resolving any conflict between the parts of a water body are allocated for brush- operators of fixed and mobile gears. On the piles. The lessees authorize owners of such other hand, the lessees can reduce common- rights to prevent fishing by others (who do pool losses during the dry season katha and not have those rights) in their designated area kol fishing (less mobile state) by strictly en- until their operation has finished (Kremer forcing the property rights of the agents. 1994, 81). The extent of katha fishing in the Let us now turn to the floodlands that vary heels is described in a report from the Minis- substantially in physical characteristics. The try of Irrigation, Water Development and floodlands are composed of low-lying areas Flood Control: that remain under water for a considerable period. Within these low-lying areas there During the dry seasons of the lease period, brush are even lower and deeper regions such as shelters are placed and maintained in the deeper the beels. The beets are leased out by the portions of the beets to provide shelter for larger varieties and state. These beets remain under water during of fish-major carps, large cat fishes feather backs. The brush shelters are placed dur- the peak of the monsoon and most of the ing the first dry season of the lease, and they are floodlands become a de facto open access re- consolidated and maintained during the subse- source system. It is only during the process quent two dry seasons. The large varieties of fish of receding water that the agents are able to are harvested only once towards the end of the reclaim their rights-either on the beels lease period during the months of Poush, Magh, leased in or on the privately owned land that and early Falgun of the Bengali year (middle of temporarily turns to a fishing ground. It is ob- December to March or early April). In each dry served in FAP 17, the Flood Action Plan by season of the three years lease period, however, the Bangladesh Ministry of Water Resources, minor carps and small sized fish are harvested and Kremer (1994, 54) that the area under private pro -erty rights has been increasing " As recognized by the above quote from FAP 17 over time. To quote from FAP 17 (1994, 93): (1994), flooded agricultural land is in principle an open access fishery. However, territorial rights over these In areas not under lease, which, when flooded, are fisheries are often privatized either by the owner of the theoretically open access for fishing, the growing land that lies at the bottom of the (flood) water or by awareness of the value of the fisheries resource the fishers living in the neighborhood of the land. This is encouraging landowners on the floodplain and type of access rights is referred to by Kremer (1994) as along the banks of khal [a canal, translation mine] "territorial." On the other hand, in some cases no such rights are informally established either by the relevant to establish claim to fish. the Through excavation landlord or by the fishers. This type of access rights is of fish pits (kua) and the placement of brush-piles referred to as open. See Kremer (1994, 71-72) for de- (katha) which aggregate the fish as the flood water tail. recede, landowners are able to extract a sizeable 'Z Carps are expensive in Bangladesh and are gener- portion of the benefits of the fishery. ally beyond the purchasing power of the poor. 74(3) Toufique: hiland Fisheries 415 from portions of the beel not covered by brush units within a physical environment (e.g., shelter. (MPO undated, 32) across the water bodies within a river) as well as across different types of physical Where the tract of water can support fish- environments (e.g., from the rivers to the trapping with barricades, the placement lines floodlands). We cannot expect the agents to are known by name and allocated by the les- restock (or not to use a destructive gear in) see or sub-lessee specifically to a group of an open water body because, as Kremer fishers (Kremer 1994, 81). Kremer distin- emphasizes in the above quote, the benefits guishes between territorial (fixed) technique of restocking (or not using a destructive (such as the brush-piles) and mobile tech- gear) will be internalized by others. As nique (such as the seine, gill, and push net reported in a National Water Plan Project fishing). The right to fish is sold by area for (NWPP 1984, 3-36), it is only in the case territorial technique. There is generally no re- of culture fisheries such as the "oxbow striction on effort. As Kremer (1994, 82) lakes and pond fisheries, that some stocking rightly points out, "effort is obviously re- in the traditional manner is done by lease- stricted by the fact that sites can not be dou- holders." ble-booked." For mobile techniques, fishing These issues came out clearly when we rights are sold by "man and the boat" units asked 115 fishers to answer whether catch and "in this way the margin between total in the water bodies they were fishing in cost and the value of catch is maintained" was declining over the last five years. All (Kremer 1994, 82). of them responded in the affirmative. Then But Kremer (1994, 83) also makes the fol- the respondents were asked to mention the lowing observation: factors they thought were responsible for the decline in catch. The fishers came up However, the owners of fishing rights make no with eight causes which we have classified apparent attempt to maintain the growth and re- as appropriation and provision problems. production of the fish stock. Leaseholders and Provision problems have been further clas- sub-leaseholders are not known to ban monofil- sified as internal (i.e., those that can be ament or narrow-mesh gill nets during the flood addressed by the agents con- season, to establish reserves or to maintain a potentially breeding stock for the following season. Because cerned) and external (i.e., those that can- of the areas of the jalmohols and sub-lessees are not be potentially addressed by the agents small and the fish are mobile between them, espe- concerned). The responses are presented in cially during the flood-season, the benefits of Figure 1. stock maintenance would be largely externalised. Only 16 percent of responses relate falling catch to the appropriation problem of rent Thus the appropriators in the floodlands dissipation (economic overfishing). Though appear to have failed to solve the provision economic overfishing may not affect fish problems mentioned in the quote above. But stock, the fishers think that catch is declining we need to look at the root cause of this because there are "too many fishers." failure: is it due to the inability of the agents Eighty-four percent of the responses relate to prohibit the use of the destructive gear falling catch to provision problems. or to establish the reserves and so on that Where internal factors are concerned, leads to provision failure or the nature of some of the fishers thought they had failed the resource system itself is such that provi- to prevent the operation of the current jal. A sioning does not pay. The externalities that current jal is a gill net with fine mesh. This the agents have attempted to internalize in particular gear is banned because it is the floodlands are related mainly to the thought to be detrimental to the growth of physical characteristics of the resource sys- fish stock. The fine mesh gills fry and finger- tem-they are appropriation externalities. In lings. There exists no work on the effects of the inland fisheries of Bangladesh internal- current jal on the stock of fish. But it is gen- ization of provision externalities is con- erally agreed that the small fish caught by strained by the mobility of the resource this gear also include those that have a fair 416 Land Economics August 1998

- Economic Overfishing** t APPROPRIATION PROBLEM (16%) t PROBLEMS FACED BY THE APPROPRIATORS*

PROVISION PROBLEM (84%)

INTERNAL FACTORS (25%) EXTERNAL FACTORS (59%) 1 l

- current jai (14%)*** - inadequate flood (39%) - spawn fishing (11)**** - use of chemicals in farming (9%) - ulcerative disease in fish (8%) - declining fish stock (2%) - fishing by dewatering closed water bodies (1 %)

FIGURE I PROBLEMS FACED BY THE APPROPRIATORS IN THE INLAND FISHERIES *Percentages refer to sample responses. **The typical way of expressing overfishing by the fishers is the response "too many fishers." ***Current jal is a gill net with fine mesh. This particular gear is banned because it is detrimental to the growth of fish stock. The fine mesh gills fry and fingerlings. ****Spawn fishing is the capture of 5-7 mm carps approximately 5 days old. chance to spawn." For example, Alam, Ali, problem as a provision problem. It should and Tsai (1997, 143) mention that the current also be recalled that the appropriation prob- jal has the "potential capability for capturing lem associated with fishing gears, including juveniles of large fish like rui, catla, mrigal the current jal, has been generally tackled and kalbasu . . . [and] with small meshes well by the fishers. This also justifies our cat- seems to damage fish resources of flood- egorization of this response as a provision plain." The Fish Act of 1950 (and its subse- problem. But again, the important thing to quent amendments) bans use of this gear on notice is that this is a problem that has to do the grounds that "they are invisible in the more with the nature of the resource system water and entangle everything that touches rather than with a collective action failure them" (Kremer 1994, 13). Note that the Fish problem from the part of the fishers. In fact, Act of 1950 prohibits the use of current jal there is a boundary problem in the sense that but not its production or sale. For the Fish fish are not staying in one place because the Act of 1950 to be effective there should be resource itself is characterized by a flow sys- either be a prohibitive tax on the sale or pro- tem. Note that similar explanation is pro- duction of this gear or a total ban on its sale vided by Kremer (1994) with respect to the or production. floodplain fisheries in his previous quotation. We need to take a closer look at the nature Some of the fishers also thought they had of the current jal problem for it could as well represent an appropriation problem." But the "Personal discussion with those (including fishery corresponds to the response of the fishers biologists) who are knowledgeable on this issue. current jai's impact on the stock of fish, " I am indebted to the referee for drawing my atten- which rationalizes our consideration of this tion to this problem. 74(3) Toufique: Inland Fisheries 417 failed to prevent spawn fishing. Twenty-five the stock of major carps has declined for percent of the responses attribute falling many reasons including overfishing. Melvin catch to internal factors over which they have (1984) reports of a large expansion in effort limited control due to the flow nature of the in the hilsa fishery with only a marginal gain resource system. in catch. This may be indicative of biological Fifty-nine percent of all responses hold overfishing, but no conclusive statement can external factors responsible for the falling be made on the size of the fish stock. The catch. Note that the response of declining fish marginal gain in catch may be due to other stock refers to the fishers' view that exces- factors not considered. On the other hand the sive application of fishing effort by them in the Bay of Bengal Programme (BOBP 1987) water bodies where they fish had already re- study found no evidence of overfishing of sulted in a decline in the fish stock. The most hilsa. Similarly, the Marine Resources As- important external provision factor men- sessment Group (MRAG 1994), based on tioned by the respondents is the lack of flood- data collected from Hail Haor, has found that ing (about 39 percent of all responses). These it is not only the current catch rates that are findings, therefore, support the hypothesis sustainable but also the current level of effort developed by Schlager, Blomquist, and Tang that could also be sustained. A similar find- (1994) that the appropriators are less likely ing is also reported in FAP 17 (1994). Thus to address provision problems in fisheries. there is no conclusive evidence on biological overfishing. But we have to take this conclu- Evidence on Rent and Overexploitation in sion with some reservation because these Inland Fisheries studies are not based on long-term data. For example, the MRAG study covers a period of A relative success in internalizing exter- less than two years and while FAP 17 (1994) nalities by the agents would imply that (i) a found a reduction in biodiversity, the long- large amount of rent should exist in the in- run implications could not be taken into con- land fisheries and (ii) the resource system is sideration because of the absence of longitu- not overexploited. In this section we will pro- dinal data. vide evidence on these issues. As presented in Table 1, Ahmed (1991), A Further Analysis of the Nature of Khaled (1985), S. Rahman (1989), and Tou- Property Rights fique (1996) have found that a significant amount of rent exists in the inland fisheries The nesting of common-property rights of Bangladesh. The figures vary from 11-56 within private property rights and their coex- percent of the gross value of the catch. It istence as a possibility has already been rec- should be noted that these rents mainly ac- ognized in the literature. For example, Runge crue to the fishers. states that "institutions may play a key role The evidence regarding biological over- in the effective management of scarce natural fishingt5 is somewhat inadequate and incon- resources, complementing and combining clusive. Tsai and Ali (1985) have found that with private rights" (1986, 624; emphasis mine). Similarly, van den Brink, Bromley, and Chavas recognize that "property rights TABLE 1 theory suggests that a pluralistic property RENT IN THE INLAND FISHERIES OF rights regime under which both private and BANGLADESH common property regimes could coexist and Rent as a co-evolve would be efficient in capturing the Sample Percentage Study Size of Gross Catch

13 Wilson (1982, 421) defines biological overfishing Ahmed (1991) 415 24 as a situation where "the effects of exploitation of cur- S. Rahman (1989) 199 56 rent stocks causes a decline in the size of future stocks Khaled (1985) 81 11 possibly leading to depletion and, in some special in- Toufique (1996) 60 40 stances, to the extinction of the species." 418 Land Economics August 1998 economic value of different adaptive strate- compete for such posts. On the other hand gies" (1995, 387). One reason why this may the resource system is not remote; it virtually be the case is that the appropriators may have passes by the doorsteps of the rural house- detailed and accurate information about the holds. The fishers usually live in the neigh- resource system and about the appropriation borhood of the water bodies where the les- technology since they have been using it year sees also reside. In fact, a formal lease- after year. These rules can either be main- contract explicitly stipulates that with respect tained or modified by an explicit or implicit to access rights preference should be given agreement between the appropriators and the to the fishers who live closest to the water external agent when the resource system is bodies. These factors also make the relation- privatized by the latter. Larson and Bromley ship between the two parties less anonymous. (1990, 240) make the following comment: We have already provided evidence to the fact that the fishers receive a significant if joint costs exist under common property ar- amount of rent (Table 1). Kremer (1994, 95) rangements, for example in agricultural or range calls this rent a "loyalty premium" and land, dividing the land into parcels controlled by admits that the lessee needs "physical and private property regimes will not necessarily re- political support of a number of follow- duce or eliminate the technical interdependence ers.... this `loyalty' is obtained by allowing that exists between (among) the users. his clients ... fishermen ... somewhat more than the market-clearing private opportunity The important thing to notice in the case costs of labor." Kremer (1994) compares of the inland fisheries in Bangladesh is that this "loyalty premium" with efficiency the rules to address common-pool losses are wage. Loyalty premium implies that fishers mostly developed by the appropriators them- will receive rent even when they may not be selves. In a sense these rules have some re- in short supply. Ullah (1985) and Toufique semblance with those usually observed in re- (1996) observed that the fishing teams spying source systems held in common. The role of on behalf of the lessees pay a lower toll. This the lessees can be evaluated as an enforcer is another expression for a loyalty premium. of these rules. It is not in the interest of the To quote from a report by BCAS (1988, 25): lessees-or for that matter in the interest of the fishers-to ignore these rules. On the in the pursuit of their self-interest, the ijaradars contrary, it is in their interest to enforce and have managed the fisheries of Bangladesh for strengthen these rules because their income nearly two centuries. Fishermen have, therefore, is determined by the productivity of the wa- come to depend on the ijaradar for management ter body as well as by the amount of rent gen- of the fishery which includes not only rent collec- tion, but also the granting fishing rights, the de- erated by the water body. of marcation of segments, prevention of incursion by We have observed in Sirajganj that the re- outsiders, arbitration of conflicts and to a much lationship between the lessees and the fishers lesser extent financial assistance, though in the is also not anonymous (Toufique 1996). It form of usurious loans. Through the performance can be more appropriately described as of pa- of these functions, the ijaradars have emerged as tron-client type. The fishers look at the les- patrons of the fishing community and as produc- sees as a provider of fishing rights. The les- tion leaders. This role of the ijaradar cannot be sees also get benefit from the fishers when underestimated. they need support in rural politics. Such sup- port may take the form of votes in local elec- V. MAJOR PUBLIC POLICY ISSUES tions, provision of manpower during border disputes with neighboring lessees, and so on. In this section we will discuss two major In Sirajganj, most of the lessees have been public policy issues that follow from our found to be the past or present members or analysis. First, there is an equity problem in chairmen of the local level institutions of the the inland fisheries of Bangladesh (Ahmed state. In the rural context it is the most afflu- 1997; Ahmed, Capistrano, and Hossain ent and powerful people who, in general, 1992; Capistrano, Hossain, and Ahmed 74(3) Toufique: Inland Fisheries 419

1996; Toufique 1997a). The state has been VI. CONCLUSION attempting to get more of the benefits to flow directly to the fishers mainly by transferring The creation of property rights in the in- property rights over water bodies to them. As land fisheries of Bangladesh has presented us mentioned above, effective property rights with an opportunity to investigate in a field- are eventually transferred to the lessees or level setting whether creation of such rights ijaradars who do not come from the commu- helped to minimize common-pool losses. By nity of fishers. Though this is the right direc- analyzing the physical characteristics of the tion or indeed the first step towards making resource system and the existing nature of an improvement in the welfare of the fishers, property rights we have come to the conclu- the mere granting of property rights is not sion that the appropriators have been suc- enough. One has to empower the fishers so cessful in internalizing appropriation exter- that they can permanently retain their rights nalities such as rent dissipation, assignment over the water bodies. In the longer term this problems, and technological externalities. requires gradual breaking up of the power There exists a significant amount of rent and structures and the patron-client type relation- there is no conclusive evidence on biological ships that define the property rights institu- overfishing in the inland fisheries of Bangla- tions in the inland fisheries of Bangladesh. In desh. However, the agents have failed to in- the short run, and as an ongoing process, the ternalize provision problems due more to the state should try to enforce the already ex- flow nature of the resource system than to isting rules and laws with respect to property any serious collective action failures. We rights over the water bodies and make neces- have also observed that the relationship be- sary changes whenever required." tween the lessees-the holder of property The second major problem relates to the rights over the water bodies-and the fishers Flood Action Plan (FAP 17) and the associ- is not anonymous. It is rather of patron-client ated engineering works that have created a type. In the inland fisheries of Bangladesh monster problem of habitat fragmentation, private property rights have nested in and playing havoc with the life cycle of fish. This maintained many rules historically devel- is reflected in the opinion of the fishers la- oped by the fishers. Such outcome is not ob- beled as "inadequate flood" in Figure 1. In- vious from the analysis of those who think adequate flood, which is the direct effect of that private property rights are the only solu- the engineering works designed to control tion to common-pool problems. water, has been ranked as the number one problem associated with falling catch as per- ceived by the fishers. The adverse impact of References flood control structures on fish is relatively well documented in the literature." This Agiiero, M. 1989. "Inland Water Fisheries of prompted the government to rely more on Bangladesh: Management Options and Na- culture-based fisheries as a compensatory tional Interventions." In Inland Fisheries policy. But, unfortunately, the rural poor Management in Bangladesh, ed. M. Aguero, S. have very limited access to culture-based Huq, A. Rahman, and M. Ahmed. Dhaka: De- partment of Fisheries, Bangladesh Centre for fisheries (Ahmed 1997). To quote from Advanced Studies and International Centre for Ahmed: Living Aquatic Resources Management. Ahmed, M. 1991. A Model to Determine Benefits Obtainable from the Management of Riverine Floodplains are the last remaining common prop- Fisheries of Bangladesh. Manila: International erty which provides the needed protein and in- come supplement for the millions of poor fami- lies. The impacts of culture based fishery 16 These issues are discussed in detail in Toufique development as well as flood protection and em- (1997b). bankment projects must be viewed in terms of "See Asaduzzaman and Toufique (1997) and the how they affect the millions of poor households. references cited there for more information on this (1997, 96) issue. 420 Land Economics August 1998

Centre for Living Aquatic Resources Manage- Gardner, R., E. Ostrom, and J. Walker 1990. ment. "The Nature of Common-Pool Resource Prob- 1997. "Socioeconomic and Policy Issues lems," Rationality and Society 2 (3):335-58. in the Floodplain Fisheries of Bangladesh." In Government of Bangladesh (GOB). 1995a. Re- Openwater Fisheries of Bangladesh, eds. C. port on the Household Expenditure Survey. Tsai and Y. Ali. Dhaka: University Press Lim- Dhaka: Bangladesh Bureau of Statistics. ited. 1995b. 1994 Statistical Yearbook of Ban- Ahmed, M., D. Capistrano, and M. Hossain 1992. gladesh. Dhaka: Bangladesh Bureau of Statis- "Redirecting Benefits to Genuine Fishers: tics. Bangladesh's New Fisheries Management Pol- Khaled, M. 1985. "Production Technology of the icy." NAGA 15 (4):31-4. Riverine Fisheries in Bangladesh." In Small- Alam, S., Y. Ali, and C. Tsai. 1997. "Fishing Scale Fisheries in Asia: Socioeconomic Analy- Gears of Floodplain Fisheries in Bangladesh: sis and Policy, ed. T. Panayotou. Ottawa: In- A Case Study of Chanda, BKSB and Halti ternational Development Research Centre. Beels." In Open water Fisheries of Bangla- Kibria, G. 1983. Shrimp Resources and Shrimp desh, eds. C. Tsai and Y. Ali. Dhaka: Univer- Culture. Khulna: Jahan Printing Press. sity Press Limited. Kremer, A. 1994. "The Generation and Alloca- Anonymous. 1991. "Field Report." Dhaka: Pro- tion of Incomes in the Hail Haor Fishery." In- shikha. terim Report to ODA. Asaduzzaman, M., and K. Toufique 1997. "Rice Larson, B., and D. Bromley 1990. "Property and Fish: Environmental Dilemmas of Devel- Rights, Externalities and Resource Degrada- opment in Bangladesh." In Growth or Stagna- tion: Locating the Tragedy." Journal of De- tion? A Review of Bangladesh's Development velopment Economics 33 (2):35-62. 1996, ed. R. Sobhan. Dhaka: Centre for Policy Melvin, G. 1984. Investigation on the Hilsa Fish- Dialogue and University Press limited. ery of Bangladesh. Rome: Food and Agricul- Bangladesh Centre for Advanced Studies ture Organization. (BCAS). 1988. Third Quarterly Report Exper- Master Plan Organization (MPO). No date. imental Project for New and Improved Man- "Open Water Capture Fishery Resources." agement of Open Water Fisheries in Bangla- Technical Report No. 16, MPO, Ministry of Ir- desh. Dhaka: BCAS. rigation, Water Development and Flood Con- Bay of Bengal Programme (BOBP). 1987. Hilsa trol, Dhaka. Investigations in Bangladesh. FAO/UNDP MRAG. 1994. "Poverty, Equity and Sustainabil- Report 36. Colombo: BOBP. ity in the Management of Inland Capture Fish- Capistrano, A., M. Hossain, and M. Ahmed. eries in South and South-East Asia. Biological 1996. "Poverty Alleviation, Empowerment Assessment of the Fisheries." Bath: Centre for and Sustainable Resource Use: Experiences in Development Studies, University of Bath. Inland Fisheries Management in Bangladesh." National Water Plan Project, Ministry of Irriga- In Environmental Sustainablility: Practical tion (NWPP). 1984. "Water Development and Global Applications, ed. F. Smith. Delray Flood Control." Second Interim Report,.Vol. Beach, FL: St. Lucie Press. 7, Fishery. Chowdhury, A. 1987. "A Review of Problems of Ostrom, E. 1990. Governing the Commons:The Fishing Communities in Bangladesh." Paper Evolution of Institutions for Collective Ac- presented at the BARC-CIRDAP Workshop tions. New York: Cambridge University Press. on Socio-Economic Aspect of Fishing Com- Ostrom, E., R. Gardner, and J. Walker. 1994. munity in Bangladesh and Its Role in Rural Rules, Games and Common Pool Resources. Development, Dhaka. Ann Arbor: University of Michigan Press. Department of Fisheries (DOF). 1990. Fish Catch Planning Commission. 1978. The Two-Year Plan: Statistics of Bangladesh. 1987-1988. Dhaka: 1978-80. Dhaka: Government of People's Re- DOF. public of Bangladesh. Bangladesh Fisheries Resource Survey Pokrant, B., P. Reeves, and J. McGuire 1997. System (DOF/BFRSS). 1995. Fish Catch Sta- "Riparian Rights and the Organisation of tistics of Bangladesh 1992-93. Dhaka: DOF/ Work and Market Relations among the Inland BFRSS. Fishers of Colonial Bengal." In Openwater Flood Action Plan (FAP 17). 1994. Flood Plan Fisheries of Bangladesh, eds. C. Tsai and Y. Coordination Organization (FPCO), Final Re- Ali. Dhaka: University Press Limited. port. Main Volume. Dhaka: Ministry of Water Rahman, A. 1987. "The New Fisheries Manage- Resources, Government of Bangladesh, FAP. ment Policy and Its Effects on Fishing Com- 74(3) Toufique: Inland Fisheries 421

munity." Paper presented at the BARC-CIR- 1997a. "Some Observations on Power DAP Workshop on Socio-Economic Aspect of and Property Rights in the Inland Fisheries of Fishing Community in Bangladesh and Its Bangladesh." World Development 25 (3): Role in Rural Development, Dhaka. 457-67. 1989. "The New Management Policy of 1997b. "Property Rights and Power Open-Water Fisheries in Bangladesh Under Structure in Inland Fisheries in Bangladesh." Experimental Monitoring and Evaluation." In Paper presented at the National Workshop on Inland Fisheries Management in Bangladesh, Policy for Sustainable Inland Fisheries Man- eds. M. Aguero, S. Huq, A. Rahman, and M. agement organised by ICLARM/DANIDA Ahmed. Dhaka: Department of Fisheries, Ban- and held on 23-24 March, 1997, Dhaka, Ban- gladesh Centre for Advanced Studies and In- gladesh. ternational Centre for Living Aquatic Re- Tsai, C., and Y. Ali. 1985. "Open Water Fisher- sources Management. ies (Carp) Management in Bangladesh." Fish- Rahman, S. 1989. "Fishing Activity and Distri- eries Information Bulletin 2 (4):1-51. bution of Benefits in Bangladesh." In Inland Ullah, M. 1985. "Fishing Rights, Production Re- Fisheries Management in Bangladesh, eds. M. lations, and Profitability: A Case Study of Ja- Aguero, S. Huq, A. Rahman, and M. Ahmed. muna Fishermen in Bangladesh." In Small- Dhaka: Department of Fisheries, Bangladesh Scale Fisheries in Asia: Socioeconomic Analy- Centre for Advanced Studies and International sis and Policy, ed. T. Panayotou. Ottawa: In- Centre for Living Aquatic Resources Manage- ternational Development Research Centre. ment. van den Brink, R., D. Bromley and J.-P. Chavas. Runge, C. 1986. "Common Property and Collec- 1995. "The Economics of Cain and Abel: tive Action in Economic Development." Agro-Pastoral Property Rights in the Sahel." World Development 14 (5):623-35. The Journal of Development Studies 31 (3): Schlager, E., W. Blomquist, and S. Tang. 1994. 373-99. "Mobile Flows, Storage, and Self-Organized Wilson, J. 1982. "The Economical Management Institutions for Governing Common-Pool Re- of Multispecies Fisheries." Land Economics sources." Land Economics 70 (Aug.):294- 58 (Nov.):417-34. 317. World Bank. 1991. Fishery Sector Review of Ban- Toufique, K. 1996. "Property Rights and the Is- gladesh. Washington, DC: The World Bank. sue of Power: The Case of Inland Fisheries in Bangladesh." Ph.D. diss., University of Cam- bridge.

People n Participation in Sustainable Development

Understanding the Dynamics ()I Natural Resource Systems Organizational Structure, Performance, and Participation: Forest User Groups in the Nepal Hills

Neeraj N. Joshi, Nazaruddin Mohd. Jali, and Abdul Halin Hamid

Introduction

Organizational structure is an important factor affecting both the participation of members of an organization and performance of the organization itself. There remains a considerable debate that concerns the particular aspects of organizational structure and how they affect participation and performance. The growing body of literature on organizational structure demonstrates the differential effect of variables such as mode of formalization, type of structures chosen for decision making, and size of an organization. The debate about the impact of these variables is important in the realm of the sociology of organization. Empirical studies are needed to examine the effect of these variables in diverse types of organizations including those related to the governance and management of forest resources. This debate also has substantial practical relevance. Recent forest policy of Nepal (e.g., the Forest Act of 1993) has emphasized the importance of creating forest user groups as local organizations that are expected to be able to govern and manage local forest resources effectively. Substantial effort is being devoted to the creation of new forest user groups. Questions that concern the mode of formalization, executive structure, and size of groups are raised as both forest users and forest officials face questions about the organization of these groups. In this paper, we first discuss some relevant theoretical underpinnings and pose questions concerning the effects of three types of structural elements of organizations. Second, we briefly describe recent efforts in Nepal to organize forest user groups. Third, we describe the region of Nepal where the field study was conducted, and the methodological steps employed in the study. Fourth, we present data about the effect of organizational structure on participation and performance. The paper concludes with remarks on the relevance of the study's findings for he theoretical understanding of organizational structure and the further establishment of forest user groups in Nepal.

Theoretical Underpinnings and Questions

The structure of organizations is dynamic. Organizations change in size, adopt new technologies, face changing environments and cultures, adopt new strategies or refine old ones, and adjust to other organizations in their field. Much has been written about the importance of voluntary local organizations and the relationship of their structural elements with performance. One such type of local organization, the forest user group, is the subject of this study. In many situations, a local organization is needed so that it acts as a channel through which local people can participate in the development and implementation of rural development programs. In a study, Korten (1983) noted the lack of an appropriate local organization within the community as one of the obstacles to generating people's participation in development programs. It is also where institutions are reckoned to affect the participatory process that local organizations are considered to be the most powerful and progressive forces. This is because local organizations act as intermediaries between the bureaucracy of a nation state and the local people in development tasks. The studies of Esman and Uphoff (1984) clearly point out local organizations as a necessary condition for accelerated rural development, and that such organizations must be seen in terms of a system of institutions performing

55 DIVERSE FORMS OF PARTICIPATION various functions in the rural sector of a particular country. There is a considerable debate in the literature about what kinds of local organizations are more effective. This is often because the strength of local organizations is affected by the appropriateness of their structural elements. The discussion in the following section follows the views of different authors with regard to three structural elements: (1) mode of formalness, (2) decision-making structure, and (3) size of a local organization.

Mode of Formalness

With regard to mode of formalness, there are two schools of thought. One group of scholars argues that more formal organizations are more effective, and the other argues that they are less effective. Those who argue in favor of more formalized organizations are: Dore (1971), cited in Esman and Uphoff (1984), and Sherraden (1991). They recognize the following propositions as positive outcomes of an organization being more formal: (1) a formal structure of community participation increases villagers' involvement in local programming, and (2) less formal organizations such as those based on traditional norms and power structure, will be either unable to perform more modern development tasks or unwilling to direct efforts in ways that will help the less advantaged sector of the community. On the other hand, Hunter (1976) and Esman and Uphoff (1984) show considerable support for more informal modes of organization in matters involving rural people in development activities. In the context of Nepal, Gilmour and Fisher (1991) point to both formal and informal forest management systems. They indicate that it is quite common to find highly visible formal systems that are totally ineffective in carrying out any sort of forest management. The implication of the above discussion is that it is important to ask what type of forest user groups (FUGs) will be more effective under the present forest management policy of the government of Nepal. Over the years, there have been both formal as well as informal systems or arrangements applied to forest management. This study seeks to determine whether or not an FUG should be a formal organization, as perceived by the forest users themselves. This is because the FUGs operating in the study area can be seen as being on a continuum going from less formal to more formal types of organization. The basis for calling them either less formal or more formal is: the extent to which the FUGs studied were under the administrative jurisdiction of the DFO. For the purpose of this study, the FUGs that were registered with the DFO were treated as `more formal' and the ones that were not yet registered as `less formal'. It is pertinent to mention here that for an FUG to get registered with the DFO, it must have an `operational plan', prepared by the FUGs themselves.

Decision-making Structure

There are two opposing schools of thought regarding decision-making structure. One of these asserts that local organizations need strong executive leadership to deal decisively with their environment, while the other relies more on maximum consensus and widespread participation of individuals in decision making. Although the more democratic approach in decision-making process is the one in which the whole assembly is involved, success on the part of an organization, in terms of decision-making structure, is also an important aspect. If a decision is important for the organization, a nonparticipative style is likely to be appropriate; if a decision is important for the subordinates in terms of their work, a more participative approach would be taken (Heller 1973, cited in Hall 1987). It is not the same decision-making structure that is effective in all the situations. From a study of rural local organizations in Botswana, Willett (1981), cited in Esman and Uphoff (1984), concludes that it would be better to vest responsibility in one person in a committee. But, a relatively more democratic and egalitarian approach than this has been reported by Kneerim (1980), cited in Esman and Uphoff

56 Organizational Structure, Performance, and Participation

(1984). Kneerim found a combination of committees and assembly as the most successful decision- making structure in the Mraru Women's Organization in Kenya. However, Esman and Uphoff (1984) assert that, purely executive form of decision making is unsuccessful. They associate assembly forms of decision making with effective performance of specific tasks, indicative of better overall performance by those local organizations who have more participatory arrangements. The core of the decision-making structure in local organizations, like the forest user groups, is not only the decision, but also its consequences for the organization. More specifically, the consequences of decision making have bearing on the organization's effectiveness, which in turn affects the interest of its members to participate. Thus, one concern of this paper is to discover what type of decision-making structure makes the FUGs more effective in the tasks of forest management.

Size of Organization

Like in the case of structural elements discussed above, there also exist two different lines of argument as regards the size of a local organization. Many scholars (Krech and Ballacy 1962, cited in Douglas 1976; Douglas 1976; Hunter 1976; Doherty and Jodha 1979; Golladay 1983, cited in Esman and Uphoff 1984) contend that local organizations need to be small in size. It may be that the smaller a local organization, the greater its effectiveness. The following are the points of argument in favor of a small size local organization: (1) domination is less likely in small groups; (2) conflict is less frequent; (3) with the increase in group size, members tend to become less satisfied, the gap between the most frequent participants and others tends to become a problem; (4) a group which is too large for its purpose and organization pushes certain members out to the periphery and denies them access to lines of communication and control, thus leading to a situation favorable to the formation of dissident subgroups; (5) group cohesiveness and member satisfaction tend to be greater; and (6) large groups inhibit participation of some members. In the context of the community forestry program in Nepal, the forest user group is a vital local organization, through which participation of local people is to occur. King et al. (1990) present the case of an FUG composed of 544 households, covering many villages and a wide range of caste groups, as not possessing any sense of "common identity" or "conununity" and lacking any organizational base for acting together. The foregoing discourse indicates that the effectiveness of a group is often influenced by its size. It also shows that there are advantages and disadvantages of both small as well as large groups. Given this dilemma, the obvious question that arises is: what is an appropriate group size for an FUG to act together more effectively?

Organization of Forest User Groups

With the advent of the Community Forest Development Program (CFDP) in Nepal in the 1970s, the concept of user group, better known as the forest user group (FUG) was introduced. Community forest is one category, aside from national and other categories of forests. Each community forest has a particular name, after which the FUGs are been named. The CFDP holds the FUGs responsible for the management of community forests to which they are formally affiliated. An FUG, as a local organization, acts as a body through which its members govern activities, organize themselves for participation, and, at the same, time, articulate their problems. According to the new Act (Forest Act 1993), a user group must be registered with the respective District Forest Office (DFO) for the former's recognition by the latter as a legal body to manage and utilize community forests.

57 DIVERSE FORMS OF PARTICIPATION

The group is conceived as a formal local organization, that lies outside the government bureaucratic organization, formed at the community level for the purpose of development, conservation, and utilization of forest resources. It is a corporate body organized largely on the basis of settlement of households. In the study area, the members of FUGs are the households who have since long been managing and utilizing the forest resources accessible to them. Hence, the membership in the user group is largely on the basis of residential proximity to a forest and historical affiliation to a particular forest. Once an FUG is formed, it is incumbent upon the group to manage and utilize the forests in accordance with the rules and regulations set by the group itself. However, according to the Forest Act of 1993, it is only upon the approval of the `operational plan' for the forest, developed by user groups in conjunction with field staff of the DFO, that the users' right to manage and utilize the forests is legitimized. An operational plan is a formal agreement between the users, the Forest Department, and the Village Development Committee VDC, which specifies all aspects related to the management of specified areas of community forest, including the regulation of utilization and distribution of benefits. An operational plan is legitimized by the endorsement of the representative of the FUGs and the DFO. It contains a description of the forest, the composition of the user group, and the rules under which the forest will be managed. The recent forest act also decrees that the community forests are to be managed with the active support and involvement of forest users. During the fieldwork for the study, the process of forming FUGs all over the district, as well as their registration with the DFO, was going on. At the same time, handing over of the forest land, by the DFO to the FUGs for management, was also going on. As of April 1993, a total of 190 FUGs from fifty VDCs had been registered with the DFO in Dhading (DFO/Dhading 1993). There were three distinct categories of FUGs seen in the studied villages. In the first category, were user groups which had been registered with the DFO and to whom the forest land had been handed over by the DFO. The second category of user group were those which had already been registered with the DFO, but were still waiting for the forest land to be handed over to them. The user groups in the third category were those which had neither been registered with the DFO nor had obtained management and utilization authority for forest land from the DFO. Of the 10 FUGs in the study, only 2 came under the first category mentioned above. The number of groups falling under the second category was 6, and the remaining 2 groups belonged to the third category. As mentioned earlier, the analysis in this study combines the first two categories and treats them together as "more formal"; and the third as the "less formal" category.

Some Fieldwork Details

Location of Fieldwork

The present study was undertaken in Dhading district which is located in the Bagmati Zone of the Central Development Region of Nepal. It contains fifty village development committees covering a total area of 192,478 hectares. Dhading is bordered by the districts of Rasuwa in the northeast, Nuwakot in the east, Kathmandu in the southeast, Makawanpur in the south, Chitwan in the southwest, and Gorkha in the west. To the north, lies the Tibet Region of the People's Republic of China. The lowest elevation in Dhading is found in the southwest (300m), while the highest is in the north (over 7000m). Within this range, the highest point the researcher reached was about 2200m, of Toplang Community Forest in Naubise. Of the total area in the district, 46 percent is covered with agricultural land, 48 percent with forest, and the remaining 6 percent of the land falls under other categories of land use (riverbeds, rocks, roads, settlements, and snow cover). Dhading has a total population of approximately 305,887 with 156,564 (51.18%) males and

58 Organizational Structure, Performance, and Participation

149,334 (48.82%) females. Of the total 49,438 households, the percentages of households owning more than one hectare, up to one hectare, and no land are 23.29, 74.17, and 2.54, respectively (HMGN/GTZ 1994).

Site Selection and Sampling Procedure

The field study was carried out from February through May 1993. The research site encompasses three VDCs, namely, Murali Bhanjyang, Naubise, and Nilkantha in Dhading district. The following were the criteria for selection of the site: (1) existence of Community Forest Development Program in the district; (2) the district chosen represented a typical hill region in Nepal, in terms of forest situation and socioeconomic background of the rural hill population; (3) sites chosen had local people involved, through FUGs, in the management activities of community forests; and (4) residents of the site were receptive to the research and the researcher. From the above selected VDCs (of Murali Bhanjyang, Naubise, and Nilkantha) the number of FUGs selected came to be 3, 4, and 3, respectively. These 10 FUGs were purposively selected for the following reasons: (1) to ensure desired proportion of respondents in the sample for the study, (2) to have all the three categories of FUGs described earlier, (3) to include multiethnic groups, and (4) to ensure sufficient response on the part of the FUG members to the researcher. At the time of the survey, there were a total of 190 FUGs already registered with the DFO from all the fifty VDCs of Dhading district. There also existed many other FUGs that were yet to be registered with the DFO. In a sense, the members of FUGs all over the district formed the population of the study. A list of FUGs was prepared for each selected VDC with the help of records provided by the DFO and by making preresearch visits to the respective VDCs. It is pertinent to mention here that the process of formation of FUGs in various parts of the district was ongoing at the time. After selecting the FUGs, the list of member households of each FUG was obtained from the chairperson or the executive members of the respective forest user committees (FUC). In all, there were a total of 570 member households under the FUGs selected for the study. All the member households belonging to the selected FUGs thus served as the sampling frame. Sample selection was done by adopting a proportionate random sampling technique in order to ensure equal proportion from all selected FUGs in the final sample. The total number of member households in selected FUGs ranged from a minimum of 23 to a maximum of 117. The final sample size comprised of 200 households. From the households so selected, an adult member who had been involved in CFDP, was chosen for the purpose of the interview. So, the household representatives to FUGs served as the unit of inquiry as well as of analysis for the purpose of the study. The main instrument used for gathering the information from the respondents was an interview schedule consisting of both close-ended as well as open-ended questions. In order to gather relevant information, the pretested interview schedule was administered to the respondents by adopting a personal interview technique. In addition, information was also collected by conducting impromptu interviews with local people-of whosoever that came across, whenever, and wherever felt convenient by the interviewer and the interviewees. Those being informally interviewed included old and young, men and women, people from high or low economic classes and castes, school teachers, village leaders, elders, and elites.

Effect of Organizational Structure on Group Performance

In this section, FUGs are analyzed in terms of organizational structural attributes that are likely to be favorable or unfavorable for performing group tasks. The information obtained is based on the perceptions of respondents.

59 DIVERSE FORMS OF PARTICIPATION

Mode of Formalness

Unlike the traditional forest management systems of the past, where there was no central authority involved, the current forest management system is a sponsored system in which the government is an outside agency that facilitates the setting up of a local organization of forest users. Under this management system, a forest user group would be entitled to legally manage and use forest products only after it has fulfilled all the official formalities outlined by the DFO. This means that a collection of local people is considered adequate to be registered as a formal organization with the DFO. The available literature, however, shows a controversy regarding the effectiveness of local organizations when formal or informal. Against this backdrop, the present study sought the respondents' opinion about the necessity of formalization of their user groups in order to function more effectively. It was found that an overwhelming majority (98.5%) of the respondents were in favor of having their forest user groups as formal organizations, although some of these FUGs were yet to gain the status of fully formal organizations. The FUGs included in the present study were classified into two distinct categories: `more formal' and `less formal'. This was conveniently done, based on whether or not the FUGs had been registered with the DFO. The FUGs which had already been registered with DFO were considered as `more formal'; in contrast to `less formal' ones, which had yet to be registered. For FUGs to get registered with the DFO, they need an `operational plan', which signifies the degree of formalness in terms of the characteristics of a formal organization. Viewed from this standard, it was found that of the 10 FUGs selected, 8 were more formal while only 2 of them came under the category of less formal. Regarding the effectiveness of FUGs, a great majority of the respondents were in favor of having formal organization. It is interesting to note that users who had managed forests for the past several decades without strong formal organizations had now come to opt for more formal organizations. The reasons stated by them, for this preference of a formal organization over an informal one, was that, by being formal, the user groups would get official recognition from the DFO. This is because, in the context of national law, the forest users get use rights only when the FUGs are formally recognized. Besides, a local organization would be in a better position formally organized in terms of clearly specified responsibilities to be borne by each member of the FUG (including FUC members). The local residents, having referred to their earlier experience, cited the example of the popular proverb "everybody's responsibility is nobody's" to stress the need for a more formal organization. It is also now emphasized in existing legislation, that FUGs move towards formal recognition.

Decision-making Structure

In any organizational situation, various types of decision-making structures generally exist. The type of structures in FUGs included `general assembly', `executive body', and `group leader'. The respondents were asked about their preferences concerning decision-making structures in their FUGs. Of the three structures, an overwhelming majority (95%) of the respondents had strongly opted for the `general assembly' kind over other kinds. The respondents believed that this system of decision making would make every member feel more accountable to that which had been decided by them all. There were only a small minority (5%) of respondents who preferred the FUC structure to be a decision-making body for their groups. Interestingly, none of the responses indicated that an authoritative decision-making structures was preferred. The perception of the respondents concerning the effectiveness of the decision making indicated that the general assembly kind of decision-making structure used to be the most influential one. There was a negligible portion of the respondents who perceived the executive body, i.e., FUC and the chairperson of the FUC, as the influential decision-making entities in matters related to imposing

60 Organizational Structure, Performance, and Participation sanctions, selection of group representatives for dialogue with government personnel, and implementation of group activities. The information obtained from informal interviews held with the local inhabitants also indicated that a few persons who were so-called influential in the local rural community, by virtue of either higher socioeconomic status or numerical strength, had played a dominant role in decision making. This type of situation was also observed in two forest-related meetings conducted by local people during the period of data collection. During those meetings, only a few influential persons were found to be more articulate and dominant than others in the discussion.

Group Size

As mentioned earlier, the size of user groups included in this study varied from a minimum of 23 to a maximum of 117 member households. In the study area (district), there were also other user groups which were not included in the sampling frame, but had memberships of more than 500 households. The present study focused on the size of FUGs and their relative effectiveness in forest management. It was noticed that irrespective of the group size that existed, only about 11 percent of the respondents felt their groups as small and, hence, opted for a size little bigger than their respective groups. Obviously, those who expressed their group as small came from the FUGs whose membership did not exceed 55. One reason stated for preferring a bigger group was that if they had more people, the group would be strengthened, in numerical terms, to protect and manage the forests. However, the number of members suggested by the respondents in any of the groups did not exceed more than 110 households. Interestingly, none of the respondents belonging to a group that consisted of 117 households considered their group as small. Instead, 17 percent of the respondents from this very category perceived their group as being too big for them to function effectively. The group size preferred was from 80 to 100 members, for three main reasons: (1) conflict would be less frequent in smaller groups, (2) more convenient in handling, and (3) effective communication among members. The group size preferred by the respondents is quite close to that of the size suggested by Doherty and Jodha (1979). According to these authors, a group of 100 farmers is suited to long-term maintenance and can enforce the rules necessary to keep up group action on their own. It is apparent that, despite a wide variation in terms of the number of members that make up a group, most respondents considered their respective group size as neither "too large nor too small." This could be due to the existence of greater per capita forest area (a ratio in which proportion of forest land is relatively higher than that of man). The other reason could be compatibility, in terms of working conditions in a group situation, among the members.

Effect of Organizational Structure on Participation

Participation has been operationalized in this study to mean the actual involvement of respondents in the decision making and implementation activities of community forestry. The extent of participation has been described in terms of three levels-high, medium, and low, given existing variations among the members of FUGs in participation. The level of participation was measured by developing suitable scales (see Joshi 1995). The scale for measuring the level of participation in decision making comprised nine activities in which the respondents were required to indicate their extent of involvement in terms of a four-point scale that ranged from `never' to `always'. Similarly, for measuring the level of participation in implementation, the respondents were asked to indicate whether or not they contributed labor and materials at preplanting, planting, and postplanting stages of afforestation programs. The level of participation was determined on the basis of the rule for normal distribution.

61 DIVERSE FORMS OF PARTICIPATION

Analysis of the extent of participation, with respect to mode of formalness of forest user groups, showed a greater proportion of respondents at a high level of participation from more formal groups as compared to the proportion of respondents from less formal groups. The Chi-square analysis supports the finding that respondents coming from more formal FUGs tended to differ from less formal FUGs in their level of participation in decision making as well as in implementation. This finding signifies that a, still greater extent of participation could be achieved provided that the process of registration of FUGs with the DFO is further accelerated. This implies a need for official recognition of the FUGs by the concerned agency. Correlational analysis was used in order to investigate the effect of group size on participation. The outcome of the analysis showed a significant inverse relationship between group size and participation in decision making (r = -.308, p = .01) and implementation (r = -.253, p = .01). These findings indicate that the extent of participation tended to be enhanced with the decrease in the size of the FUG.

Conclusion

Given the dynamic nature of an organization, the variation in the structural elements in those FUGs that lie outside the frame of an ideal FUG has demonstrated a varying degree of organizational performance. This indicates a need for adjustment in those elements so as to enhance FUG capabilities to act together more effectively in the management of forests. Such an adjustment would require that the FUGs be formed with appropriate elements such as decision-making structure, formalization, and size of organization. The foregoing discussion has illustrated the effects of structural elements on the users' perception of performance of FUGs, and on the extent of participation in the activities of community forestry program. The most preferred structural elements of an FUG as perceived by the forest users are: (1) general assembly kind of decision making, (2) a membership of 80 to 100, and (3) a formal mode of organization. The core of community forestry is people's participation. Efforts to enhance the level of participation would require favorable actions by all concerned. Forest user groups, as local organizations at the community level, could act as vehicles for forest development; and their effectiveness can be enhanced by making them more ideal in terms of structural elements as indicated by the findings of the study. The leaders of the influential few FUGs could encourage members to get involved in the decision making so that they can give voice to their needs and interests. The idea behind this is the concern for empowerment, whereby every member would have some say and thus enhance participation. The smaller groups may be combined, while the larger groups may be split for forming appropriate group sizes. By doing so, on one hand, the smaller groups that feel weak in terms of numerical strength will be fortified. On the other hand, the problems with larger groups such as ineffective communication among members and frequent conflicts, will also be lessened. The man-to- forest ratio could be on the same basis that is followed in splitting or combining groups. The forest user groups that have not yet registered themselves with the DFO need to prepare operational plans for obtaining formal status, and the concerned agencies need to provide assistance to FUGs to help in preparing operational plans and in other areas as and when needed. The timely service of DFO field staff will be a significant contribution to the formation of user groups and to the preparation of operational plans, while the forest users themselves play an active role in this endeavor.

62 Organizational Structure, Performance, and Participation

References

District Forest Office, Dhading. 1993. Official Record. Dhading: District Forest Office. Doherty, V.S. and N.S. Jodha. 1979. "Conditions for Group Action among Farmers." In Wong, J. (ed.) Group Farming in Asia.. Pp. 207-223. Singapore: Singapore University Press. Douglas, T. 1976. Group Practice. London: Tavistock Publications. Esman, J.R. and N.T. Uphoff. 1984. Local Organizations. Ithaca and London: Cornell University Press. Fisher, R.J. 1992. "Local Organizations in Community Forestry." In Veer, C. and J. Chamberlain (eds.) Local Organizations in Community Forestry Extension in Asia. Bangkok: Regional Wood Energy Development Programme in Asia, FAO. Gilmour, D.A. and R. J. Fisher. 1991. Villagers, Forests and Foresters - The Philosophy, Process and Practice of Community Forestry in Nepal. Kathmandu: Sahayogi Press. Hall, R.H. 1987. Organizations: Structures and Process. 3rd ed. New York: Prentice Hall Inc. HMGN/GTZ. 1994. Dhading District Village Profile 1994/94. Kathmandu: Dhading District Development Project (GTZ). Hunter, G. 1976. "Organizations and Institutions." In Hunter, G., A.H. Bunting and A. Bottrall (eds.) Policy and Practice in Rural Development. New York: Allanheld, Osmun and Co. and Universe Books. Joshi, N.N. 1995. Factors Influencing Participation of Members of Forest User Groups in Community Forestry in the Hills ofNepal. Ph. D. dissertation, Universiti Pertanian Malaysia. King, G.C., M. Hobley, and D.A. Gilmour. 1990. "Management of Forests for Local Use in the Hills of Nepal: Towards the Development of Participatory Forest Management." Journal of World Forest Resource Management. 5: 1-13. Korten, F.C. 1993. "Community Participation: A Management Perspective on Obstacles and Options." In Korten, D.C. and F.B. Alfanso (eds.) Bureaucracy and the Poor: Closing the Gap. Pp. 179-200. Connecticut: Kumarian Press. Sherraden, M.S. 1991. "Policy Impacts of Community Participation: Health Services in Rural Mexico." Human Organization, 50(3): 256-263.

63 The Symbolic Making of a Common Property Resource: History, Ecology and Locality in a Tank-irrigated Landscape in South India

David Mosse

ABSTRACT

Today there is a pervasive policy consensus in favour of 'community manage- ment' approaches to common property resources such as forests and water. This is endorsed and legitimized by theories of collective action which, this article argues, produce distinctively ahistorical and apolitical constructions of 'locality', and impose a narrow definition of resources and economic interest. Through an historical and ethnographic exploration of indigenous tank irrigation systems in Tamil Nadu, the article challenges the economic- institutional modelling of common property systems in terms of sets of rules and co-operative equilibrium outcomes internally sustained by a structure of incentives. The article argues for a more historically and politically grounded understanding of resources, rights and entitlements and, using Bourdieu's notion of 'symbolic capital', argues for a reconception of common property which recognizes symbolic as well as material interests and resources. Tamil tank systems are viewed not only as sources of irrigation water, but as forming part of a village 'public domain' through which social relations are articulated, reproduced and challenged. But the symbolic 'production of locality' to which water systems contribute is also shaped by local ecology. The paper examines the historical and cultural production of two distinctive 'cultural ecologies'. This serves to illustrate the fusion of ecology and social identity, place and person, in local conceptions,* and to challenge a currently influential thesis on the ecological-economic determinants of collective action. In short, develop- ment discourse and local actors are seen to have very different methods and purposes in the 'production of locality'. Finally, the article points to some practical implications of this for strategies of 'local institutional development' in irrigation.

INTRODUCTION

Indigenous locally controlled irrigation systems have for long been over- looked in favour of a focus on the role of the state in constructing, main- taining and managing irrigation. Indeed, there has been a deep rooted pessimism about the possibility of preserving such `commons' resources other than through centralized state control (Hardin, 1968). However, a current policy trend emphasizing community-based management heralds something of a reversal of this. In several sectors, including irrigation, local

Development and Change Vol. 28 (1997), 467-504. L) Institute of Social Studies 1997. Published by Blackwell Publishers Ltd, 108 Cowley Rd, Oxford OX4 IJF, UK. 468 David Mosse control is viewed as a preferable strategy for avoiding Hardin's `tragedy of the commons' - preferable, that is, to the options of state control or privatization (Blair, 1996). In support of this, attention has been drawn both to the widespread existence of effective decentralized farmer-managed irriga- tion systems, and to the benefits of encouraging local control through properly constituted `water users associations' and the transfer of operation and maintenance to them, even in large-scale systems. Several rather different concerns underlie the emerging policy consensus on `irrigation management transfer'. On the one hand, at a time when state- managed systems are seen as performing poorly or in fiscal crisis, community management offers opportunities for reducing subsidies, introducing cost recovery, enhancing farmer involvement in operations and maintenance, and in other ways transferring resources management to local users and so relieving the state of financially burdensome obligations (cf. Maloney and Raju, 1994; Meinzen-Dick et al., 1994; Sengupta, 1993; Svendsen and Gulati, 1994). On the other hand, in non-governmental and environmentalist circles, `resource management transfer' forms part of a discourse in which com- munity resource rights and control are asserted vis-a-vis the state as part of a critique of the modernizing development strategies of the centralized state and the dominance of `western' technical over indigenous community perspectives. Whatever the underlying agenda, the operational emphasis on `community management' often finds its justification in the existence of long term success- ful indigenous community managed irrigation systems. Indeed, analysis of these systems has been the principal means by which models for the explana- tion and promotion of collective action in resource management have been generated (e.g. Ostrom, 1990; Wade, 1988). The overall aim of this paper is to examine some of the limitations of these models or theories of collective action in accounting for the evolution, structure and operation of an indigen- ous farmer-managed irrigation system in south India. Conclusions on this will have implications for the development of strategies for the promotion of new or alternative community systems of irrigation management.

MODELS OF COMMUNITY RESOURCES MANAGEMENT

Theoretical explanations of communal action in common property resources management have appeared partly in answer to the widely-held pessimistic view that the actions of individually rational villagers would tend to militate against the evolution of co-operative solutions to the management of local commons (Olson, 1965; Popkin, 1979). There are two main schools of thought which have proposed an alternative view. The first (and most recent) draws on an institutional-economic analysis of local forms of co-operative action (successful and unsuccessful) to derive generalizable principles for farmer-managed irrigation (Ostrom, 1990, 1992; Tang, 1992). This analysis uses formal models derived from the theory of repeated games to challenge History, Ecology and Locality in Tank-irrigated South India 469 the dominant thesis on the unlikelihood of collective action among rational self-interested individuals. Focusing on costs and benefits to individual actors, incentives and penalties, institutional analysis demonstrates the eco- nomic rationality of co-operation and the possibility of co-operative equi- librium outcomes from competitive games (Ostrom et al., 1994; Sengupta, 1991). Moreover, institutional-economic analysis provides some answers to the question, `under what conditions will people co-operate?'. It claims to be able not only to show that farmers can co-operate in managing common resources, but also to predict the circumstances under which this is most likely to happen. Institutional-economic analysis therefore offers the possib- ility of the kind of predictive and generalizing theory of co-operative action which development agencies require in order to generate predicable out- comes from planned inputs. The second school emphasizes the force of tradition, social rights, value systems and moral codes in generating and preserving co-operative resources management to ensure, among other things, a minimum food security for community members. Co-operative solutions to common property use derive not so much from individual rationalism as from Scott's (1976) `moral economy', or from a `moral conscience' arising from the small community's need to cope with risk and its collective dependence on local resources, which is often institutionalized in religion, folklore and tradition (Guha, 1991: 29-34; cf. Blair, 1996: 485). Development interventions seek to harness these traditions, or to revive them where they are in decline (e.g. Gadgil and Guha, 1994: 39). Notions of community tradition have, in fact, long been part of the way in which external (often state) agencies conceive of local collective action. As I have argued elsewhere, in the case of tank irrigation in south India, the nineteenth century colonial government required the exist- ence of autonomous traditions of tank maintenance for the administration of tens of thousands of irrigation tanks. Where these traditions appeared no longer to exist they had to be re-invented, and where necessary (although futilely) upheld by the force of law (see Mosse, 1997a). These two schools of collective action arise from two long opposed traditions in social science. In the first, associated with Thomas Hobbes and Adam Smith, a person is first of all a rational self-interested individual (homo econornicus), while in the second, associated with Durkheim (among others), a person is firstly a social being (homo sociologicus) guided by social norms and only secondly an individual (Blair, 1996; Elster, 1989; see also discussion in Douglas, 1986). Despite these deep-rooted contrasts, within the discourse on common property resources development, both `rational choice' and `moral economy' schools construct rather similar images of community and indigenous collective action. Firstly, these images of community emphasize the autonomy and the independence of local resource use arrangements from state systems. Secondly, these models of community are synchronic, ahistorical and do not deal with change. On the one hand, rational choice models present local 470 David Mosse institutions as equilibrium outcomes of a structure of individual incentives. On the other hand the social and political forces acting on traditional systems are not seen as constituting, underpinning, reproducing or extending these systems; they simply erode, dissolve or undermine them (Mosse, 1997a). Community traditions do not change or evolve, they decay (Spencer, 1990: 140). Thirdly, these images of collective action institutions are often narrowly utilitarian and economistic, effecting a separation of resource management from other aspects of social life. Fourthly, planning models tend to homo- genize collective action institutions, disallowing location specificity. In short, models of community resource management generated by the policy emphasis on community resources management tend to construct a specific kind of ahistorical, apolitical 'locality' as a (perhaps necessary) basis for planned interventions. The analysis of community institutions of resources management offered here tries to move beyond the limitations of these models in several ways. Firstly, community institutions of resources use are not viewed as isolated from wider political structures of region and state. Secondly, the analysis emphasizes the specificity of local socio-cultural forms and their historical particularity. Apart from criticism of the naive methodological individualism of formal economic models, the sociological-historical perspective from which this paper draws challenges the ahistorical and apolitical view of local institutions as `steady state' systems. This perspective is especially important when it is realized that today's strong policy thrust towards community institutions is occurring precisely at a time when `traditional' institutional arrangements for sustainable resource use are reportedly in a state of collapse (Bromley and Cernea, 1989; Gadgil and Guha, 1992). In this light, resources are to be viewed as defined by changing ideas of property, rights, and entitlement. Thirdly, it is assumed that historically-specific structures of power, rather than simply calculated pay-offs (or traditional wisdom) underlie the norms and conventions of collective resource use, and account for the occurrence and persistence of local institutions of resource use (Bardhan, 1994: 174; cf. Douglas, 1986). These do not only manage resources, they also serve to reproduce relations of dominance and depend- ence, and provide the context for political strategy and status competition.'

1. The distinction between economic-institutional and sociological-historical explanations of local institutions is deliberately simplified here. Recent and sophisticated institutional models dispense with some of the more unrealistic assumptions about people as self- interested individual economic maximizers. Game theoretical modelling has begun to focus on the relationships between actors and the perception of these, and has begun to take account of ethics, norms, expectations and other forms of `assurance', 'trust' or 'habit' which influence or sustain co-operative solutions (Runge, 1986; Seabright, 1993). There is, moreover, a shift away from individual game scenarios. Ostrom, for example, refers to the need to complement individual-focused 'games theory' with a collective 'rules theory' to predict the structuring of games (Ostrom et al., 1994: 12). Also to the extent that models examine the influence of past actions, a historical perspective is introduced. History, Ecology and Locality in Tank-irrigated South India 471

While the burden of this article is to illustrate the way in which the models of collective action misrepresent common property resources systems as autonomous, internally sustained equilibrium and rule governed systems, there is another parallel argument (developed elsewhere; Mosse, 1997a) which looks at the specific reasons why these constructions have come to dominate policy discourse. The first and most' obvious fact is that policy science, including its theories of local collective action, is not value-free. Environmental and development discourse involves arguments which `elide empirical assertion and moral judgement' (Jacobs, 1994: 81). Following Foucault, it is possible to look at the particular interactions between knowledge and power which accord validity to economic explanations, equilibrium models, and certain ideas of the `community' to the exclusion of others (cf., Ferguson, 1994; Mosse, 1997a; Sachs, 1992). In brief, only certain models of common property resources management are easily compatible with centralized development planning systems. In particular it is those which meet requirements of generality, predictability, standardization and more generally have the capacity to fix and order the flux of social life so that it can be subject to planning and administrative control. More specifically, the newly theorized `community management' ideas stressing locally autonom- ous, internally sustained and self-reliant community institutions have emerged within a global discourse (policy and practice) oriented towards finding community solutions to the perceived problems of state and market- based irrigation management; solutions that are capable of addressing the policy imperatives of cost-sharing, recovery, and reducing the financial liability of the state. Such modelling of rural society is not recent but as old as centralized planning, administration or revenue collection itself. Indeed, I have elsewhere argued that in south India the identification, description and codification of autonomous village `traditions' of tank maintenance occurred precisely from the point in the mid-nineteenth century when a centralized public works administration came into being (Mosse, 1997a). Colonial interest in com- munity `custom and traditions' of resource management arose from concerns not so different from the present ones, namely an awareness of the financial implications arising from the state's obligation to maintain highly decentral- ized resources such as tanks over which it had recently extended its pro- prietorial rights. The mid-nineteenth century public works administration was seriously worried by the absence of such traditions or widespread reports of their recent decline. Indeed, from the 1870s this prompted a series of moves to document, demonstrate, codify and invent village traditions and even to enforce the practice of custom through legislation (ibid). Clearly, here at least `traditional' community management was a corollary of the extension of state power not its inverse. Theories of resource management institutions (whether based on village custom, moral economy, or rational choice) gain authority or `endow them- selves with rightness' (Douglas, 1986: 92) from their correspondence with the 472 David Mosse concerns of given administrative or development regimes (colonial or con- temporary). However, their validity as more than just operational strategies comes from being `substantiated according to the scientific standards of the day' - whether these be the observations of the authorized European officers who speak through the pages of colonial commissions of enquiry into village traditions of irrigation maintenance in the Enlightenment rubric of objective science (Ludden, 1993: 252), or the descriptions of collective action built on rules and incentives authorized by the economic models of human behaviour which command the position of dominance in contemporary development discourse. My argument is for a reconception of local resources systems in less general and more context specific terms. Let me be a little more precise about the issue, and the need to depart from 'economic-institutional' and `moral economy' models. I do not wish to suggest that common property resource use is other than the product of individual strategy and rational choice, but rather that there has been a failure to take cognizance of the fact that such strategizing is mediated by institutions (concepts, meanings, values) which are constituted in culturally and historically specific ways (Douglas, 1986). One of the principal points that I want to emphasize is that an institutional analysis of indigenous resources systems is unlikely to be useful unless it has first correctly characterized the social relations and categories of meaning and value in a particular resource system. In the first place this means resisting the tendency to impose a narrow definition of economic interest, utility and value (cf. Bourdieu, 1977: 177). For example, I hope to show that common property resources (in this case irrigation tanks in Tamil Nadu) are not simply sources of physical inputs (irrigation water). They are also repositories of symbolic resources. Tanks, like village temples, are public institutions expressive of social relations, status, prestige and honour. They are part of the symbolic `production of locality' (Appadurai, 1997). Through public works patron- age, privileged shares, gift or redistribution, tanks (like Tamil temples) illustrate what Bourdieu describes as the `interconvertability of economic capital (in the-narrow sense) and symbolic capital' (1977: 180). On the one hand the protection, construction, or repair of irrigation tanks long involved investments which generated `symbolic capital' in the form of honour or authority and created domains of influence for individual leaders. On the other hand, tanks themselves provide resources (water, trees, fish) which generate economic capital which is invested in symbolic capital through other public institutions, especially temples. Attempting to account for the strengths and weaknesses of institutions of resource management in terms of the balance of individual economic costs and benefits gives little recognition to the fact that material interests are often inseparable from social relationships, and that choices are mediated by shared assumptions about such things as justice, fairness and reciprocity (Douglas, 1986; Spencer, 1990: 98). History, Ecology and Locality in Tank-irrigated South India 473

The analysis in this article, then, attempts a conceptual re-integration of water resources management into the wider set of exchanges and social and political relationships involved in the Tamil villages studied. Because tanks stand for other things than a secure water supply to crops, a narrow economism is inadequate to understand their `management'. Tank common property has to be viewed as part of a `village public domain', and this means accepting a wider definition of economic interest than is assumed within conventional institutional-economic analysis. However, this does not require a `rehabilitation of altruism and cooperation' (Uphoff, 1992: 326-56), or involve conceiving of a large domain of symbolic action beyond economic analysis. Rather, such analysis has to be extended to embrace the range of symbolic interests normally rejected as economically irrational. As Bourdieu (1977: 177-8) puts it:

... contrary to naively idyllic representations of 'pre-capitalist' societies, practice never ceases to conform to economic calculation even when it gives every appearance of disinterestedness by departing from the logic of interested calculation (in the narrow sense) and playing for stakes that are non-material and not easily quantified ... The only way to escape from the ethnocentric naiveties of economism, without falling into populist exaltation of the generous naivety of earlier forms of society, is to carry out in full what economism does only partly, and to extend economic calculation to all the goods, material and symbolic, without distinction, that present themselves as rare and worthy of being sought after in a particular social formation - which may be `fair words' or smiles, handshakes or shrugs, compliments or attention, challenges or insults, honour or honours, powers or pleasures, gossip or scientific information, distinction or distinctions etc.

My study of indigenous irrigation in south India suggests, then, the need to overcome the distinction between rational action and `custom', and to reconceptualize common property resources as part of the village public as a material and symbolic domain in Tamil villages. If one line of argument in the article is that the economics (in the narrow sense) of water management does not exclusively define the form and meaning of tank institutions, a second, and rather different, one is that these institutional arrangements are none the less ultimately constrained by the facts of ecology. As Leach (1961) made amply clear in his study of a tank- irrigated Sinhalese village, social structure is not a thing in its own right independent of elementary facts such as that water evaporates and flows downhill. If there are dangers in a narrow economism, there are equally dangers in an over-socialized view of natural resources. The issue is not simply a need to recognize that ecology is not infinitely malleable (Benton, 1994; cf. Wade, 1995), but also to give explicit recognition to the ecological and territorial dimension of all social structures. Social structures are both embedded in and create particular landscapes which are therefore, the product of both social imaginings and ecological necessities (Redclift and Benton, 1994). Strategies for accumulating `symbolic capital' exist because they address ecologically determined needs, for example, to mobilize labour for a short but labour-intensive cultivation period, to control water runoff 474 David Mosse

from concentrated monsoonal rainfall, or to cope with the periods of uncertainty and crop failure. If, as Croll and Parkin (1992) put it, persons and their environment are `reciprocally inscribed' this `reciprocal inscription' takes place in ecologically as well as socially specific contexts. Since no ecology is uniform, the task is to identify meaningful local categories of social and ecological variation which produce distinctive systems of common property resource use. These are seen as generated in culturally specific ways, not as co-operative equilibria spon- taneously sustained by the long-run interests of foresighted self-interested individuals (Bardhan, 1993: 89). The focus of discussion is on systems of `tank' irrigation in Tamil Nadu in south India.2 These ancient and complex systems are a response to ecological facts - limited but concentrated rainfall, impermeable subsoil layers, and critical population density (von Oppen and Subba Rao, 1980a, 1980b). They also make demands on social organization. Tank water is a common pool resource whose joint use is subtractive. Tanks therefore require institutional frameworks for inter-household co-operation for water use, as well as higher-order linkage for construction, maintenance and arbitration. But, as Ludden (1978: 4) points out, irrigation does not generate any specific mode of agrarian organization, `its institutional base is shaped by the cultural and historical context within which the irrigation economy is developed'. My research firstly attempts to understand tank irrigation as a social system, and secondly to understand the considerable variation in institu- tionalized co-operative action within a given region. The simple question is, why in some villages do farmers co-operate to maintain and operate local irrigation systems, while in others these apparently critical common resources are neglected and abandoned? This is a question posed in Wade's (1988) influential study, which suggested the primacy of variables of ecology and risk over those of social structure as factors accounting for variation in the strength of institutionalized collective action. While I will show that condi- tions of drainage and soil properties do indeed influence the patterns of institutionalized co-operation and generate specific types of water manage- ment, it will be clear that local variation is also the product of distinctive and historically-specific forms of social organization. In describing variation in common water resources management I will draw on indigenous categories which elide the factors of ecology and society in defining distinctive local `cultural ecologies'. I now turn to a thumbnail sketch of a southern Tamil

2. Tank systems have only recently become the focus of research attention. Particularly important are a series of studies from the Madras Institute of Development Studies: overview studies of tank irrigation in Tamil Nadu include MIDS (1986), Sengupta (1985, 1993), Vaidyanathan (1992); empirical studies include Janakarajan (1989, 1991a, 1991b), Sengupta (1991), Sivasubramaniam (1995a, 1995b), Vaidyanathan and Janakarajan (1989). History, Ecology and Locality in Tank-irrigated South India 475 tank system which generalizes from detailed descriptions available elsewhere (Mosse, 1997c).

BEYOND THE `VILLAGE REPUBLIC': THE HISTORICAL DEVELOPMENT OF A HYDROLOGICAL AND POLITICAL LANDSCAPE

In the gently sloping southern Tamil plains districts of Sivagangai and Ramnad, heavy run-off from concentrated rainfall is captured, stored and controlled in some 10,000 reservoirs or `tanks' which provide virtually the only source of irrigation. Over 2000 of these tanks, formed by taking advantage of the natural depressions of the relief and erecting crescent- shaped earth dams across the drainage flow, fall under the Sarugani minor river basin. These tanks, which depend entirely on rainfall or supply from small seasonal rivulets are interconnected into extensive chains or `cascades'. These tanks are at the tail end of the larger drainage system and, partly because of up-stream irrigation wQrk, suffer chronic water scarcity. The modern state holds proprietary rights over tanks and undertakes sporadic repairs, but otherwise has relatively little to do with their management. Indeed, the social systems which operate these tanks are widely understood as part of the self-government of the traditional Indian village. However, for several reasons it is a misconception to see tanks as isolated village systems. The inter-connected tank systems studied (in Ramnad and Sivagangai areas), evolved as part of a pre-colonial state system which regional historians

Map 1. Ramanathapuram (Old) District showing Sarugani River Minor Basin

Source: Public Works Department Karalkudl, Tamil Nadu. 476 David Mosse n i as B nor IM.nkalanatn) Mi i so y d van an 3 K C S --4 m e c th th a f c o s Scaleiamlles pp-, k U an Basin T 2: p Minor Ma `k a Sarugani IN- I\\XA-Y_ciA7 the . of .- S'- . 0- : t 10 ` , Tanks 'O 2. .J. a d'' 'f . f Map d °P- . T . s V - ° . M Department, - U Works ;.CeD TI i Public from Sivaganga. Adapted History, Ecology and Locality in Tank-irrigated South India 477 now characterize as decentralized or `segmentary' (Dirks, 1987; Ludden, 1985; Price, 1996; Stein, 1980). This departs from both the established idea of the centralized hydraulic state and its currently popular opposite, the self- governing village republic. Tank irrigation and settlement expanded in the region between the fourteenth and sixteenth centuries as the southern plains areas were integrated into a region-wide warrior polity. Within this, local Maravar caste warrior chiefs created territorial and political domains of authority for themselves by extending tank systems, while ensuring an upward flow of resources to political overlords. These territorial-political segments, or natus, probably grouped communities dependent upon particu- lar water sources requiring maintenance, repair and dispute arbitration by Maravar clan chiefs or nattampalars (head of the natu). These segments were in turn part of larger kingly domains, being integrated through ritual transactions focusing on regional temples and articulated in the idiom of honour (Appadurai, 1981; Dirks, 1987; Price, 1996). The first point to stress is that tank irrigation systems did not come into being as autonomous village systems isolated from the state. Tank works were part of a political process in which rival chiefs extended and maintained domains of control. Rights to tanks or to control the flow of water were frequently the objects of royal donation and largesse to Maravar chiefs and kinsmen, warriors, temples, Brahman communities and others. For example, in 1654 a deed from Rengunatha Cethupathi (raja of Ramnad) granted the entire water rights of the Tenar, a seasonal river, to Periyatampi, a Cervai caste chief, to fill tanks in the Unjanai area of Sivagangai.3 The idiom of royal grant and entitlement persists today in articulating informal rights over water flow. Secondly, tanks and their catchments were not static physical or social systems. Breaches, diversions and new water courses evoked competing claims to title that required constant arbitration in inter-village disputes and in the nineteenth and twentieth centuries generated a mass of litigation. Thirdly, the resources for tank operation, and remuneration for villager services, were generated through a system of shares in the produce of cultivated wet land in villages which brought the interests of the state deep into the organization of individual villages. Irrigation costs were borne equally by villagers and the state in a system which organized agricultural production and redistribution as a public activity linked to the state, even though this was controlled locally by Maravar village heads or ampalars. In the nineteenth century the political system underlying tank irrigation was dismantled and replaced by Zamindari estates accountable to a colonial

3. A palm leaf (cadjan) copy of this record, together with one showing the transfer of the water right to the villagers of Unjanai following an oath at the temple of `Sevuga Perumal Ayyanar' at Singapuneri in 1863, were produced as evidence during litigation between 1935 and 1938. 478 David Mosse government and largely defined by eighteenth century British ideas of land holding. The ensuing changes were complex and well beyond the scope of this article. Suffice it to say, firstly, that changes in administration, political and legal institutions and property rights gradually dissolved segmentary political relations and broke up earlier resource flows and the `existing circuit of investment and social reproduction which had long obtained under pre- colonial regimes' (Stein, 1985: 412). On the one hand, as Dirks points out, `honours and local position could no longer be secured through the integrated relation of community benefaction and patronage, and [instead] local magnates began to invest ... in new economic opportunities afforded by colonial trade' (Dirks, 1987: 357, emphasis added); and on the other hand, as Washbrook argues, authority and social position no longer required such benefaction. In other words British rule in various ways emancipated local political chiefs and big men from the obligation to invest in community resources and public institutions such as tanks (Washbrook, 1976: 332, 1988: 92). The shortfall was not made good by the government's own public works. Secondly, the burden of much contemporary research has been to demonstrate that the creation of Zamindar estates did not, as the British expected, lead to a transition to agrarian capitalism, but rather allowed the continuation of an earlier pattern of political relations and kingly rule, based on redistribution and largesse (Baker, 1984; Price, 1979, 1996). As Price argues, the former rajas (the Zamindars of Ramnad and Sivagangai) continued to treat productive resources (land and water) as political assets to rule - i.e., to gift, exchange or redistribute - rather than as property to manage for profit. The extensive `alienations' of land and tanks in the form of grants and concessionary leases to temples, pilgrim centres, royal relations, or creditors which resulted from this (Rajaram Row, 1891) generated a highly complex landscape of insecure property rights and tank maintenance responsibilities. In addition, a series of litigations over royal succession further diverted resources away from investment in irrigation maintenance and repair. Thirdly, the nineteenth and twentieth century courts replaced the personalized networks of chiefly arbitration in disputes over water. The courts evoked older notions of royal grant and customary water flow, but struggled hopelessly to lock an endlessly changing ecology into a fixed set of rights. The evidence also strongly dispels the notion that tank systems faired better in Zamindari areas because spared the atomizing effects of British ryotwari tenure,4 as well as the assumption that these village resource use systems declined because the assertion of state rights over them led to a dereliction of community obligation (Sengupta, 1985). The physical deterioration of this tank system (extensively reported in the nineteenth and early twentieth

4. A revenue system introduced in territories under British administration which brought the government into direct relation with individual 'tenants' (ryots) who paid land tax based on detailed field-level settlement surveys. History, Ecology and Locality in Tank-irrigated South India 479 century) had more to do with wider political and economic changes brought by colonialism than the collapse of community co-operation, or a `tragedy of the commons'. This challenges the view, now firmly part of the new community management discourse, that present levels of resource degrada- tion are the result of the dissolution of traditional arrangements for sustain- able resource use and that the principal cause of this is the intervention of the state, and particularly the colonial state's assertion of proprietary right over non-private resources such as water and forests (Gadgil and Guha, 1992). Indeed, it was the creation (or invention) of isolated village tradition (in the sense of the dismantling of former overarching segmentary political structures) rather than its erosion which underlay a weakening of tank systems in this part of Zamindari south India. In fact, an ideology of autonomous village systems rooted in the exigencies of colonial administration (Ludden, 1993) continues to divert attention away from persisting links within tank systems. These links have been explored in detail for one network of tanks linked to a rainfed seasonal stream - the Nattar Kal (Map 3) - and falling within a former natu (see Mosse, 1997c). In this local tank network inter-tank links arise both from hydrology and the social connection between predominantly Maravar dominated villages. This is evident, for example, in the patterns of inter-village conflict which show a clustering around Maravar villages asserting or contesting rights to the control of water flowing in the shared channel. This inter-village and inter-tank competition for water still evokes the idiom of royal grant and entitlement especially in Maravar caste villages with former kinship or political ties to the raja of Sivagangai. Perhaps even more significant than this conflict are the co-operative water ties between villages. In relation to need, water is always unevenly distributed across the local inter-lined tank system. Locally specific factors of rainfall, hydrology and cropping produce conditions of surplus as well as scarcity. In response to this there is an informal market in which water common property is traded (as well as gifted and exchanged) and transferred between villages, over distances of up to 40 km through complex networks of channels. These transactions pre-suppose some social connection which, if not based on proximity, may involve relationships of kinship and caste. Indeed in the detail of individual cases it is clear that kin and caste links have played a significant part. Moreover, leaders with wide-ranging influence are able to act as `water brokers', arranging the purchase of water and its delivery or sale to villagers en route.

VILLAGE LEVEL WATER MANAGEMENT: `RULES IN USE' AND THE VILLAGE `PUBLIC DOMAIN'

Once it is secured in a tank, water is shared and rationed by the group of users. Such user-managed systems have been the focus of much description 480 David Mosse

Map 3. Section of Valani Tank Group showing Nattar Kal and selected Mankalanatu Tanks

Map 3: Selected Tanks, Channels & Streams

Mankalanatu tanks surveyed.

® Selected Karisalkatu tanks surveyed.

Centre of the naW terrritory. Village of the nattampalar - 'head of the natu'

Location of Vayalur' & 'Alapuram' tanks (Maps 4 & 5)

Scale. !mile

Source: Topographical Sheets, Public Works Department, Sivaganga, Tamil Nadu. History, Ecology and Locality in Tank-irrigated South India 481 and theoretical modelling. In the present study, 86 tanks were included in a rapid survey of water management practices in the area. Considerable variation was found in the extent of institutionalized co-operation among these, which I will discuss below. First, however, I want to comment on the nature of local systems of water control, sharing and rationing in the 58 tanks where they were found to exist. These systems are well adapted to water scarcity. At times of shortage an area-based rationing is adopted. Firstly, and most commonly, farmers cultivate a proportionate share (ivu) of their wet land holding, say 10-50 `cents' per acre. In 1995/6, a drought year, the proportion of land to be irrigated from the `Vayalur' tank (situated towards the end of a local tank network) was jointly decided for each watering, starting at 20 cents and reducing to 10 cents by the fourth watering. Under conditions of extreme scarcity a different system operates in which a fixed and equal acreage per household is allowed for irrigation. This entitlement is restricted to members of the village and is not extended to farmers from other villages owning land in the tank command. From documentary evidence in one village we know that this system was operating in the area in the mid-nineteenth century and almost certainly long before that. A few points should briefly be made about these systems in view of the tendency to view them either as generated by a community co-operative ethic (the `moral economy' view), or as a set of `rules in use' (Ostrom, 1990), or as expressing a consensual equilibrium outcome of self-interested actors (the institutional-economist view). Firstly, while the social control over access to tank water does indeed serve a collective need to manage risk and protect subsistence livelihoods, this does not mean that these systems are sustained by a community moral ethic. Social controls on the water commons may just as easily serve to establish relations of dominance and control, in this case principally those of Maravar caste headmen and their kin. Secondly, water use practices are not clearly governed by a system of agreed rules. To be sure rules are articulated; but what they describe are publicly accepted norms or official codes. Often, these rules `encode' the interests of some people better than others (cf. Bourdieu, 1977). Typically, the interests of poorer, lower caste or tail-end farmers and women farmers are poorly encoded in official rules (see footnote 5) - although in this case official rules are, in fact, strikingly equitable. However, rules are not invariably followed. Some individuals possess the capital of authority to deviate from the rule without attracting public notice or sanction, or, through `second order strategies' to `regularize' action which does not conform to the rules (ibid). Others, again typically poorer low caste and women farmers, do not. Even in villages where water sharing and rationing systems fail to operate or have been abandoned, areas of potential conflict, such as the use of tank water for dry land cash crops (chillies) rather than to raise the staple paddy crop on wet land, are hedged around with `codes of conduct' and a `morality of water use'. But again such rules or codes do not govern behaviour in any 482 David Mosse straightforward way. Rules are evoked publicly as much to justify individual interests as to order collective behaviour (Elster, 1989). Even more so than in villages with clearly articulated water distribution systems, in those without these systems individuals vary in their capacity strategically to manipulate rules to serve individual needs, to `put themselves in the right' (Bourdieu, 1977: 22) and to lay claim to irrigation water under one principle or another. Consider the following example in which public approval is sought for the diversion of residual tank water from its publicly acknowledged priority use, namely communal use by low caste residents along the tank bund (for cattle, bathing and other domestic usage) to a wholly illegitimate use, namely the irrigation of private cash crops.

In Alapuram village, in January 1995 a group of Pallar (a low caste) men objected to a Maravar man turning residual water from their tdvu (low place) to irrigate his crop, rather than leaving the water for their domestic use. In a public meeting to resolve the dispute, the Maravar man and his relatives appealed for and (after much argument) won public support for his right to the water on the grounds that it was needed to secure a late subsistence crop of paddy on his wet land and not for his cash crop chillies. He sought further to underline his moral right by pointing out that the Pallar farmers who raised objection had themselves illegitimately pumped tank water (by diesel engine) through his wetland fields to irrigate crops on their dry land. However, it subsequently became clear that under cover of this moral right, the Maravar farmer was able, privately and without incurring public dis- approval to sell much of this water to three farmers to irrigate cash crop chillies. The cunning and skill of these latter farmers in having secured closely protected residual water for their chilli crops was privately pointed out to me with a mixture of admiration and disapproval.

Clearly, a few farmers were able to manipulate publicly endorsed (but weakly enforced) codes of water allocation to divert common water to meet personal needs.5 The case only serves to highlight the need to challenge the pervasive

5. The fact that women did not participate in this public teashop discussion is important but not culturally surprising. The way in which women and their interests were represented by men in the dialogue is, however, significant. In the dispute (Pallar) women had clear interests - namely preserving accessible tank water for washing clothes, bathing and other domestic uses, and preventing its diversion to cultivation. In fact, at one level this is precisely what the dispute was about - namely Pallar protection of their domestic water supply. The first point is that despite this, at no point in the meeting were Pallar women's interests articulated. Pallar men did not stake their claim to retain tank water in terms of the needs or rights of women. Indeed, the terms of the public discussion, the set of norms and codes evoked, appear to have precluded a focus on women's interests. Discussion concerned rights to water in terms of land type (dry/wet), crop (rice/chillies), and legitimate/illegitimate diversion of water and the use of pumps, not the problems of carrying water from distant sources. Women's interests could not be evoked to win the argument, and in fact the argument was lost. This merely illustrates the point that women's interests in tank water are not well encoded in publicly evoked rules and norms. The second point is that although women were referred to in this public discussion, they featured not as holders of particular interests or rights, but rather as the source of disruption and disorder. In the course of the discussion it was pointed out that women History, Ecology and Locality in Tank-irrigated South India 483 concept of 'rule-governed' behaviour applied to common property resources systems, which tends to reify the rule, to falsely attribute motivating power to it and to overlook the different interests and strategies which rules conceal. It may be more useful to adopt an agency-focused analysis of praxis in which rules are viewed as part of a legitimizing discourse and in which the capacity to `operationalize rights by rules' is influenced by power and social position (Li, 1996: 509-10). Thirdly, rule conformity is often mistakenly identified as the principal indicator of institutional success or efficiency. Apart from the fact that (because of subtle secondary strategies) rule compliance may be self-fulfilling among local resource users (i.e., `officially' rules are always followed), there is a further problem when rule conformity is taken as a measure of successful collective action, independent of factors such as productivity or equity. Indeed, when community water management institutions are evaluated - as they often are - by the measure of rule conformity, success becomes self- fulfilling, defined in terms of the prescriptions of the model itself (Palmer- Jones, 1995, has criticized Tang, 1992, in this vein). However, as Rajivan (1991: 221) points out, `cohesive institutions may often lead to inefficient or inequitable area and productivity outcomes. Some irrigation institutions are cohesive enough to ensure that there is no cultivation at times of shortage'. We know, for example, that under conditions of extreme scarcity, as in the years 1865, 1876 and 1879, the ivu rationing system in Sarugani village failed to avert massive crop failure. Indeed, in the famine year of 1876-7 rule conformity appears to have dissipated a minimal water supply and contrib- uted to massive crop failure in the village (see Mosse, 1997c).6 Finally, in villages where it exists, the crux of the water distribution system is in any case not a set of allocation rules, but rather a service relationship involving hereditary Pallar caste `water turners' or nirppaccis (also known as

would object and cause conflict if water was pumped for irrigation. Also provocative ideas (outrageous claims) were introduced in the form of, `women have been saying that ...'. Having embroiled themselves in heated conflict, men could agree that `without reason women will simply make trouble'. Women would fight over water and their rumours had already caused men to argue. It was as if men could resolve their disputes in public not only by drawing on and underlining `official' norms and rules, but also by evoking excluded women as the source of social conflict and ignorant misunderstanding. 6. An account of the great drought and famine of 1876-7 in Sarugani shows that under conditions of extreme shortage, the ivu system of cultivating an equal proportion of land actually increased the risk of total crop failure and therefore acute shortage and starvation. There was, in that year, a conflict of interest between the farmers who wished to preserve an equitable distribution of tank water but use it to raise ragi (finger millet) a less water demanding crop, and the inamtar (the priest and holder of the grant) who received a half share of wetland produce and who wanted to divert the entire tank water to nine fields (including church land) on which the crop would be saved and from this distribute rice as `alms' to the villagers. In the absence of an agreement, the water was distributed as usual to all fields and, as in other drought years (1865, 1876 and 1879), only a few fields (1-6) actually ripened. 484 David Mosse matai kutumpan or matai variyan),7 paid by farmers in grain or cash on a per- acre basis. The nirppaccis open and close the tank sluices and distribute water to the fields.8 These `water turners' and not farmers themselves are permitted to irrigate fields of standing crop, according to their own finely graded judgements on field-specific conditions of moisture, soil and crop condition. This serves to mitigate the inequity between head and tail end plots which is inherent in irrigation systems, and elsewhere resolved through arrangements of land tenure (Leach, 1961). The `water turner' role is part of an inferior public office whose status is derived from its position within an older pre-colonial caste-based social order organizing agricultural production and caste society (including the system of rights to shares in the village harvest). This village office (usually known as kutumpan) was held by members of the Pallar caste and principally involved two roles: the distribution of water from the tank, and the distribu- tion of grain shares on the threshing floor. The kutumpan office hierarchically complemented that of the Maravar headman or ampalar. While Maravar ampalars ruled, ordered, and controlled the production processes in the village, Pallar kutumpans had the derivative function of executing the dist- ribution of productive resources (water) and harvested grain. Moreover, symbolically speaking, while Maravars were (and are) at the ordered centre of village affairs, Pallar public roles involve them with marginal, liminal, and dangerous spaces, for example, as gravediggers and woodcutters. Along with the graveyard and the forest, the tank and its sluices are dangerous liminal boundary zones which attract spirits and demons, while dawn and dusk - the time for opening and closing the sluices - are liminal dangerous times. It is for this reason that an animal sacrifice is commonly required at the tank sluices to propitiate the `sluice deity' (matai muni).9 The complementary relationship between the Maravar ampalar and Pallar kutumpan is articulated in roles and prestations at village festivals, and reproduced at certain regional temples of the natu in which ranked honours are given to the eighteen ampalars and the twenty-two kutumpans (Pallar leaders) who notionally constitute a natu.

7. Such low caste roles, variously named as nirppacci, nirkatti, nirani, among others, and having varying responsibilities and remuneration, are widely described in the literature on south Indian tank systems, e.g., Janakarajan (1991a: 10-14, 1991b), MIDS (1986: 111ff), Sengupta (1991), Shankari (1991: 97-8), Vaidyanathan and Janakarajan (1989: 89). Occasionally nirppaccis also have responsibility for guarding crops, although more often this is a separate function. 8. In a few villages the roles of opening and closing the sluice and that of irrigating the fields are separated and performed by different people. 9. In `Vayalur' the sacrifice is made at one sluice and blood sprinkled on other sluices to keep the dangerous spirits and demons (pey-picacu) at bay. Several cases of spirit possession encountered in the area involved munis (demon-deities) associated with the tank, such as Kalunku Muni - muni of the overflow weir, or Tamaramatai Muni - muni of such and such a tank sluice. History, Ecology and Locality in Tank-irrigated South India 485

More could be said on this, but the point to emphasize is that while at one level the tank is an irrigation structure, at another it has long been a public institution (like the village temple) expressive of an order which implies relationships of dominance, dependence, and caste rank. This is not only the case in the Ramnad and Sivagangai area but more widely in Tamil Nadu, although interestingly the idiom in which irrigation tanks, as institutions of public order, express existing relations of authority, varies regionally. In the northern districts characterized historically by landholding collectives (termed mirasi by the British), rights and obligations in tank water (as in temples) are expressed in terms of privileged shares (pankus) (CWR, 1990; Mosse, forthcoming; Sivakumar and Sivakumar, 1996). In areas such as Pudukkottai where resource control was grounded in the clan institutions of the ruling Kallar caste, rights to tank water and temple honours are articu- lated in the idiom of ranked lineages (karai) (Krishnan and Mohanraja, 1995: 225; cf. Dirks, 1987: 210-12). In Ramnad and Sivagangai, common property is circumscribed by hierarchical relations of service, the model for which is provided by the Maravar headman (ampalar) and the Pallar kutumpan. The point here is not only that resource entitlements are defined by social position (e.g., by caste and kin) but also that publicly managed resources (and the institutions involved) provide idioms of political relations and social standing (Spencer, 1990: 100-1). As Granda, working on Tamil property rights concepts derived from inscriptional sources from the fourteenth century, concludes, land and water in Tamil Nadu are not only exploitable resources but also media through which a variety of social relations have been structured (Granda, 1984: 3). The more general and pragmatic point is that the cultural and symbolic construction of `common property' has a bearing on the way that these resources are (or have been) managed or mismanaged. This can be illustrated by other facets of the southern Tamil tank systems. Firstly, social conflicts arising from the symbolic dimension of tank systems influence water management. Where water management systems are embedded in authoritarian caste structures, they are also affected by chal- lenges to these structures. In a number of villages surveyed, Pallar nirpaccis have withdrawn their services, often following disputes at temple festivals. These ritual occasions give public focus to service roles and dependencies and are therefore occasions to raise the `problem of honour', to make public protest and advertise the more widespread collapse of hierarchical depend- encies. Put together these social protests have had a significant (negative) impact on water management and productivity. Perhaps even more signifi- cant, however, is the fact that in villages with strong water management institutions, the economic need for `water turner' roles has resulted in their re-negotiation, re-constitution and re-definition in socially acceptable forms. The changes in payment, roles and obligations involved are part of a wider re-negotiation of identity and status among low castes often involving a shift towards non-hereditary, non-caste specific contractual forms of employment. The disengagement of water management from caste authority has its costs, 486 David Mosse however. New systems are often complex, time-consuming and involve elaborate supervisory systems described elsewhere (Mosse, 1997c). Secondly, non-water tank resources are managed in culturally distinctive ways. As well as water, fish and trees are managed as part of village level tank systems. These resources - and especially auctioned trees - generate significant common financial resources. In `community management' models such resources are often viewed as important for investment in tank maintenance - for desilting, clearing channels, repairing bunds and sluices, paying watermen or even the costs of litigation. What is striking, however, is that in the villages surveyed common funds generated from tank resources are hardly ever used for these or any other practical public works. In the vast majority of cases (73 per cent of tanks surveyed) income from all sources is expended on temples - temple construction, repair, inauguration or the celebration of festivals (Mosse, 1997c). Public funds in these villages are not primarily managed in ways which maximize utility and ensure accountability, but rather in ways which minimize social conflict and serve to enhance the prestige and credibility of existing leadership, as religious donors. Indeed public funds are not managed (rarely are they placed in banks with accessible accounts) but rapidly expended on a narrow range of culturally acceptable public goods, foremost among which is the temple. As Tamil political leaders at all levels have long understood, while conferring honour and public esteem on its donors, the temple demands generosity, but not accountability. Overall, the point to reiterate is that common property resources manage- ment cannot (as is so often the case) be isolated from context and viewed as a distinctive type of economic activity; at least not without overlooking critical factors. While responding to ecological necessity and the need for co- operative control, tank management arrangements in Tamil Nadu are also shaped and circumscribed by culturally specific notions of the public domain, public service and public action. In Tamil villages (some at least) the `village public' (ur potu) happens to be the domain of authority and rank and status, which makes the management of common property an eminently political action. This also makes tank management a heavily gendered action in that women (and their interests, see note 5) are largely excluded from this public domain and its decision making.1°

10. The exclusion of women from irrigation as public action is often mistakenly read as implying that women do not have (different) interests in this area. This is manifestly not the case (cf. Zwarteveen, 1995). Women have distinctive interests and perspectives on water use in terms of the choice of crops and how they are irrigated as well as the need for water for domestic purposes. For example, where women are interested in maximizing subsistence yields, or reducing the demand for weeding, or securing crops over which they have more control they may desire irrigation practices different from men. Moreover, despite the `official' model of male control, women sometimes are able to exert influence over water use or to subvert `public' organizations controlling water management (Zwarteveen and Neupane, 1995). History, Ecology and Locality in Tank-irrigated South India 487

Even where ritualized systems of rights/obligations no longer exist, or where local authority has been largely undermined, matters of the tank remain public affairs full of the possibility of conflict, and tanks continue to symbolize public interests and have a role in political strategizing. For example, contenders in panchayat elections have stated their intention to improve the tank as a way of validating claims for political support on a non- caste, non-religious basis. Equally, organizing action in relation to the village tank (e.g., voluntary repairs, clearing weeds, or complaints against con- tractors) provide the means for new forms of association, especially those of non-dominant groups (low caste youth or women) conventionally excluded from influence over the public domain, to assert a claim to wider legitimacy, authority and status. In these cases the village tank is not a context for the expression of authoritative control but part of an increasingly contested political domain. Ironically, where unitary caste authority no longer exists in villages, individual political ambition and competition generates a good deal of tank maintenance activity, even though sporadic and symbolic. Because tank resources - rights to fish shares, the use of common funds, service roles etc. - continue to provide ways of talking about the distribu- tion of power in the community, they tend to generate disputes out of proportion to their material value (which may be no more than a handful of fish). This only underlines the fact that the economic benefits of co-operation do not in themselves define or exhaust the meaning of the social institutions which produce co-operation. I emphasize this point because others take a different view. Wade (1988), for example, in his influential study insists that village studies in India have long neglected the strong tendency among villagers to produce corporate organizations meeting practical needs and to isolate these from caste and factional politics. The village public realm, Wade insists, is about getting things done (1988: 196). In this context his study makes much of the existence of functionally focused village councils concerned with managing non-privatizable village resources. But while high levels of co-operative action exist in 58 out of 86 tanks surveyed, only two had any recognizable `council'. Such examples are often seized upon by policy makers who strongly desire the existence of such bodies. In this region, however, these are truly the exception. Evidence from Ramnad/Sivagangai suggests that contrary to Wade's conclusions, village organizations are not widely characteristic of rural Tamil Nadu, despite the widespread official belief in their existence, or the ideology of their recent decline. The absence of corporate organizations cannot be taken as evidence of their decline: on the contrary, while the evidence for the existence of corporate organizations in the past is at best sketchy, many of those which do exist can be shown to be of recent origin. Secondly, the absence of corporate organization does not mean the absence of collective action. Corporate organization is not, as Wade suggests, an all or nothing option (ibid: 187). 488 David Mosse

ECOLOGICAL VARIATION AND COLLECTIVE ACTION

I will now turn briefly to the question of local variation in the extent of institutionalized collective action in the operation of tank irrigation. Perhaps it is true to say that anthropologists have been more concerned to understand the nature of social institutions than to account for local variability in them. At one level the very particularity of the local instance is acknowledged as the product of such a complex set of historical circumstances as to defy explana- tion; at another the individual instance is seen as informed by cultural material having far wider general applicability. In the context of concern about common property resource use, a more intermediate set of questions is increasingly the focus of attention - more often of economists than anthro- pologists - namely the question mentioned above, why in some villages do farmers co-operate to manage common resources (e.g., to maintain and operate irrigation tanks) while in others apparently critical productive resources are neglected? Like Wade, I am interested in identifying a systematic pattern of variation in institutions of (water) resource control - in this case institutions of specialist water turners (nirppaccis), systems of water sharing and rationing at times of shortage and mechanisms for water acquisition (channel main- tenance, water dispute and trade) - and to examine this within in a variable physical environment, i.e., across a catchment. To do this, a large number of interconnected tanks (86 tanks in 76 villages) stretching across a catchment were included in a survey of tank institutions. Strikingly, and somewhat unexpectedly, I found a sharp contrast between the social organization of tank irrigation in the upper and the lower parts of the catchment studied. In almost all tanks (in 54 out of the 56 villages surveyed) in the upper part of the catchment, water is controlled through the sort of sophisticated water rationing systems and nirppacci roles which have been described above. In every one of the 21 tanks surveyed in the lower part of the catchment these institutions are wholly absent. Here, tank water use is individually negotiated rather than embedded in social relations. Access to water depends upon skill, private social networks and capital. Weakly enforced public codes of water use more often provide a means to legitimize individual practice than to order collective behaviour. In these `non-co-operative' villages farmer strategies are also shaped by changing technology. Today secure irrigation depends upon access to diesel pumps and lengths of pipe which allow direct pumping of water from the tank into field channels and over ineffective and poorly maintained tank sluices. The successful pursuit of strategies of individual `opt out' from the commons water resources (which is akin to the use of private energized wells elsewhere) depends not only upon capital to purchase equipment, but also on the complex business of securing water patrons (or clients). In 'co- operative' villages the use of such technology is limited and strongly controlled. History, Ecology and Locality in Tank-irrigated South India 489

I have argued that irrigation institutions have both economic and socio- political/symbolic dimensions. In accounting for variation in the level of institutionalized co-operation in tank systems it will be necessary, therefore, to take both perspectives into account. Let me begin by explaining the ecological-economic perspective. In the tank system discussed here, the contrast in commons use arrangements coincides with a sharp divide between two micro-ecological regions distinguished, inter alia, in terms of catchment and runoff characteristics, soil type and cropping patterns. There is a sharp boundary within the Sarugani minor basin between the light sandy soil in the northwest (upper) and heavy clay soil in the southeastern (lower) part (see Map 2). This distinction is marked by a sharp change in tank density, shape and orientation. In the upper part of the catchment lateritic soils produce high runoff and densely packed tanks interspersed with paddy land. The pattern at village level can be seen in the village maps (Map 4). On these sandy soils, cultivation has long been exclusively transplanted rice, domin- ated today by high yielding short duration varieties. Ground water is often saline and there are only isolated pockets of land irrigable by wells. Only if the tank is full can a harvest be assured. Rather as Wade predicts it is the scarcity and high dependence on tank irrigation associated with rice cultivation on porous sandy soils which corresponds to the sophisticated and socially embedded village water management arrangements described earlier. In particular Pallar nirppaccis and water rationing systems are only found in these villages. Also, sandy soil

Map 4. Tanks and Wet Land in `Vayalur' Revenue Village

Map 4: Tanks and wet land in 'Vayalur' Revenue Village. 0 Tank Irrigated wet land. © Unirrigated dry land ® Flooded land cultivated as tank water recedes. Government land RE Tank water spread area

Source: Village survey maps & A'Register for the year 1949 81983. 490 David Mosse

Map S. Tanks and Wet Land in `Alapuram' Revenue Village

Map 5: Tanks & wet land In 'Alapuram' Revenue village (Tank irrigated area ®Unirrigated dry land Flooded land cultivated as tank water recedes

Source: Village survey maps & A'Reglster for the year 1949 & 1983 villages maintain tank supply channels, they are closely inter-linked through patterns of water flow and it is only in this area that the inter-village trade in water mentioned above takes place. In the lower southeastern area more widely spaced tanks on water-retentive black cotton soils irrigate a smaller proportion (18 per cent) of agricultural land and are separated by large tracts of dryland on which rice, chillies, cotton, linseed, gingelly, coriander, vegetables and a wide range of other crops are grown (see Map 5). Tanks have a less critical part in the black clay soil farming system, in which rainfed cultivation has a far greater role. Overall, the benefits of tank irrigation (increased profit and reduced risk) are less in black soil than red-soil areas. Tank institutions are correspondingly weaker. The wider set of contrasts associated with these two ecological micro- regions is summarized in Table 1. It appears almost undeniable that the pattern of collective action in the Sarugani basin tanks is an expression of ecological variation; that it is shaped by a structure of payoffs arising from ecologically determined costs, benefits and risks and therefore strongly complies with the predictions of institutional-economic theory. In the one (sandy soil) area ecological conditions - rice cultivation under sandy-soil induced scarcity - are such that both the benefits from co-operative water use and costs of not co-operating (in terms of crop loss and social conflict) are high. Moreover, landholding patterns give almost all farmers direct interest in irrigation as owner-cultivators of wet land. In the other (clay soil) History, Ecology and Locality in Tank-irrigated South India 491 area, risks and the benefits of co-operation are low and land interests dispersed through extensive rainfed cultivation and complex tenure systems. All this would appear to confirm Wade's general thesis on the importance of ecology and risk in accounting for collective action, namely that collective action is predicted in commons situations where joint use and subtractive benefits are coupled with scarcity (1988: 184). 'Interestingly, while Wade's general principle is confirmed, my research reverses his observations. Wade observed that villages lower in a catchment (tail-end irrigated villages) face greatest risk and have strongest institutions. However, in the Saragani basin tanks, those at the tail-end have the weakest water management institutions. The difference is easily explained. While, in Wade's case, it is position within the distributary (head or tail-end) which determines risk and scarcity in terms of varying degrees of reliability and adequacy in water supply, in the segmented rainfed tank systems described here it is water demand determined by the soil type which is the critical factor. Indeed the lower demand for water in tail-end clay soil villages has resulted in neglect and the further reduction in an already reduced water supply (e.g., through siltation and encroachment of supply channels) and tank capacity (through silted and raised tank beds; cf. MIDS, 1986: 122).

LOCAL CULTURAL ECOLOGIES

While the analysis of tank institutions in terms of ecological variables and individual economic incentives is both compelling and useful in setting out the conditions for collective action, in this section I will argue that it is inadequate to understand the particular form and dynamics of existing co-operative institutions or to plan for new ones. For this, it is necessary to analyse the specific interactions of social, cultural and historical circum- stances with ecology. There are several aspects to this. First of all, it cannot be assumed that all tank institutions are derived or sustained as co-operative solutions by autonomous user communities. As I have already pointed out, the notion of spontaneous equilibrium institutions internally sustained by a structure of economic (in the narrow sense) incentives diverts attention away from the fact that, historically, state and village have been inter-linked and that, certainly in Ramnad and Sivagangai, there have always been strong external interests in the organization of agricultural production and water management. Systems of revenue which defined rights to a share in the produce of tank irrigated land, the obligations to maintain a complement of village officers through grain shares or land grants, and the duty to keep tanks in repair, all influenced the operation of village tank institutions. Moreover, the nature of external interests in tank irrigated paddy production was not uniform but varied across the Zamindaris depending on the particular form of local tenure and the local history of village and tank gifts, grants or leases. The different types of right Table 1. Contrasting Characteristics of Sandy Soil (Mankalanatu) and Black Cotton Soil (Karicalkatu) Areas

Mankalanatu (Sivagangai Taluk) Karicalkatu (Ilayangudi Taluk)

Soil/land Light, sandy red lateritic soil Heavy, clay soil, black cotton soil Porous Water retentive High proportion of cultivated area irrigated Low proportion of cultivated area irrigated (86% Taluk-wide), nearly all by tanks (18% Taluk-wide), nearly all by tanks More uncultivable and forest land Less uncultivable and forest land Large price difference between wet and dry land Less price difference between wet and dry land

Cropping Mostly irrigated cultivation Mostly dryland cultivation Tank-irrigated paddy cultivation Diverse cropping adapted to moisture conditions 100% of paddy irrigated (Taluk-wide) 20% of paddy irrigated (Taluk-wide) Only paddy grown on wet land Cash crops (e.g., chillies) on 30-60% wet land Transplanted paddy Direct sown paddy Dominance of a few paddy varieties Diversity of paddy varieties Exclusively high yielding rice varieties `Local' rice varieties widely grown Short cultivation season (3 months) Long cultivation season (6 months) Surplus tank water used to raise a second crop of Only one cropping season, surplus water used paddy or ground nuts for chillies Very limited dryland cropping - millets Diverse dry cropping: rice, chillies, pulses etc. No inter-cropping Inter-cropping to reduce risk Variable area cultivated Area cultivated constant, yield variable High risk Lower risk High dependence on tanks Lower dependence on tanks Long-term continuity of cropping pattern Cropping pattern changed in C19th and C20th

Land tenure All farmers own wet land Few better off farmers cultivate wet land Landholding pattern relatively fixed Landholding pattern changeable Little tenancy Complex land tenure arrangements: share-crop and land mortgage tenure Tanks Number and density of tanks high Number and density More of tanks low small tanks (86% under 40ha. in Taluk) Fewer small tanks (47% under 40ha. Interconnectivity tanks in Taluk) of high Interconnectivity of tanks low Supply channels maintained Supply channels Tank abandoned/encroached bunds eroded and repaired No repair needed or performed on Tank silt not applied to bunds fields Occasionally silt applied to fields Fewer trees in tanks More trees in tanks Water management Pallar caste `water distributors' No `water distributors' Paraiyar caste channel watchers No channel watchers Water rationing during shortage No systems of rationing No diversion of water to dryland Diversion of tank water to Water dryland for chillies management embedded in social relations Water management individually negotiated Inter-village conflict over diversion of supply channels Less inter-village conflict, focuses on encroachment problems Inter-village trading of tank water No tank water trading Surplus tank water sold Surplus water used for dry crops History of social relations Continuity in Maravar `warrior' caste dominance No continuity of caste dominance - Utaiyar cultivator castes settled in C18th, Maravar dominance challenged in C19th Limited options for mobility through land acquisition Considerable opportunities for low caste mobility through land acquisition (lease, purchase); Low caste service relations challenged in 1990s Low caste service relations abandoned in 1960s, history of caste conflict 494 David Mosse also implied different degrees of authoritative control over the production process in tank irrigated lands, ranging from the diffuse authority of the king/Zamindar, mediated through local Maravar headmen, to the more direct authority of the beneficiary of gifted or leased villages and tanks. These historical and institutional factors have produced distinctive local types of tank operation. However, they have not been independent of eco- logical constraints. For example, while share-based revenue systems penal- ized wet land cropping everywhere, only in black cotton soil areas could nineteenth and early twentieth century farmers reduce or abandon paddy cultivation under tanks and switch to dry grain (millet) production." In these areas, as one administrator observed in 1910, `the tenants [cultivators] have no incentive to improve their wet lands and they throw all their energy and money into their dry lands which bear a fixed cash rent and bring greater out turn'. 'Z In such localities, the nineteenth century also brought new opportunities for cultivating profitable commercial crops on dry land, especially cotton and groundnut, for which there were expanding markets (Baker, 1984; Ludden, 1985). The result was a steady reduction in the importance of cultivation under tanks in black cotton soil areas as `the productivity of dry lands increased in relation to wet lands' (Ludden, 1985: 135-40, 149-56). This had the effect of weakening tank structures and institutions. In sandy soil areas, by contrast, ecological constraints have ensured a remarkable continuity of paddy cultivation under tanks over a two hundred year period. The persistent dominance of tank irrigated paddy cultivation has allowed a continuity of authoritative control over wet land rice production and redistribution, and this authority has been expressed in culturally specific conceptions of authority, service, protest and mobility (i.e., of the village public domain, ur potu) which have in turn given tank resources their symbolic significance in the articulation of hierarchical social relations. While tank institutions function to distribute power and symbolic resources as well as water, they do so in locally specific ways. In some villages (especially in sandy soil upper catchment villages) tank common property remains part of a `public domain' expressing relations of Maravar caste dominance and control. In others (lower catchment clay soil villages), however, the tank is more often part of a contested public domain providing symbolic resources for individual and group political strategies. The difference between the two areas is not simply that self-interested farmers are rationally constrained to follow rules in one local ecology but not

11. This practice was widely reported and condemned by British officers from the beginning of the nineteenth century, e.g., Diary of the Superintendent of tank repairs and water courses for the month of January 1801 (Tamil Nadu Archives). 12. Note from Estate Collector of Uthumalai Zamindari to Secretary of Court of Wards 14 Dec 1910, commenting on Virakkeralampudur, a tank irrigated black loam soil village (Sivagangai Record Office). History, Ecology and Locality in Tank-irrigated South India 495 in another. Rather, it is that in one set of villages power and authority tend to be articulated through public institutions (the temple, the tank, service roles etc.), while in another set, power operates through more diffuse private networks of patronage, alliance and personal obligation. If in the one type of village the dynamics of power still support public institutions of resources management, in the other type, the public domain is more often charac- terized by contest, competition and the secondary strategies used to win public opinion in support of the pursuit of individual interests. Thus there are differences in the way that tank resources symbolize social relations, and these have to do with both history and ecology. As far as can be known from folk history and early colonial records, in the eighteenth and early nineteenth centuries public institutions for the control of tank irrigation and agricultural production - including grain shares and nirppacci ('water turner') roles - characterized all of the Ramnad and Sivagangai regions. For the most part these were based on strong Maravar territorial institutions of control (i.e., natus). The question is, why did agricultural and social changes in the late nineteenth and twentieth century erode water manage- ment institutions in one ecological region (lower catchment clay soil villages) while preserving them in another (upper catchment sandy soil villages)? One factor has already been mentioned, namely the opportunities to diversify away from irrigated rice presented in clay soil areas. Another factor was the displacement of Maravar controlled public systems of resource use as a result of the immigration of successful dryland cultivator castes (Utaiyars) into black cotton soil areas in the late eighteenth century. In the nineteenth century, secure property rights and new economic opportunities from dry- land cash cropping (as well as remittances from Burma), encouraged by British government, gave economic power and land-based economic relations (including land tenure arrangements and associated forms of patronage and dependence) a new ascendancy over older forms of political dominance. In certain localities, the growing economic power of Utaiyar cultivators (and also Chettiyar bankers) came to challenge the authority of the old order, based as it was on Maravar kinship, political office and control of wetland land production and revenue. This social change was reflected in local struggles for caste power and status in the form of `honours' in regional temples and shrines in the nineteenth century (Mosse, 1997b; cf. Baker, 1984; Dirks, 1987). As the political leadership of Maravar areas came to be replaced in the late nineteenth and early twentieth century by land-based dominance of Utaiyar (and other) landlords, the village public institu- tions - tank and temple - through which Maravar dominance had earlier been asserted were eroded. Clearly, the erosion of `public' systems of resource use in black cotton soil villages, and their retention in sandy soil villages, is the result of complex factors of settlement, caste power, property rights and economic opportun- ities, revenue systems, and colonial markets. This complexity has to challenge any oversimplified ecological determinist reading of the pattern of collective 496 David Mosse action in tank management in the Sarugani river basin even where this appears to be firmly aligned to physical factors of soil type or position in the catchment. The final, and perhaps most compelling, reason for rejecting a narrow ecological-economic view is that this does not accord with the categories of villagers themselves. The physical and institutional characteristics of the two ecological micro-regions are defined locally by two indigenous categories: the first mankalanatu refers to the area of sandy soil (manal man) and wetland rice cultivation (natu) in the upper catchment area; the second karicalkatu refers to the area of black cotton soil (karical man) and predominantly dryland (katu) cultivation in the lower part of the catchment. These categories are not simply indigenous eco-types, however; they also convey information about `human types' including social institutions and beha- vioural propensities. When villagers explain the existence of strong water control institutions, they say that the people in mankalanatu have kattuppatu; a term meaning order and discipline, and conveying both the notion of unity and co- operation (the ability to agree rules, resolve differences and impose sanctions) and that of strong caste authority and associated service and subordination. The English word `control' is often used in discussing kattuppatu. This characterization refers to a public domain; it does not at all imply individual prudence and thrift. On the contrary mankalanatu men are characterized as spendthrifts, prone to lavish display, extravagant expenditure on clothes, festivities and entertainment. They are to be found in the market place sporting terylene shirts and flashy watches rather than labouring in the fields. They also fall easily into debt and sell their harvest to repay loans. The concerns of marikalanatu men are public status and honour, not hard work and, as they proudly point out, unlike karicalkatu men, `mud will not stick to our moustaches'. Underlying this characterization is an ecological truth - namely that farmers in the marikalanatu area can only cultivate in one short (three month) period and cannot extend season or profits by skill or hard work. What is far more crucial to success is the accumulation of the `symbolic capital' in the form of social position, honour and authority necessary to mobilize individual and communal labour and to reproduce dependent service roles to work and irrigate fields in a highly concentrated `labour period' (Bourdieu, 1977: 179).13 In this way, risk and a short cultivation season have supported the historical evolution and continuity of strong public institutions and systems of honour. In karicalkatu villages, by contrast, whether in terms of large landowners opting out of community tank systems, the dispersal of landed interests

13. Following Bourdieu's analysis, `symbolic capital' - networks of alliance, the capital of rights, duties and honours - can be understood as credit encashable as rights over labour and service at critical times (1977: 181). History, Ecology and Locality in Tank-irrigated South India 497 beyond tank wet land, diversified rainfed cropping and complex land tenure arrangements, conditions have long militated against sustaining collective action for tank water management. This much is recognized by villagers who hold a general belief that in karicalkatu there is no co-operation, no public order (kaquppatu), no rules (mural) and `no control'. As a social type the karicalkatu person is thrifty, hardworking, strong and individualistic. Karicalkatu men are cautious, they do not like to take loans and never sell grain until the next season's rain is assured. They avoid public display, and conceal their profits. If a karicalkatu man harvests 40 kalams14 he will claim that his field yielded only 25, whereas if a marikalanatu man harvests only 10 kalam he will boast 40. In karicalkatu, status and public honour count for less. Agriculture on karical soil is a never ending task likened to the making of rope which rewards skill and individual effort. In karicalkatu, ecological characteristics have encouraged a long term and historically shaped erosion of public institutions and a tendency towards the expression of power and status through more private forms of accumulation, patronage and alliance. In conclusion, the categories of marikalanatu and karicalkatu imply both the ecological conditions producing different degrees of risk (related to water scarcity, growing season, cropping pattern) and the social and symbolic systems which respond to this and which shape specific kinds of resource use. In their articulation of caste stereotypes and inter-relations, these categories also imply distinctive local histories of settlement and conflict. From the point of view of (male) social agents themselves, therefore, it would be quite wrong to attempt, as Wade does, to separate out the ecological-economic from the social symbolic aspects of common property systems and to apply an institutional analysis only to the former, or to discount as economically irrational the complex sets of meaning and symbolic interests involved in resource management systems. It should be obvious, however, from the use of gendered pronouns that this is a pronouncedly male discourse. The seasonality of women's workloads does not generate the same contrasts, and public institutions and public honour do not have the same significance from their point of view. As cultural ecological categories marikalanatu and karicalkatu point to a distinctive discourse on locality and resources. In this, physical resources are quite inseparable from social identities; notions of place and person fuse. As Daniel (1984) shows, in Tamil conceptions soil itself is a medium through which social identity is moulded and transformed. Soil has person-centred as well as ecological characteristics and its qualities effect those of the persons living from it. As Vasavi (1994) points out with reference to farming in Karnataka, the practice of agriculture too is founded on `precepts' (concepts and actions) which incorporate social and moral concepts. The concept of `abundance', for example, differs from the economic concept of productivity

14. A volumetric measure of paddy. 498 David Mosse as yield or worked output in that it encompasses specific moral ideas of social obligation, share and gift. `Technical knowledge' is imbued with `relational properties' (ibid: 287, citing Marriott 1989: 2). To borrow Appadurai's language, what we have here are instances of the `production of locality' (1997). The operation of tanks, the working of soils and the growing of crops do not only produce material outcomes, they are also `moments in a general technology of localization', `producing reliably local subjects as well as ... reliably local environments' (ibid: 183). They are part of the socialization of space and the `production of locality', both in itself and in opposition to other localities as social and ecological places. Development policy and projects are also in the business of the `produc- tion of locality'. This is amply evident in the emphasis on `local knowledge', and `local institutions' in programme planning and implementation. How- ever, the point that I have tried to emphasize is that this external production of locality has quite different purposes and products. In particular (as I have argued elsewhere - Mosse, 1996), in largely unacknowledged ways, the development programmes produce a distinctive type of `locality' (whether in theoretical model, project design, participatory plans, or new institutions), one which has to satisfy the operational demands of agencies, whether government departments, NGOs or their international funders. One of the most critical issues in any programme of common property resources development, then, has to do with the articulation of these competing definitions of locality.

SOME PRACTICAL IMPLICATIONS

We can now return to the point where this discussion began. There is today considerable interest in promoting new formal institutions to rehabilitate tanks and enhance their management. These are usually viewed as financially self-supporting corporate organizations which, if given unambiguous rights, would be able sustainably to manage tank resources - water, fish, trees - and enhance community involvement in system maintenance. What lessons can be learned from the analysis of successful indigenous systems in Ramnad and Sivagangai? Overall, considerable caution is necessary in extending lessons from indigenous systems to new institutions. As indicated, the difference between the two in terms of structure, purpose and meaning may be greatly under- estimated. Firstly, the indigenous tank systems discussed here do not involve corporate organizations. The existence of traditions of collective action may not, therefore, predict the success of new formal asset-holding organizations demanding procedures for public accountability or financial management. In fact, water allocation in mai kalanatu is managed very effectively without such bodies. The transaction costs of corporate organization are high and the benefits uncertain. This is precisely why mankalanatu villagers consistently History, Ecology and Locality in Tank-irrigated South India 499 avoid the accumulation of public funds, demanding planning and account- ability. Secondly, successful co-operation in water sharing does not imply collective responsibility for managing the wider resources of the tank (trees, fish, silt). Not only is fund management avoided in the indigenous systems discussed here, but villagers conspicuously avoid investment in tanks as common productive resources. The moral claims on commons resources for expenditure on temples, festivals and other culturally defined public goods is high. This penalizes collective investments for production and militates against collective entrepreneurial activity (cf. Platteau, n.d.). The point is that resource mobilization is linked to investment in system maintenance or income generation in culturally specific ways.t5 Thirdly, indigenous institutions are characterized by transactions which emphasize the social identity, position and status of actors. They are not between free and independent `appropriators'. The organization of new democratic `societies' based on the principle that all individuals, irrespective of identity, have equal rights is often in contradiction with the very logic of social relations in indigenous systems' (Platteau, n.d.). Given that `rules' of resource use are the product of social relationships, new rules isolated from old interests are unlikely to achieve authoritative enforcement simply because publicly agreed. Fourthly, and relatedly, existing systems of water use are supported by structures of authority and not democratic decision making, and often (through locally specific idioms) serve symbolically to express and reproduce this authority and the accompanying gender and caste exclusions. Attempts to establish alternative (democratic and accountable) institutions often result in: (1) the co-option of new institutions by local leaders or the reproduction of existing (caste and gender) structures inside new organizations; (2) the dismissal or side-lining of new organizational forms; or (3) the undermining and disruption of new organizations through powerful local resistance. As documented cases illustrate, because new institutions are usually the focus of outsider attention and potentially involve new resources (material and symbolic), and because they therefore expand the village public domain, there is often more at stake in their control and they are therefore more often the subject of more prolonged dispute (Mosse, forthcoming). Fifthly, external agencies and programmes never confront a static consensual `tradition'. Rather, they `interact with already contested domains of power and meaning' (Li, 1996: 515). Institutional innovation in tank systems takes place in the context of change, where resource flows expressing

15. S. Parthasarathy recently pointed out to me that in Gujarati villages tank resources had readily been taken as opportunities for entrepreneurial activity. The Tamil tanks discussed here carry distinctive cultural `baggage' which appears to make this more difficult. 500 David Mosse existing authority are breaking down and where new, more democratic arrangements are sought. Indeed, external agencies promoting new associ- ations based on principles of equity and democracy often (unintentionally) precipitate such crises in local authority as well as providing sources of support for subordinate groups and offering new institutional solutions. It sometimes happens that new irrigation organizations become intensely political, providing the context for strategies to challenge as well as retain caste power, which in some cases tie up development efforts in extended litigation (see Mosse, forthcoming). Sixthly, and more optimistically, social processes disrupting authoritative structures may encourage new forms of `enlightened', accountable and disinterested leadership compatible with new patterns of resource use. However, the point is that it is often the decline and erosion of indigenous forms of collective action which actually makes the appearance of new forms of association possible. Contrary to common assumption, successful co- operative institutions are not then the expression of enduring institutions of village self-government but a sign of the replacement of indigenous author- itative control. Seventhly, I have repeatedly pointed out that tank management has not, historically, been an autonomous village affair. In new water users' associ- ations and irrigation management transfer, also, the role of external agencies is important and frequently overlooked. Collective action theory persistently constructs as spontaneous and self-supporting, institutional solutions which are in reality conceived, imposed, and in most cases sustained by substantial external authority and resources. Moreover, farmers themselves rarely see water users' associations as meeting a demand for autonomy or isolation from the state. The meaning and purpose of new associations for the farmers concerned often has to do with new linkage to external authority. `There is not a demand for a reduced state but for a better state, one which is more responsive to their needs' (Li, 1996: 522). To conclude, there is today, as Bardhan points out, a `difficult transition' in which `traditional' co-operative institutions are on the decline while participatory alternatives have yet to take root (1993: 91). In this context, external development interventions of all kinds offer material and symbolic resources for use in the on-going re-negotiation of social relations. In this sense new institutions, tank societies and others, serve political purposes, being used, especially by subordinate groups, to acquire resources (from programmes or NGOs), to lobby and forge external links as much as for the management of village resources. The best and most reflective of recent tank development efforts go beyond the idea of tradition and begin to examine the dynamic and contested meanings of community underlying resource use in Tamil Nadu today; and then strategically support subordinate groups to enhance access to, and control of water resources, taking `operational clues' from on-going struggles, knowledge and strategies (Li, 1996: 515) rather than constructing steady-state models of co-operation. History, Ecology and Locality in Tank-irrigated South India 501

ACKNOWLEDGEMENTS

This article is a preliminary attempt to order a large and unwieldy body of research material from oral history, village surveys, field and crop surveys, and archival records. Some of the empirical and historical material is presented more fully elsewhere (Mosse, 1997c). The paper draws on fieldwork conducted in 1993-6. The research was supported by a Fellowship from the Economic and Social Research Council under the Global Environmental Change Programme (Grant No. L320273065) and a grant from the Ford Foundation (New Delhi). I am grateful for support from the Madras Institute of Development Studies (MIDS), for field assistance from Mr M. Sivan (Farmer, `Alapuram' village, Pasumponmuthuramalingam Tevar district, Tamil Nadu). I am also grateful for comments on different versions of this material at seminars and conferences at MIDS, and the universities of Edinburgh and Linkoping (Sweden). Transliteration in the text follows the conventions of the Madras University Tamil Lexicon.

REFERENCES

Appadurai, A. (1981) Worship and Conflict under Colonial Rule. Cambridge: Cambridge University Press. Appadurai, A. (1997) Modernity at Large: Cultural Dimensions of Globalization. Delhi: Oxford University Press. Baker, C. J. (1984) An Indian Rural Economy 1880-1955: The Tamilnad Countryside. Oxford: Clarendon Press. Bardhan, P. (1993) `Symposium on Management of Local Commons', Journal of Economic Perspectives 7(4): 87-92. Bardhan, P. (1994) `Rational Fools and Cooperation in a Poor Hydraulic Economy', in K. Basu et al. (eds) Choice, Welfare and Development, pp. 169-81. Oxford: Clarendon Press. Benton, T. (1994) `Biology and Social Theory in the Environment Debate' in M. Redclift and T. Benton (eds) Social Theory and the Global Environment, pp. 28-50. London and New York: Routledge. Blair, H. W. (1996) `Democracy, Equity and Common Property Resources Management in the Indian Subcontinent', Development and Change 27(3): 475-99. Bromley, Daniel W. and Michael M. Cernea (1989) `The Management of Common Property Natural Resources: Some conceptual and operational fallacies'. World Bank Discussion Papers, No. 57, Washington, DC: The World Bank. Bourdieu, P. (1977) Outline of a Theory of Practice (Trans. R. Nice). Cambridge: Cambridge University Press. Croll, E. and D. Parkin (1992) `Anthropology, the Environment and Development' in E. Croll and D. Parkin (eds) Bush Base: Forest Farm - Culture, environment and development. pp. 3-10. London and New York: Routledge. CWR (1990) Annual Report (1988-89) of the Ford Foundation funded research project `Alternative Approaches to Tank Rehabilitation and Management - A Proposed Experi- ment'. Madras: Anna University, Centre for Water Resources. Daniel, E. V. (1984). Fluid Signs: Being a person the Tamil Hay. Berkeley, CA: University of California Press. Douglas, M. (1986) How Institutions Think. London: Routledge and Kegan Paul. 502 David Mosse

Dirks, N. (1987) The Hollow Crown: Ethnohistory of a south Indian little kingdom. Cambridge: Cambridge University Press. Elster, J. (1989) 'Social Norms and Economic Theory', Journal of Economic Perspectives 3(4) 99-117. Ferguson, J. (1994) The Anti-Politics Machine: 'Development', depoliticization and bureaucratic power in Lesotho. Minneapolis, MN: University of Minnesota Press. Gadgil, Madhav and Ramachandra Guha (1992) This Fissured Land. An Ecological History of India. Delhi: Oxford University Press. Gadgil, Madhav and Ramachandra Guha (1994) Ecology and Equity: The Use and Abuse of Nature in Contemporary India. London and New York: Routledge. Granda, P. A. (1984) 'Property Rights and Land Control in Tamil Nadu: 1350-1600'. PhD Dissertation, University of Michigan. Guha, R. (1991) The Unquiet Woods: Ecological change and peasant resistance in the Himalaya. New Delhi: Oxford University Press. Hardin, G. (1968) 'The Tragedy of the Commons', Science 162: 1243-8. Jacobs, M. (1994) 'The Limits to Neoclassicism: Towards an institutional environmental economics' in M. Redclift and T. Benton (eds) Social Theory and the Global Environment, pp. 67-91. London and New York: Routledge. Janakarajan, S. (1989) 'Characteristics and Functioning of Traditional Irrigation Institutions', in Management of Renewable Resources, Special section of Bulletin, Madras Development Seminar Series XIX (12): 81-101. Janakarajan, S. (1991a) 'In Search of Tanks: Some hidden facts'. Working Paper No. 97. Madras: Madras Institute of Development Studies. Jankarajan, S. (1991b) 'The Extent and Impact of Ground Water Irrigation in Tamil Nadu: Some macro and micro evidences'. Working Paper No. 102. Madras: Madras Institute of Development Studies. Krishnan, V. and M. Mohanaraja (1995) 'Social Barriers in Practising Traditional Water Management in Tank Irrigation Systems', in N. V. Pundarikanthan and L. Jayasekhar Proceedings of the National Workshop on Traditional Water Management for Tanks and Ponds, pp. 225-35. Madras: Anna University. Leach, E. R. (1961) Pul Eliya. Cambridge: Cambridge University Press. Li, T. M. (1996) 'Images of Community: Discourse and Strategy in Property Relations', Development and Change 27(3): 501-527. Ludden, D. (1978) 'Ecological Zones and the Cultural Economy of Irrigation in Southern Tamilnadu', South Asia 1(1): 1-13. Ludden, D. (1985) Peasant History in South India. Princeton, NJ: Princeton University Press. Ludden, D. (1993) 'Orientalist Empiricism: Transformations of colonial knowledge', in C. A. Breckenridge and P. van de Veer (eds), Orientalism and the Post-Colonial Predicament: Perspectives on South Asia, pp. 250-78, Philadelphia, PA: University of Pennsylvania Press. Malony, C. and K. V. Raju (1994) Managing Irrigation Together: Practice and policy in India. New Delhi: Sage Publications. Meinzen-Dick, R., M. Mendoza, L. Sadoulet, G. Abiad-Shields and A. Subramanian (1994) 'Sustainable Water Users Associations: Lessons from a literature review'. Paper presented for World Bank Water Resources Seminar (13-15 December). MIDS (Madras Institute of Development Studies) (1986) Tank Irrigation in Tamil Nadu: Some Macro and Micro Perspectives. Madras: Madras Institute of Development Studies. Mosse, D. (1996) 'The Social Construction of "People's Knowledge" in Participatory Rural Development', in S. Bastian and N. Bastian (eds) Assessing Participation: A Debate from South Asia, pp. 135-80. New Delhi: Konark Publishers. Mosse, D. (1997a) 'Colonial and Contemporary Ideologies of Community Management: The Case of tank irrigation development in south India', Modern Asian Studies 13. Mosse, D. (1997b) 'Honour, Caste and Conflict: The Ethnohistory of a Catholic festival in rural Tamil Nadu (1730-1990)', in Christianisme et Islam en Milieu Hindou, in Collection Purusartha No. 19. Paris: Ecole des Hautes Etudes en Sciences Sociales. History, Ecology and Locality in Tank-irrigated South India 503

Mosse, D. (1997c) `Ecological Zones and the Culture of Collective Action: The History and social organization of a tank irrigation system in Tamil Nadu', South Indian Studies 3. Mosse, D. (forthcoming) `The Ideology and Politics of Community Participation: Tank irrigation development in colonial and contemporary Tamil Nadu', in R. L. Stirrat and R. Grillo (eds) Discourse of Development: Anthropological Perspectives. Olson, M. (1965) The Logic of Collective Action. Cambridge, MA: Harvard University Press. von Oppen, M. and K. Subba Rao (1980a) `Tank Irrigation in Semi-arid Tropical India. Part 1: Historical development and spatial distribution'. Progress Report 5, Economics Programme. International Crops Research Institute for the Semi-Arid Tropics (ICRISAT), Hyderabad, India. von Oppen, M. and K. Subba Rao (1980b) `Tank Irrigation in Semi-arid Tropical India. Part 2: Technical features and economic performance'. Progress Report 8, Economics Programme. International Crops Research Institute for the Semi-Arid Tropics (ICRISAT), Hyderabad, India. Ostrom, E. (1990) Governing the Commons: The Evolution of institutions for collective action. Cambridge: Cambridge University Press. Ostrom, E. (1992) Crafting Institutions for Self-governing Irrigation Systems. San Francisco, CA: Institute for Contemporary Studies Press. Ostrom, E., R. Gardner and J. Walker (1994) Rules, Games and Common-Pool Resources. Ann Arbor, MI: University of Michigan Press. Palmer-Jones, R. (1995) 'Water Markets and Water Management in South Asia', paper presented at International Conference on The Political Economy of Water in South Asia, Joint Committee on South Asia Social Science Research Council American Council of Learned Societies and Madras Institute of Development Studies (5-8 January). Platteau, J-P. (n.d.) 'The Disappointing Performance of Rural Cooperative Organizations in the Third World: A Diagnosis inferred from the "New Institutional Economics"', Namur, Belgium: University of Namur (mimeo). Popkin, S. L. (1979) The Rational Peasant: The Political Economy of Rural Society in Vietnam. Berkeley and Los Angeles, CA: University of California Press. Price, P. G. (1979) 'Resources and Rule in Zamindari South India, 1802-1903: Sivagangai and Ramnad as kingdoms under the Raj', PhD Dissertation, Wisconsin. Price, P. G. (1996) Kingship and Political Practice in Colonial India. Cambridge: Cambridge University Press. Rajaram Row, T. (1891) Manual of the Ramnad Samastanam. Madurai: Cleghorn Press. Rajivan, K. (1991) 'Irrigation Institutions and Technical Change: A Case Study from south India'. PhD Dissertation, University of South California. Redclift, M. and T. Benton (1994) 'Introduction', in M. Redclift and T. Benton (eds) Social Theory and the Global Environment, pp. 1-27. London and New York: Routledge. Runge, C. F. (1986) 'Common Property and Collective Action in Economic Development', World Development 14(5): 623-35. Sachs, Wolfgang (ed.) (1992) The Development Dictionary: A Guide to knowledge as power. Johannesburg: Witwatersrand University Press; London and New Jersey: Zed Books Ltd. Scott, J. C. (1976) The Moral Economy of the Peasant: Subsistence and rebellion in Southeast Asia. New Haven: Yale University Press. Seabright, P. (1993) 'Managing Local Commons: Theoretical issues in incentive design', Journal of Economic Perspectives, 7(4): 113-34. Sengupta, N. (1985) 'Field Systems, Property Reform and Indigenous Irrigation'. MIDS Offprint, 3. Madras: Institute of Development Studies. Sengupta, N. (1991) Managing Common Property: Irrigation in India and the Philippines. New Delhi: Sage Publications. Sengupta, N. (1993) User-friendly Irrigation Designs. New Delhi: Sage Publications. Shankari, U. (1991) 'Major Problems in Minor Irrigation: Social change and tank irrigation in Chittor district of Andhra Pradesh', Contributions to Indian Sociology (n.s.) 25(1): 85-111. 504 David Mosse

Sivakumar, S. S. and C. Sivakumar (1996) `Organisation of Agriculture, Society and Ideology: An Essay on the meaning of social order in the Tamil Country' in P. Robb (ed.) Meanings of Agriculture: Essays in South Asian History and Economics, pp. 332-85. New Delhi: Oxford University Press. Sivasubramaniyam, K. (1995a) `Tank Irrigation Management under Different Conditions: Study of two large tanks in Tamilnadu'. International Conference on The Political Economy of Water in South Asia, MIDS (January). Sivasubramaniyam, K. (1995b) `Maintenance of Traditional Tank Irrigation Systems: Case of Kaveripakkam and Dusimamandur Tanks', in N. V. Pundarikanthan and L. Jayasekhar (eds) Proceedings of the National Workshop on Traditional Water Management for Tanks and Ponds, pp. 110-32. Madras: Anna University. Spencer, J. (1990) A Sinhala Village in a Time of Trouble: Politics and change in rural Sri Lanka. Delhi: Oxford University Press. Stein, B. (1980) Peasant State Society in Medieval South India. Delhi: Oxford University Press. Stein, B. (1985) `State Formation and Economy Reconsidered', Modern Asian Studies 19(3): 387-413. Svendsen, M. and A. Gulati (1994) Strategic Change in Indian Irrigation. New Delhi: Indian Council of Agricultural Research; Washington, DC: International Food Policy Research Institute. Tang, S. Y. (1992) Institutions and Collective Action: Self-governance in irrigation. San Francisco, CA: Institute for Contemporary Studies Press. Uphoff, Norman (1992) Learning from Gal Oya: Possibilities for Participatory Development and Post-Newtonian Social Science. Ithaca, NY, and London: Cornell University Press. Vaidyanathan, A. (1992) Strategy for Development of Tank Irrigation. Madras: Madras Institute of Development Studies. Vaidyanathan, A. and S. Janakarajan (1989) Management of Irrigation and its Effects on Productivity under Different Environmental and Technical Conditions: A Study of two surface irrigation systems in Tamil Nadu. Madras: Madras Institute of Development Studies. Vasavi, A. R. (1994) `Hybrid Times, Hybrid People: Culture and agriculture in south India', Man 29(2): 283-332. Wade, R. (1988) Village Republics: Economic conditions for collective action in south India. Cambridge: Cambridge University Press. Wade, R. (1995) `The Ecological Basis of Irrigation Institutions: East and South Asia', World Development 23(12): 2041-9. Washbrook, D. (1976) The Emergence of Provincial Politics: The Madras Presidency 1870-1920. Cambridge: Cambridge University Press. Washbrook, D. (1988) `Progress and Problems: South Asian economic and social history c. 1720-1860', Modern Asian Studies 22(1): 57-96. Zwarteveen, M. (1995) `Linking Women to the Main Canal: Gender and irrigation manage- ment'. Gatekeeper Series No. 54. International Institute for Environment and Development. Zwarteveen, M. and N. Neupane (1995) `Gender Aspects of Irrigation Management: The Chhatti Mauja irrigation system in Nepal', Asia-Pacific Journal of Rural Development 5: 1-26.

David Mosse is a social anthropologist at the Centre for Development Studies (University of Wales, Singleton Park, Swansea SA2 8PP, UK). From September 1997 he will be at the Department of Anthropology, School of Oriental and African Studies, University of London. He has lived and worked in India frequently since 1982 as a researcher and consultant in rural development. His ethnographic work in Tamil Nadu has focused on popular religion, social organization, dalit movements, and indigenous irrigation. During the last six years he has worked intensively with a participatory natural resources development project in tribal western India. CBNRM Social Science Resource Kit Institutional Analysis Resource Book C. Bibliography

This section provides a list of selected annotated/abstracted references related to institutional analysis that might be of interest to CBNRM researchers. The source of the annotations is in most cases the author or the publisher; however, some have been edited or adapted for the CBNRM researcher audience.

1. Afikorah-Danquah, S. 1997. Local Resource Management in the Forest-Savanna Transition Zone: The Case of Werenchi District, Ghana. IDS Bulletin 28 (4): 36-46.

Abstract: An institutional analysis is at the core of the environmental entitlements approach. In this case study from Ghana, researchers gained a better understanding of the institutions which mediated indigenous inhabitants' and immigrants access to and control over forest and agricultural resources, finding that the combination and interaction of these institutions was at the heart of the conflict over visions of appropriate environmental management. With a more complete understanding of institutions and organizations, some of the issues surrounding conflicts over resource use were more easily uncovered, and a range of resource conflicts between immigrants and indigenous people were found to have an institutional origin (IDS).

2. Agrawal, Arun, Charla Britt and Keshav Kanel. 1998. Decentralization in Nepal: A Comparative Analysis. A Report on the Participatory Development Program. San Francisco: Institute for Contemporary Studies Press.

Abstract: In the past two decades, decentralization has gained increasing attention as a necessary and valuable instrument for enhancing efficiency and promoting democratic participation-yet progress has been slow. A detailed analysis of the Participatory Development Program in Nepal, this book identifies conditions vital to the success of decentralization and shows how they may be fostered. At the same time, it skillfully places the program in a comparative perspective: by examining the political economy of previous decentralization efforts, Agrawal and his colleagues identify the impediments that decentralization has faced in so many countries. This even-handed study has important implications for projects of decentralization not only in Nepal, but also in developed and developing countries around the world (ICS Press). CBNRM Social Science Resource Kit Institutional Analysis Resource Book

3. Agrawal, A. 1998. Group Size and Successful Collective Action: A Case Study of Forest Management Institutions in the Indian Himalayas. In Forest Resources and Institutions. Gibson, C., McKean, M.A., Ostrom, E. (Eds.). Forest, Trees and People Programme. Working Paper No. 3. FAO: Indiana University.

Abstract: Agrawal investigates how local-level variation within the Indian van panchayat (forest council) system of community forestry leads to substantially different outcomes for the management of forest resources. He begins by analyzing the legislation which underlies the van panchayat system, and then demonstrates how local factors-such as size, organization, age and resource endowments-affect management success. He concludes that the size of the van panchayat is an important factor, with very small councils being at a disadvantage over moderate-sized ones in their efforts to generate sufficient human and other resources to monitor and enforce local rules (eds.).

4. Agrawal, A. 1997. Forest Management Under Common Property Regimes in the Kumaon Himalaya. In People and Participation in Sustainable Development. G. Shivakoti et al (eds.). Proceedings of an International Conference held at the Institute of Agriculture and Animal Science, Rampur, Chitwan, Nepal, 17-21 March 1996. Bloomington, Indiana and Rampur, Chitwan.

Abstract: This paper explores local level forest use and management in the Kumaon Himalaya. It seeks to situate the ongoing research on forest resource use in the Kumaon Himalaya in the context of a larger conversation on common property use and management in mountain regions. More specifically, it examines why it is necessary to look at communities as relevant units of social organization for understanding resource use; the need to analyze the effects on resource use of stratification and differences within communities; and the importance of subjecting concepts such as "community", "local", and "indigenous" to further reflection and refinement (author).

5. Alston, Lee J., Thrainn Eggertsson, and Douglass C. North, eds. 1996. Empirical Studies in Institutional Change. Cambridge University Press.

Abstract: Book is a collection of nine empirical studies by fourteen scholars. Dealing with issues ranging from the evolution of secure markets in seventeenth-century England to the origins of property rights in airport slots in modern America, the contributors analyse institutions and institutional change in various parts of the world and at various periods of time. The volume is a contribution to the new economics of institutions, which emphasises the role of transaction costs and property rights in shaping incentives and results in the economic arena. To make the papers accessible to a wide audience, including students of economics and other social sciences, the editors have written an introduction to each study and added three theoretical essays to

C-2 CBNRM Social Science Resource Kit Institutional Analysis Resource Book the volume, including Douglass North's Nobel Prize address, which reflect their collective views as to the present status of institutional analysis and where it is headed

(Cambridge University Press) .

Chapter Contents: Introduction; A note on theory, Thrainn Eggertsson; Empirical work in institutional economics, Lee J. Alston; 1. Towards an understanding of property rights, Gary D. Libecap; 2. Impediments to institutional change in the Soviet system, Jan Winiecki; 3. Transaction costs and economic development, Andrew Stone, Brian Levy and Ricardo Paredes; 4. The evolution of the modern institutions of growth, Douglass C. North and Barry R. Weingast; 5. Regulation in a dynamic setting, Anne O. Krueger; 6. Price controls, property rights and institutional change, Steven N. S. Cheung; 7. Regulating natural resources: the evolution of perverse property rights, Robert Higgs; 8. The politics of institutional change in a representative democracy, William H. Riker and Itai Sened; 9. The economics and politics of institutional change, Lee J. Alston and Joseph P. Ferrie; Epilogue: Economic performance through time, Douglass C. North.

6. Bandaragoda, D.J. 1998. Design and practice of water allocation rules: Lessons from Warabandi in Pakistan's Punjab. Research report no. 17. International Irrigation Management Institute (IIMI ). Colombo.

Abstract: The conventional wisdom of the application of water allocation rules in irrigation systems is rarely questioned. The major reason for this complacence is a lack of interest in considering water distribution as a dynamic socio-technical process. A study on a sample of 22 irrigation subsystems (water-courses) located in 6 canals in 3 different major canal systems of Pakistan's Punjab Province clearly showed that the actual practice of warabandi water allocation schedules differed substantially from the design. This report focuses on the existing gap between the original design of warabandi as a water allocation method and the way it is practiced in Pakistan today (author).

7. Becker, C. and Ostrom, E. 1995. Human Ecology and Resource Sustainability: The Importance of Institutional Diversity. Annual Review of Ecological Systems 26, pp. 131

Abstract: The authors define the concept of a common-pool resource and present similarities and differences among common-pool resources in regard to their ecological and institutional significance. The design principles that characterize long-surviving, delicately balanced resource systems governed by local rules systems are discussed, as is a synthesis of the research on factors affecting institutional change. Initial findings from pilot studies in Uganda related to the effects of institutions on forest conditions are presented as well (author-adapted).

C-3 CBNRM Social Science Resource Kit Institutional Analysis Resource Book

8. Borrini-Feyerabend, G. 1996. Collaborative Management of Protected Areas: Tailoring the Approach to the Context. IUCN, Gland, Switzerland.

Abstract: 'Collaborative management' is a partnership where various stakeholders agree on sharing the management functions, rights and responsibilities for a territory or a set of resources under protected areas. This publication addresses conservation professionals-in particular governmental agency staff-offering a broad definition of the approach and provides examples of how it has been specifically tailored to different contexts. General assumptions, consequences, benefits, costs and potential drawbacks of collaborative management are reviewed and a process by which an agency in charge of a protected area can pursue the approach is illustrated (IUCN).

9. Bromley, Daniel et al. (eds). 1992. Making the Commons Work: Theory, Practice, and Policy. San Francisco: Institute for Contemporary Studies Press.

Abstract: The contributors to this book take an approach which emphasizes not the tragedy but the possibilities of the commons, both in theory and in practice. Drawing on case studies from countries around the world, they develop a new set of constructs to explain how small-scale common property systems can be successful and durable. The book shows that people on the local level can handle complex social and economic issues successfully and equitably. Often, economic aid to developing countries has ignored these small-scale but effective local institutions, mistakenly relying on privatization or state ownwership to solve development and natural resource problems (R. Hawkins).

10. Brown, D. and Korte, C. 1997. Institutional Development of Local Organizations in the Context of Farmer-Led Extension: the Agroforestry Programme of the Mag'uugmad Foundation. Network Paper No. 68, Agricultural Research and Extension Network, UK Overseas Development Institute (ODI).

Abstract: This paper reviews the attempts of the NGO, World Neighbors, and its indigenous successor, the Mag'uugmad Foundation Inc., to develop the local institutional capacity of farmer organizations on the island of Cebu in the Philippines, building on a highly successful programme of farmer-led extension of agroforestry technologies. The paper describes the history of the programme from its inception in 1982, and considers the grounds for its success in the transformation of the farming system over a wide area of the Cebu uplands. It then examines progress to date in the area of local institutional development, and assesses the likelihood of the substantial transfer of management functions to the community (authors).

C-4 CBNRM Social Science Resource Kit Institutional Analysis Resource Book

11. Carney, D. 1996. Formal Farmers' Organizations in the Agricultural Technology System: Current Roles and Future Challenges. Natural Resource Perspectives, Number 14, November. London, U.K. Overseas Development Institute (ODI).

Abstract: Farmers organizations (FOs) need to be relatively sophisticated and well-funded to become involved in agricultural technology development and transfer. This is because of the complexity of understanding members technological needs and of building productive partnerships with other technology suppliers. Only small-scale initiatives are likely to be possible for organizations which have limited capacity. Such initiatives usually rely on leaders existing knowledge of or access to improved technologies. Organizations with relatively homogenous membership and with close links to the market (which helps both to set quality standards and to generate money for the organization itself) are generally better able to get involved in technology than their larger, more political counterparts. The attitude of the public and private technology suppliers is also likely to be a critical factor in determining whether farmers organizations will be successful in their technology-related activities, as is the support of donors and/or NGOs on the capacity-building and financial sides (author).

12. Centre for Economic Development and Administration. 1991. Inventory of Rural Institutions in Rapti Zone Nepal. Kirtipur, Kathmandu, Nepal: Center for Economic Development and Administration, Tribhuvan University.

13. Chettri, R.B. and Pandey, T. 1992. User Group Forestry in the Far-Western Region of Nepal. International Centre for Integrated Mountain Development Monograph, Kathmandu, Nepal.

14. Clague, C. (ed). 1997. Institutions and Economic Development: Growth and Governance in Less-Developed and Post-Socialist Countries. Baltimore: Johns Hopkins University Press. (The Johns Hopkins Studies in Development).

Abstract: The puzzles of economic development and post-communist transitions, according to Christopher Clague and his colleagues, can be illuminated by a serious economic analysis of institutions. Economic performance is strongly dependent on the economic policies selected and on the manner in which these policies are implemented by government agencies. Performance is also affected by property rights and contract enforcement mechanisms in the business community and by patterns of participation in community organizations. These and other institutional arrangements are analyzed in this book under the rubric of the New Institutional Economics. Topics include the effects of democratic political institutions on economic performance, the determinants of success or failure in community organization, the institutional challenges facing formerly communist societies, and the use of the economics of information to improve

C-5 CBNRM Social Science Resource Kit InstitutionalAnalysis Resource Book government administrative performance. Contributors include economists Christopher Clague, Robert Klitgaard, Peter Murrell, Mancur Olson, Vernon Ruttan, and Vito Tanzi, and political scientists Stephan Haggard, Margaret Levi, and Elinor Ostrom (Johns Hopkins University Press).

15. Cortner, H.J. 1998. Institutions Matter: The Need to Address the Institutional Challenges of Ecosystem Management. Landscape and Urban Planning 40 (1-3):159-166.

Abstract: As management philosophy focused on sustainability, ecosystem management calls for a reassessment of how we approach nature, science and politics. Implementation of ecosystem management will require changes in society's institutions. However, the exact requirements for institutional change are unclear. Therefore, in order to move toward implementing ecosystem management, a better understanding of the relationships between institutions and natural resource management will be required. The purpose of this paper is to examine institutional barriers and incentives to ecosystem management and to encourage a dialogue on this subject among resource managers, the public, and researchers. To this end, the paper identifies five problem areas where additional understanding of the institutional requirements for implementing ecosystem management is needed (author).

16. Cramb, R. A., and I. R. Wills. 1990. The Role of Traditional Institutions in Rural Development: Community-Based Land Tenure and Government Land Policy in Sarawak, Malaysia. World Development 18(3):347-360.

Abstract: Traditional local-level institutions are frequently considered obstacles to rural development and so attempts are made by the state to impose "dissonant" institutional forms from above. In contrast, this paper argues that traditional institutions should be viewed as the building blocks of a modern, development-oriented institutional structure. The argument is applied to the case of the Than system of community-based land tenure and its relationship with government land policy in the Malaysian state of Sarawak (author).

17. Edwards, V.M., Steins, N.A. 1998. Developing an Analytical Framework for Multiple-Use Commons. Journal of Theoretical Politics 10 (3): 347-383.

Abstract: This paper extends the analytical framework put forward by Oakerson for application to multiple-use pools, where multiple types of use are made of the resource system. Four components are introduced: (1) multiple-use analysis of physical and technical attributes; (2) multilevel analysis of decision-making arrangements; (3) social characteristics of the broad user community; and (4) analysis of contextual factors. The multiple-use framework facilitates understanding of multiple-use commons in a chosen time period and institutional change over time (author).

C-6 CBNRM Social Science Resource Kit Institutional Analysis Resource Book

18. FAO; SEAGA. 1998. Field Handbook: Socioeconomic and Gender Analysis Programme. Rome: FAO. Available on-line at http://www.fao.org/sd/seaga/SEfh000l.htm

Notes: This SEAGA Field Handbook is written for development agents who work directly with local communities in developing countries. It is intended for outsiders such as extensionists, government and non-government field workers, and private- and public-sector development consultants, and for insiders such as community organisers and leaders of local groups and institutions. The purpose of this Handbook is to support participatory development planning at the community level. It is based on actual experiences in agriculture, forestry and fisheries, but can be used by those working in all sectors of rural development. The Handbook offers three toolkits: Toolkit A The Development Context is for learning about the economic, environmental, social and institutional patterns that pose supports or constraints for development; Toolkit B Livelihood Analysis is for learning about the flow of activities and resources through which different people make their living; and Toolkit C. Stakeholders' Priorities for Development is for planning development activities based on women's and men's priorities. The first two focus on learning about the current situation ("what is"), while the third focuses on planning for the future ("what should be"). Each toolkit is designed to answer important questions. (In Chapter Nine several additional tools are provided to facilitate adaptation of these toolkits, as needed) (FAO).

19. Feeny, D. 1994. Frameworks for Understanding Resource Management on the Commons. Pp. 20-33 in Community Management and Common Property of Coastal Fisheries in Asia and the Pacific: Concepts, Methods and Experiences. R. S. Pomeroy, ed. Manila, Philippines: International Center for Living Aquatic Resources Management.

Abstract: Author reviews a number of frameworks, both conventional and more recent, for understanding resource management in the context of common property. He emphasizes that recent literature in economics has recognized a fourth category or pillar of variables for describing economic systems-institutions. Together with resource endowments, preferences and technology, institutions, the "rules of the game", shape the possibilities for economic activity. Institutional arrangements, in general, and property rights, in particular, form the nature of resource use and management of the commons (ed.).

20. Fisher, R. J. 1989. Indigenous Institutions and Organizations in the Management of Common Property Forest Resources in Nepal. Environment and Policy Institute, East-West Center, Honolulu, HI. (EAPI Working Paper, no.18).

C-7 CBNRM Social Science Resource Kit Institutional Analysis Resource Book

21. Fisher, R. J. 1991. Studying Indigenous Forest Management Systems in Nepal: Towards a More Systematic Approach. EAPI Working paper no. 30. Honolulu, Hawaii: East-West Center.

22. Gibson, C., McKean, M.A., Ostrom, E. (Eds.). 1998. Forest Resources and Institutions. Forest, Trees and People Programme. Working Paper No. 3. FAO: Indiana University.

Notes: In this Working Paper the authors seek to understand the complex interactions between local communities and their forests. To do so, they depart significantly from conventional national-level analyzes and offer groundbreaking efforts to identify the relationship between forest conditions, individuals, and institutions at a local level. The presumption that guides the authors is that institutions at the local level--together with the incentives and behaviors they generate--lie at the heart of explanations of forest use and condition. Case material covers Asia, South America and Africa (eds.).

Contents: (1) Explaining Deforestation: The Role of Local Institutions; (2) Common Property: What is it, What is it Good for, and What Makes it Work?; (3) Group Size and Successful Collective Action: A Case Study of Forest Management Institutions in the Indian Himalayas; (4) Successful Forest Management: The Importance of Security of Tenure and Rule Enforcement in Ugandan Forests; (5) Social Norms and Human Foraging: An Investigation into the Spatial Distribution of Shorea robusta in Nepal; (6) The Lack of Institutional Supply: Why a Strong Local Community in Western Ecuador Fails to Protect its Forest; (7) Indigenous Forest Management in the Bolivian Amazon: Lessons from the Yuracare People; (8) Coping with Changes in Population and Forest Resources: Institutional Mediation in the Middle Hills of Nepal; Appendix: IFRI Research Strategy.

23. Gupta, A.K. Rethinking Policy Options for Watershed Management by Local Communities: Combining Equity, Efficiency and Ecological-Economic Viability. Available on-line from the Society for Research and Initiatives for Sustainable Technology and Institutions (SRISTI) website at http://csf.Colorado.EDU/sristi/papers/rethinkin.html

Abstract. In this paper, I argue for certain basic re-thinking in the policy options for viable watershed management by combining local knowledge with formal science through rejuvenated or revitalized traditional institutions. In part one, I review the policy environment in light of some of the recent reports in India which have a major bearing on watershed development programs. In part two, I discuss the theory of portfolio options which can provide an effective alternative to the current approaches to watershed management. In part three, I discuss various policy changes in research,

C-8 CBNRM Social Science Resource Kit Institutional Analysis Resource Book public administration, decentralized system of self governance, and interface with voluntary organizations and people's institutions. I conclude that large scale efforts in restoration of productivity of eroded regions have to be appreciated without ignoring the fact that spreading resources thinly may give political advantage but would not generate any durable change in the resource management situation (author).

24. Gupta, A.K. 1994. Challenges in Developing Indigenous Theories of Organization and Management: An Indian Perspective. Indian Journal of Social Work, April Vol.55 No.2 Pages 220-236

Abstract: Most of the efforts around the world in the area of OB have been to discover universal theories of organization and management. Now increasing attention is being paid to culture, society or nation specific theories. This paper advocates such an approach for the study of organizations in India. It delineates some major theoretical and methodological challenges in undertaking this approach and calls for the need to have a committed academic community (author).

25. Hanna, Susan, and Mohan Munasinghe (eds.). 1995. Property rights in a social and ecological context : case studies and design applications. Washington, D.C.: World Bank. 206 p.

Abstract: This publication contains a collection of case studies and design applications from around the world that focus on the institutional dimensions of environmental sustainability. The knowledge of how property rights regimes, as particularly important types of institutions, function in relation to humans and their use of the environment is critical to the design and implementation of effective environmental protection. These studies address questions of the design of governance systems for sustainability; the relationships among equity, stewardship, and environmental resilience; the use of traditional knowledge in resource management; the mechanisms that link humans to their environments; and the role played by poverty and population (World Bank).

Contents: An introduction to property rights in a social and ecological context (Hanna,

Munasinghe); Design lessons from existing air pollution control systems : the United States (Tietenberg); Distributed governance in the Northwestern Hawaiian Islands lobster fishery (Townsend, Pooley); Enforcement of regional environmental regulations: Nitrogen fertilizer in Sweden (Gren, Brannlund); Designing incentives to conserve India's biodiversity (Gadgil, Rao); Will new property right regimes in Central and Eastern Europe serve the purposes of nature conservation? (Zylicz); Nonsustainable use of renewable resources : mangrove deforestation and mariculture in Ecuador (Parks, Bonifaz); Learning by fishing: practical science and scientific practice (Palsson);

Indigenous knowledge and resource management systems : a Native Canadian case

C-9 CBNRM Social Science Resource Kit Institutional Analysis Resource Book study from James Bay (Berkes); The role of validated local knowledge in the restoration of fisheries property rights: the example of the New Zealand Maori (Ruddle); The role of tenurial shells in ecological sustainability: property rights and natural resource management in Mexico (Alcorn, Toledo); Integrating ecological and socio-economic feedbacks for sustainable fisheries (Hammer); Parametric management of fisheries: an ecosystem - social approach (Wilson, Dickie); Environmental and socioeconomic linkages of deforestation and forest land use change in the Nepal Himalaya (Pradhan,

Parks); Environmental crisis and unsustainability in Himalayas : lessons from the degradation process (Jodha).

26. Harris, J., J. Hunter and C.M. Lewis (eds.). 1995. The New Institutional Economics and Third World Development. London: Routledge.

Abstract: The new institutional economics is the most important new body of theory to emerge in economics in recent years. The contributors to this volume address its significance for the developing world. The book is a major contribution to an area of debate still in its formative phase. The book challenges the orthodoxies of development, especially concerning the role of markets (publisher).

27. Hobley, M. and Shah, K. 1996. What Makes a Local Organization Robust? Evidence From India and Nepal. Natural Resource Perspective, Number 11. London, U.K. Overseas Development Institute (ODI).

Abstract: Authors analyze the factors that contribute to the effectiveness of local organizations as resource managers, using their experience in India and Nepal with the development of local forest management organizations to illustrate their argument. They outline gaps in present knowledge and conclude with a discussion of the implications for policy and practice.

28. Kaul, Minoti Chakravarty. 1996. Common Lands and Customary Law: Institutional Change in North India over the Past Two Centuries. Oxford University Press.

Abstract: This book seeks to dispel the notion that communally- held resources are necessarily open to private exploitation and misuse. It argues that customary law and institutions of property rights were devised by village communities to check the misuse of common property resources (Oxford University Press).

C-10 CBNRM Social Science Resource Kit Institutional Analysis Resource Book

29. Knudsen, A.J. 1995. Living with the Commons: Local Institutions for Natural Resource Management. Chr. Michelsen Institute (CMI) Report No. 2. Bergen, Norway.

Abstract: Garrett Hardin's essay "The Tragedy of the Commons" has for almost three decades stimulated research on common property regimes. This report provides an overview of this research and reviews a selection of empirical and theoretical contributions to the "commons" debate. Despite the hectic research activity, the report is critical of the tendency to reproduce well-worn arguments instead of questioning them. In order to progress beyond a rebuttal of Hardin, the report calls for an interdisciplinary approach to the study of common property regimes and advocates an analytical focus on local institutions. In particular, the report discusses those circumstances under which local institutions represent an alternative to state management of renewable natural resources. The three themes which have been selected for closer study are coastal fisheries, rangelands management and forestry management. Regionally, case studies from Africa and Asia have been preferred over material from the rest of the world (author).

30. Lam, Wai Fung. 1998. Governing Irrigation Systems in Nepal: Institutions, Infrastructure and Collective Action. San Francisco: Institute for Contemporary Studies Press.

Abstract: This study shoots a deadly arrow into the heart of economic determinists. Dr. Lam analyzes two types of governance systems for developing rural infrastructure in Nepal: state-governed technocratic systems, and indigenous self-governing systems. He finds that farmer-governed systems have significantly higher levels of productivity: water is delivered more effectively, reliably, and equitably. Unlike conventional policy analyses which view irrigation systems largely from the technical side, this thorough and convincing study emphasizes the link between irrigation performance and how individuals work together on systems. Lam's study will be of interest to scholars, policy makers, donors, and activists throughout the world who are working to devise programs that involve the participation of farmers and villagers in rural areas. With a foreword by Dr. Elinor Ostrom (ICS Press).

31. Leach, M., Mearns, R., Scoones, I. 1997. Challenges To Community-based Sustainable Development - Dynamics, Entitlements, Institutions. IDS Bulletin 28(4):4-14.

Abstract: Recent approaches to community-based natural resource management frequently present 'communities' as consensual units, able to act collectively in restoring population-resource imbalances or reestablishing harmonious relations between local livelihoods and stable environments. Arguing that these underlying assumptions and policy narratives are flawed as guidelines for policy, this article presents an alternative

C-11 CBNRM Social Science Resource Kit Institutional Analysis Resource Book perspective which starts from a perspective which sees the politics of resource access as central among diverse social actors, and sees patterns of environmental change as the outcomes of negotiation or contestation between their conflicting perspectives. The notion of 'environmental entitlements' encapsulates this shift in perspective, and provides analytical tools to specify the benefits that people gain from the environment which contribute to their well-being. The processes by which people gain environmental endowments and entitlements are, in turn, shaped by diverse institutions, both formal and informal (IDS).

32. Leach, M., Mearns, R., Scoones, I. 1997. Institutions, Consensus And Conflict - Implications For Policy And Practice. IDS Bulletin - Institute of Development Studies 28(4):90

Abstract: This article reflects on the challenges faced when the ideal of consensual communities is questioned. A more complex view of institutional relationships at the local level is envisaged, one which emphasizes conflict as much as consensus. This, in turn, suggests some implications for institutional design and processes of conflict negotiation. A number of alternatives. are explored, ranging from targeted, institutional design to more flexible, learning process approaches. Support for effective negotiation processes is highlighted, including the enhancement of claims-making capacity through processes of participation and empowerment. Due to the inherent uncertainties in both ecological and social dynamics, institutional design can never take a blueprint form. Instead, a flexible, adaptive style of dealing with institutional complexity and uncertainty is envisaged. Despite the necessity of disagreggating 'community' imagery for local-level implementation, such imagery can also be used strategically and effectively by local people and other development actors in struggles to define and direct processes of change (IDS).

33. Loehman, E.T. and D. Marc Kilgour (eds.) 1998. Designing Institutions for Environmental and Resource Management. Cheltenham, UK: Edward Elgar Publishing.

Abstract: This book addresses environmental and resource management problems that continue to emerge despite increasing attempts at regulation. It proposes a proactive approach to environmental and resource management through the design of institutions and organizations. In addition, it suggests that social rules for environmental management can be improved by taking into account the social costs of externalities and the administrative and transactions costs to reduce them (eds.).

C-12 CBNRM Social Science Resource Kit Institutional Analysis Resource Book

Contents: 1. Introduction: Social Design for Environmental and Resource Management. PART 1: CONCEPTS FOR INSTITUTIONAL AND ORGANIZATIONAL DESIGN- 2. Issues in the Design of Mechanisms and Institutions; 3. On Coordination, Externalities and Organization; 4. The Institutional Analysis and Development Approach; 5. Solving Collective Action Problems Fairly: Puzzles in Policy Design; 6. Heterogeneity and Common Pool Resource Management; 7. Environmental Change, Institutions, and Human Capital; 8. Suboptimality and Transaction Costs on the Commons; 9. Designing Institutions for Sustainability. PART 2: QUANTITATIVE MODELS IN INSTITUTIONAL DESIGN- 10. Pollution Regulation in a Political Economy; 11. Cooperation in Pollution Reduction: Design of a Policy Instrument; 12. Citizen Lawsuits and the SLAPP in US Water Disputes; 13. Environmental Standards and Liability with Limited Assets and Private Information; 14. Efficient Enforcement Systems for Water Regulations; 15. Joint Implementation Agreements and Principal-Agent Contracts; 16. Contract Design for Public Grazing Lands. PART 3: EXPERIMENTAL METHODS IN INSTITUTIONAL DESIGN- 17. The Scope for Laboratory Experiments on Marketable Pollution Permits; 18. Instrument Choice for Emissions Permit Trading: Designing a Laboratory Environment; 19. Groundwater Institutions: Models and Experiments; 20. Resource Dilemmas under Uncertainty: Testing Nash Behavior.

34. Malik, S.M.; Waheed-uz-Zaman; Kuper, M. 1996. Farmers' organized behaviour in irrigated agriculture in Pakistan's Punjab: a case study of six watercourse command areas in Junejwala Minor, Lower Chenab Canal System. International Irrigation Management Institute (IIMI) Working Paper no. 39. Colombo (Sri Lanka)

Abstract: Increased participation of farmers in the management of Pakistan's canal irrigation system is advocated by the federal government and donors to help address the low levels of performance of irrigated agriculture in Pakistan. However, past efforts in establishing water users associations have shown that establishing sustainable, robust farmers' water management organizations is not an easy task. The aim of this paper is to assess the existing organization of irrigation activities in six tertiary units in Pakistan's Punjab and to evaluate the effect of the social setup of the villages to which these units belong on the present organization of the irrigation activities. The study exposes the fact that existing irrigation institutions (e.g. warabandi) have clear rules and boundaries that limit interactions and possible conflicts between farmers. Moreover, the number of institutions that are functional is kept very small, as organizations are dissolved when targets have been achieved (IIMI).

C-13 CBNRM Social Science Resource Kit Institutional Analysis Resource Book

35. Messerschmidt, Donald A. 1988. Community Forest Management and the Opportunities of Local Tradition: A View from Nepal. In Rural Institutions and Resource Management. A. Dani and J. G. Campbell, eds. Kathmandu, Nepal: International Centre for Integrated Mountain Development (ICIMOD).

36. Mosse, D. 1996. Local institutions and farming systems development: Thoughts from a project in tribal western India. Agren Network Paper No. 64. Agricultural Research and Extension Network, UK Overseas Development Institute (ODI).

Abstract: Many rainfed farming and watershed development programmes face problems in ensuring the long-term sustainability of activities. Comparative research into this problem has pointed to the absence of adequate farmer or community participation in activity planning, implementation and management as a critical issue. This need for greater participation is increasingly addressed though the employment of participatory research and planning methods - PRA/PALM. These methods, however, are not sufficient to ensure and sustain farmer involvement. They will only contribute to sustained participation as part of a structured planning process which includes the identification and support of local institutions which can take responsibility for implementation and the long-term management of watershed resources. The difficulty is that new local institutions may themselves be among the least viable and sustainable of project interventions. This is often because participatory ideals overlook the real social costs involved in collective action. This paper discusses the experience of the KRIBP project in evolving an approach to local institutional development in a tribal region in western India. It discusses problems and lessons learned, and suggests ways of improving performance by reducing the demand for and costs of collective action in farming systems development (author).

37. Murphree, M.W. 1994. The Role of Institutions in Community-Based Conservation. Chapter 18 in Natural Connections: Perspectives in Community- Based Conservation. Western, D. et al. (eds.). Pp. 403-427. Washington D.C. Island Press.

Abstract: Author provides a broad overview of the relationship between institutions and community-based resource management. He discusses the institutional approach, describes the various categories of institutional actors that need to be considered under such an approach, and outlines some of the key issues related to implementing institution-oriented community-based conservation schemes.

C-14 CBNRM Social Science Resource Kit Institutional Analysis Resource Book

38. Murphree, Marshall W. 1993. Communities as Resource Management Institutions. London: Sustainable Agriculture Program of the International Institute for Environment and Development, Gatekeeper Series no. 36.

39. North, Douglas C. 1990. Institutions, Institutional Change and Economic Performance. Cambridge: Cambridge University Press.

Abstract: Continuing his analysis of economic structures, Douglass North develops an analytical framework for explaining the ways in which institutions and institutional change affect the performance of economies, both at a given time and over time. North first explores the nature of institutions and explains the role of transaction and production costs in their development. The second part of the book deals with institutional change. Institutions create the incentive structure in an economy, and organisations will be created to take advantage of the opportunities provided within a given institutional framework. He then explains how institutional development may lead to a path-dependent pattern of development. In the final part of the book, North explains the implications of this analysis for economic theory and economic history. He indicates how institutional analysis must be incorporated into neo-classical theory and explores the potential for the construction of a dynamic theory of long-term economic change (Cambridge University Press).

40. Nugent, J.B. 1993. Between State, Markets and Households: A Neo- Institutional Analysis of Local Organizations and Institutions. World Development Vol.21 (April) No.4 Pages 623- 632

Abstract: The current and foreseeable trend toward a diminished role of the state in most economic activities, including rural development activities, offers new opportunities for local organizations and institutions (LOIs). At the same time. many of the problems and shortcomings to which states and their activities have been vulnerable are also potentially applicable to LOIs. To be successful in themselves as well as in fostering rural economic development, rural LOIs and their activities must be carefully designed and appropriately integrated with markets, other LOIs and the state. This paper reviews the relevant literature on conditions for success and then goes on to derive from this review (including some relevant experience) some suggestions for policy and practice (author).

C-15 CBNRM Social Science Resource Kit Institutional Analysis Resource Book

41. Oakerson, R. 1992. Analyzing the Commons: A Framework. In Bromley, Daniel et al. (eds). Making the Commons Work: Theory, Practice, and Policy. San Francisco: Institute for Contemporary Studies Press.

Abstract: Oakerson presents a conceptual framework that can be used to collect and analyze information about the management of common property resources. The framework distinguishes four sets of attributes that can be used to describe the commons: (1) physical attributes of the resource and the technology used to harvest it; (2) institutional arrangements that govern decision-making and relationships between users; (3) the resulting patterns of interaction among decision-makers; and (4) outcomes or consequences (author adapted).

42. Ostrom, Elinor. 1994. Institutional Analysis, Design Principles and Threats to Sustainable Community Governance and Management of Commons. In Community Management and Common Property of Coastal Fisheries in Asia and the Pacific: Concepts, Methods and Experiences. R. S. Pomeroy, ed. Manila, Philippines: International Center for Living Aquatic Resources Management. (Workshop Reprint Series, R94-11).

Abstract: Author contributes to the discussion of the different frameworks for understanding management of common property resources by discussing the results of a study done using a specific framework for analysis. The Institutional Analysis and Development (IAD) framework, developed by the author and her colleagues at Indiana University, focuses on the identification of contextual attributes that shape various action situations, i.e., resource management strategies. The IAD links the characteristics of the physical world with those of the general cultural setting of the resource users; the specific rules-in-use that affect the incentives individuals face in a particular situation; the patterns of interaction among the resource users; and the likely outcomes (ed.).

43. Ostrom, E., Gardner, R., Walker, J.M. 1994. Rules, Games and Common-Pool Resources. Ann Arbor, Michigan: University of Michigan Press.

Notes: Explores ways that the tragedy of the commons can be avoided by people who use common-property resources. Contributors include Elinor Ostrom, Roy Gardner, James Walker, Arun Agrawal, William Bloomquist, Edella Schlager, and Shui Yan Tang. Chapter 2 Institutional Analysis and Common Pool Resources provides a clear introduction to the foundations of Common Pool Resources and a description of the Institutional Analysis and Development (IAD) framework. The IAD framework can be used as a general organizing tool that helps in the development of a long-term research program. This approach is applicable not only for research on common property

C-16 CBNRM Social Science Resource Kit Institutional Analysis Resource Book resources, but also on other problems where individuals find themselves in repetitive situations affected by a combination of factors derived from a physical world, a cultural world, and a set of rules (author adapted).

Contents: PART 1: THEORETICAL BACKGROUND- 1. Rules, Games and Common- Pool Resource Problems; 2. Institutional Analysis and Common Pool Resources; 3. Games Appropriators Play; 4. Rules and Games. PART 2: EXPERIMENTAL DESIGN- 5. CPR Baseline Appropriation Experiments; 6. Probabilistic Destruction of the CPR; 7. Communication in the Commons; 8. Sanctioning and Communication Institutions; 9. Regularities from the Laboratory and Possible Explanations. PART 3: FIELD STUDIES- 10. Institutions and Performance in Irrigation Systems; 11. Fishers' Institutional Responses to Common-Pool Resource Dilemmas; 12. Rules, Rule-Making, and Rule Breaking: Examining the Fit Between Rule Systems and Resource Use; 13. Changing Rules, Changing Games: Evidence from Groundwater Systems in Southern California; 14. Regularities from the Field and Possible Explanations. PART 4: CONCLUSION- 15. Cooperation and Social Capital.

44. Ostrom, E. 1992. Crafting Institutions for Self-Governing Irrigation Systems. San Francisco: Institute for Contemporary Studies Press.

Abstract: Many large-scale irrigation projects established in developing countries have failed, despite huge inputs of foreign aid, because the institutions for operating and maintaining the projects proved inadequate. Ostrom counsels planners to look beyond a narrow concern with engineering and to let irrigation consumers involve themselves in the design, operation, and maintenance of water-supply systems (ICS Press).

45. Ostrom, V., D. Feeny, and H. Picht, eds. 1988. Rethinking Institutional Analysis and Development: Issues, Alternative, and Choices. San Francisco: Institute for Contemporary Studies Press

Abstract: This book challenges traditional economic thinking which has explained economic growth in terms of resources, technologies, and human preferences. Noting that economists have not been able to account fully for economic growth, the editors argue that a fourth component-institutions-is the critical missing link. As the thirteen authors demonstrate, the influence of institutions is considerable: establishing the basic rules governing all public and private actions from individual property rights to the ways in which communities deal with public goods, and affecting distribution of income, efficiency of resource allocation, and the development of human resources (ICS Press).

C-17 CBNRM Social Science Resource Kit Institutional Analysis Resource Book

46. Richards, Michael. 1997. Common Property Resource Institutions and Forest Management in Latin America. Development and Change, Volume 28: Issue 1: 95 -117

Abstract: This article focuses on how common property resource (CPR) institutions managing forest resources in Latin America have responded to change, a subject relatively ignored in the English-language literature. It examines in particular the evidence surrounding the popular view that CPR institutions must inevitably break down in the face of economic and demographic pressures-an extension of the 'tragedy of the commons' thesis. The evidence shows that there have been a number of both positive and negative experiences. The negative experiences include the obvious vulnerability of Amerindian informal institutions to the individualistic incentive structures of market forces. However, many indigenous and other groups have responded positively to market pressures and there is ample evidence that, given an appropriate policy environment, community-based natural forest management can still be regarded as a 'great white hope' for forest conservation, especially considering the largely negative environmental and equity impacts of individualized resource privatization, as in the Brazilian Amazon (author).

47. Ruddle, Kenneth. 1993. External Forces and Change in Traditional Community-Based Fishery Management Systems in the Asia-Pacific Region. Maritime Anthropological Studies (MAST) 6(1-2):1-37.

48. Schlager, Edella. 1994. Fisher's Institutional Responses to Common Pool Resource Dilemmas. In Ostrom, E., Gardner, R., Walker, J.M. Rules, Games and Common-Pool Resources. Ann Arbor, Michigan: University of Michigan Press.

49. Schlager, Edella, W. Blomquist, and S.Y. Tang. 1994. Mobile Flows, Storage and Self-Organized Institutions for Governing Common Pool Resources. Land Economics 70: 294-317.

50. Schlager, Edella, and Elinor Ostrom. 1992. Property Rights Regimes and Natural Resources: A Conceptual Analysis. Land Economics 68(3):249-62.

51. Song, Y.J. 1997. New Organizational Strategy for Managing the Forests of Southeast China - The Share-holding Integrated Forestry Tenure (Shift) System. Forest Ecology and Management 91(2-3):183-194.

Abstract: This study examines the factors influencing the adoption of a unique organizational innovation for local forest management in China. The share-holding

C-18 CBNRM Social Science Resource Kit Institutional Analysis Resource Book integrated forestry tenure (or the SHIFT) system in Sanming prefecture of Fujian province, People's Republic of China, is based on a system of benefit.sharing and tenureship that provides both management responsibility and economic incentive to local villagers. The adoption of SHIFT has improved. sustainability, productivity, equity, and flexibility in the face of changes caused by biophysical and human intervention, and encouraged a more democratic rural life and more power of independent decision-making. It has also given a greater role to professional foresters and led to a more diversified set of management goals (Sociofile).

52. Tamang, Devika, Gerard J. Gill, and Ganesh B. Thapa, eds. 1993. Indigenous Management of Natural Resources in Nepal. Kathmandu, Nepal: HMG Ministry of Agriculture/Winrock International.

53. Tang, Shui Yan. 1994. Building Community Organizations: Credible Commitment and the New Institutional Economics. Human Systems Management 13: 221-232.

54. Tang, Shui Yan. 1992. Institutions and Collective Action: Self-Governance in Irrigation. San Francisco: Institute of Contemporary Studies Press.

Abstract: Tang evaluates the best conditions and relationship with government for self-governing irrigation systems (ICS Press).

55. Thapa, Ganesh. 1997. Indigenous Management of Nepal's Natural Resources: Some Policy Issues. In People and Participation in Sustainable Development: Understanding the Dynamics of Natural Resource Systems. G. Shivakoti, et al. (eds.). Bloomington, Indiana and Kathmandu, Nepal: Workshop in Political Theory and Policy Analysis, Indiana University and Institute of Agriculture and Animal Science, Tribhuvan University.

Abstract: This paper briefly reviews the performance of indigenous management systems in Nepal to discuss related policy issues. In Nepal, two systems of natural resource management can be observed. At the formal level, there is the centralized, bureaucratic system of management by government agencies, implementing laws and policies to basically regulate the society with certain perceptions of government. At the other, more informal level, local communities have been managing natural resources, supported by a network of community structures and organizations, indigenous practices, customary rules and laws, and enforced by the society or community concerned. This indigenous management of natural resources usually goes unnoticed or ignored in development activities. Only recently, due to increased awareness, the government has been slowly incorporating these institutions into the mainstream development path (author).

C-19 CBNRM Social Science Resource Kit Institutional Analysis Resource Book

56. Thomson, J.T., D. Feeny, and R.J. Oakerson. 1992. Institutional Dynamics: The Evolution and Dissolution of Common Property Resource Management. In Bromley, Daniel et al. (eds). Making the Commons Work: Theory, Practice, and Policy. San Francisco: Institute for Contemporary Studies Press.

Abstract: Using a comparative institutional analysis approach, authors analyze resource-use patterns in Niger and Thailand. The focus is on the tension between and changes in the balance between local-level management arrangements and those at the center (the national government). The intriguing aspect of this work is that it highlights the role of local learning and innovation in management regimes (ed.).

57. Thomson, J. T. 1992. A Framework for Analyzing Institutional Incentives in Community Forestry. Rome: FAO.

Abstract: This document presents a technique for analyzing community forestry problems and designing modified or new community forestry institutions. The technique has been used successfully to address renewable natural resource problems, particularly irrigation systems and fisheries (author).

58. Umans, Laurent. 1993. Analysis and Typology of Indigenous Forest Management in the Humid Tropics of Asia. National Reference Centre for Nature, Forests and Landscape of the Ministry of Agriculture, Nature Management and Fisheries, Stichting BOS (Foundation for Netherlands Forestry Development Cooperation), Wageningen, The Netherlands. (BOS Werkdocument).

59. Uphoff, N. 1994. Local Organization for Supporting People-Based Agricultural Research and Extension: Lessons from Gal Oya, Sri Lanka. In Beyond farmer first: Rural peoples knowledge, agricultural research and extension practice. Scoones, I. and Thompson, J. eds. London: Intermediate Technology Publications. Pp. 182-202.

60. Uphoff, N. 1993. Grassroots Organizations and NGOs in Rural Development: Opportunities with Diminishing States and Expanding Markets. World Development 21 (4): 607-622

Abstract: This article undertakes to deal systematically with nongovernment organizations (NGOs) and grassroots organizations (GROs), usually treated as a residual category. Ten levels for development decision-making and action are identified in the tradition of central place theory. Three of these are considered "local" or "grassroots" because of their collective action possibilities. Then, three sectors are delineated, instead of the usual two. NGOs are part of the collective action sector which differs from the public and private sectors in theoretically consistent ways. A careful distinction is made between institutions and organizations, usually conflated in the

C-20 CBNRM Social Science Resource Kit Institutional Analysis Resource Book

literature. Examples of accelerated rural development through grassroots organizations and NGOs are discussed from Asia, Africa and Latin America (author).

61. Uphoff, N. 1986. Local Institutional Development: An Analytical Sourcebook with Cases. West Hartford: Kumarian Press. 421 pp.

Abstract: The importance of local institutions for sustainable rural development has until recently been afforded inadequate attention by governments and donor agencies. This analytical sourcebook draws on experiences summarized in mini case-studies from throughout the Third World. Critical concepts and issues are explored in its examination of various sectors of local institutional development (LID) including agriculture, infrastructure, and primary health care. Intended especially for practitioners and researchers seeking a rigorous understanding of LID, the material presented is well-suited for both field and classroom use (author).

Contents: 1. Analysing Options for Local Institutional Development; 2. Local Institutional Development for Natural Resource Management; 3. Local Institutional Development for Rural Infrastructure; 4. Local Institutional Development for Primary Health Care; 5. Local Institutional Development for Agriculture; 6. Local Institutional Development for Nonagricultural Enterprise; 7. Strategies for Supporting Local Institutional Development; 8. Mobilizing and Managing Economic Resources for Local Institutional Development

62. Zein-Elabdin, Eiman. 1996. Development, gender, and the environment: Theoretical or contextual link? Toward an institutional analysis of gender. Journal of Economic Issues vol. 30 no.4 (Dec.) Pp. 929-947

Abstract: The current discourse on development, gender, and the environment has emerged from a convergence of feminist and environmentalist critiques of economic development. This discourse is dominated by two paradigms: women in development and ecofeminism. Although the two paradigms have been influential in advancing women's issues in economics, this influence has been limited by the lack of understanding of the institutional nature of gender among the proponents of the paradigms. An institutional framework is suggested to analyse the intersection of development, gender, and the environment. This framework is based on an interpretation of gender specification as an institution in the Veblenian sense of habits of thought and on Polanyi's definition of the economy as instituted provisioning processes, thereby fully legitimizing gender specification as a subject of study from an economic viewpoint, and appreciating its role in determining the relative economic and social positions of women and men in society (author).

C-21

CBNRM Social Science Resource Kit Institutional Analysis Resource Book

D. Obtaining Documents Listed in the Bibliography

IDRC Document Delivery Service

The IDRC library offers a document delivery service to all Centre-funded projects. Any project staff member may request, from the IDRC library, copies of journal articles or excerpts from books free of charge. The IDRC library will send these documents to the project via regular mail. Please note that whole books cannot be copied or loaned and only one copy of any journal article can be provided per project.

Procedure

Send a request via e-mail, fax or regular mail (address below) to Madeleine Audet. The request must include a minimum of information in order to be processed.

For a Journal Article please include: Author, Title, Date, Journal Name, Volume, Issue and Pages. For a Book Chapter, please include: Author, Title, Date, Publisher and Pages

As well, you will need to identify the name and number of your IDRC project and your institution. In order to simplify this process an order form has been attached below. You may wish to print this off and use it when ordering by fax or regular mail or complete it in electronic format and attach it to an e-mail message.

Please note that as an IDRC project recipient you are entitled to this service for any journal article or book chapter that you wish-not just those listed in the resource kit.

Using the form provided on the following page, please direct reference requests to:

Madeleine Audet Research Information Management Service (RIMS) IDRC PO Box 8500 Ottawa, ON Canada K1 G 3H9

Telephone: (613) 236-6163 ext 2257 Fax: (613) 238-7230 e-mail: [email protected] (cc your message to [email protected])

CBNRM Social Science Resource Kit Institutional Analysis Resource Book

CBNRM Journal Article Request Form

Please use this form to indicate those journal articles and book chapters you would like to have IDRC copy and deliver to you.. It may take up to 4 - 6 weeks for delivery from the date we receive your request.

Your Name:

Project Title/Number:

Institution:

Project Leader:

Mailing Address:

No. I Journal Article or Book Chapter (please include author, title, date, journal name, volume, issue and pages)

CBNRM Social Science Resource Kit Institutional Analysis Resource Book

E. Websites and Electronic Information

This section presents selected websites and mailing-lists related to institutional analysis and CBNRM that offer useful resources for researchers.

Websites

1. Workshop in Political Theory and Policy Analysis http://www.indiana.edu/workshop/

The Workshop in Political Theory and Policy Analysis, organized within the Department of Political Science at Indiana University and co-directed by Elinor and Vincent Ostrom, is a multidisciplinary institution composed of scholars who have a central interest in the study of economic and political institutions-their evolution and impact on behavior. The Workshop is home to the Program for Advanced Study in Comparative Institutional Analysis and Development. Available on-line are a brochure describing the Workshop, lists of publications, and the Workshop Newsletter "Polycentric Circles". The on-line Research Library includes the following sections: (1) CPR Virtual Library of Common Pool Resources, with CPR Bibliographies; IASCP Conference Abstracts; Online CPR-Related Articles & Books; Useful CPR-Related Links; and (2) Workshop Bibliographies including three of particular relevance: Institutional Analysis and Development; Indigenous Knowledge and Institutions; and Rules and Norms.

2. Center for the Study of Institutions, Population, and Environmental Change http://www.indiana.edu/-cipec/

The Center for the Study of Institutions, Population, and Environmental Change (CIPEC), also co-directed by Elinor Ostrom, studies processes of change in forest environments as mediated by institutional arrangements, demographic factors, and other major human driving forces. The site includes a description of CIPEC, a list of recent publications, and short fact sheets on the following topics as they relate to CIPEC's research: Institutional Analysis; Methods; Remote Sensing; GIS; Land Use/ Land Cover; and Demography. CBNRM Social Science Resource Kit Institutional Analysis Resource Book

3. International Forestry Resources and Institutions Research Program (IFRI)

http://www.indiana.edu/-ifri/

The IFRI research program was initiated in 1992 at Indiana University to address issues such as: determining how to change processes leading to deforestation in many countries of the world; assessing what type of institutions are associated with sustainable forest practices; and helping policy makers and forest users to design more effective forest policies. The Information Resources section of the site includes a collection of links to full text articles on the subject of Forestry (including many related to institutional analysis), as well as links to other forestry resources. The site also includes a description of IFRI, a list of publications (some with abstracts); and a description of IFRI's research strategy.

4. International Association for the Study of Common Property (IASCP)

http://www.indiana.edu/-iascp/

The International Association for the Study of Common Property (IASCP), founded in 1989, is a nonprofit Association devoted to understanding and improving institutions for the management of environmental resources that are (or could be) held or used collectively by communities in developing or developed countries. The website includes IASCP's mission statement, announcements by IASCP members, conferences, and a brief description of IASCAP's quarterly publication The Common Property Resource Digest which features articles, the CPR Forum (Commentaries and responses on CPR issues), Book Reviews, Bibliographies, Announcements, and Letters to the Editor. The Digest is available to members only and is not available on-line. The site also features the CPR Virtual Library of Common Pool Resources, an excellent resource available to both members and non-members which contains searchable bibliographies, conference abstracts, a listing of CPR-related articles and books available for free on-line, and useful links. For further information send an e-mail to [email protected]

5. Chr. Michelsen Institute: Development Studies and Human Rights

http://www.cmi.no/

The Chr. Michelsen Institute (CMI) is an independent, non-profit social science research institution working on issues of development and human rights. Of particular relevance is the Natural Resources Management Programme, which addresses issues related to the exploitation of natural resources in a social and institutional context. A central concern is to examine how institutions other than property institutions regulate access to natural resources to better understand the dynamics of natural resource exploitation in communities in developing countries. Recent work has mostly dealt with the

E-2 CBNRM Social Science Resource Kit Institutional Analysis Resource Book following themes: management of tropical forests, including modelling of deforestation processes; fisheries management, mainly focussing on small-scale production and including situations of multi-agent exploitation of fishery resources; environmental impact of social change. The site includes the following main sections: General Information; Announcements; Research Activities; Publications; and Library. The Publications sections offers abstracts and ordering information for External Publications; CMI Reports Series; CMI Working Paper Series; Conference Papers by CMI Staff; and Proceedings of CMI Conferences. A number of these publications deal with institutions and institutional analysis. The Library section includes a bibliographic database; thematic lists on countries/areas, including Asia and Southeast Asia; new documents lists; and bibliographic links.

6. NRM Changelinks http://nrm.massey.ac.nz/changelinks/

This site provides an excellent, comprehensive on-line guide for natural resource managers and others working to help communities identify and adopt more sustainable natural resource management practices. Links and on-site material provide approaches, information and theory related to natural resource management. The two sections on "capacity building" and "collaborative planning & management" are of particular relevance for those interested in institutional approaches. Other pages provide information on conferences, discussion groups and job opportunities in the fields of environment and development.

7. ELDIS: Institutions and Environmental Entitlements http://www.ids.ac.uk/eldis/envids/dp/instit.html

ELDIS is a "gateway" to information sources on development and the environment available free via the internet. It provides an ever increasing number of descriptions and links to a variety of information sources, including WWW and gopher sites, databases, library catalogues, bibliographies, and e-mail discussion lists, research project information, map and newspaper collections. For relevance to CBNRM, this is one of the most useful sites currently available. The above VVWW site on institutions offers a short summary of the "environmental entitlements" approach to institutional analysis, and links to several related papers and bibliographies. CBNRM Social Science Resource Kit Institutional Analysis Resource Book

8. International Center for Living Aquatic Resources Management (ICLARM) http://www.cgiar.org/iclarm/index.htm

ICLARM is an international research organization devoted to improving the productivity, management and conservation of aquatic resources for the benefit of users and consumers in developing countries. Site includes sections on News; Background on ICLARM; Research Programs; Publications and Software; and Library and Information Services (with links to other relevant information services). Under Research Programs, click on the link "Policy Research and Impact Assessment" to get information about several projects involving institutional analysis in Southeast Asia. The site also includes a link to The Asian Fisheries Society homepage, a scientific society organized in 1984 for fishery professionals in Asia to communicate, share information and cooperate with each other. This homepage offers background information, a discussion of fisheries issues, list of publications, and links to related websites.

9. Forests, Trees and People Programme & Network http://www-trees.slu.se/

This Web site is part of the Forests, Trees and People Programme's networking activities. The Network is designed to share information about improved methods of planning and strengthening community forestry activities and about on-going or planned initiatives of potential interest to its members. The Forests, Trees and People Newsletter, which forms part of the Programme's networking activities, is a quarterly publication distributed to field projects, institutions, organizations and individuals interested in and/or working with community forestry activities. The newsletter contains articles on subjects relevant to community forestry activities (based on field reports and research findings) as well as book reviews and a section on information, activities, events and courses of interest to network members. Several issues are now available on this website and future issues will be included when they become available. Members receive the newsletter free of charge (this may change). Site includes links to the following regional facilitators of the FAO:

RECOFTC (Regional Community Forestry Training Center) http://www.recoftc.org/ Based in Thailand, the main objective of RECOFTC is to organize, provide, facilitate and otherwise support training for community forestry initiatives throughout the Asia-Pacific Region. CBNRM Social Science Resource Kit Institutional Analysis Resource Book

WATCH (Women Acting Together for Change) http://www-trees.slu.se/nepal/watchindex.htm Based in Nepal, the organization's mission is to try out an alternative approach of development in which rural people, especially women and disadvantaged groups, become subjects of development rather than objects, so that they can develop enough confidence and capabilities required to make decisions and implement them as required.

10. International Institute for Environment and Development (TIED) http://www.iied.org/

IIED's principal aim is to improve the management of natural resources so that communities and countries of the South can improve living standards without jeopardising their resource base. Its work is undertaken with, or on behalf of governments and international agencies, the academic community, foundations and groups and the people they represent. The website provides information on IIED's seven programmes and associated research activities: Environmental Planning and Management; Human Settlements; Sustainable Agriculture; Forestry and Land Use; Drylands; Environmental Economics; and a European Programme. From the website users can link to the TIED Resource Centre, a unique service for those seeking practical information and support on all aspects of research on participatory methodologies, community wildlife management, environmental planning, profiles and strategies with a particular focus on their application and integration into institutional structures. The website also has a list of TIED publications and, especially interesting, an updated page on 'grey' literature which can be downloaded. TIED publishes the journal PLA Notes (Participatory Learning and Action), formerly RRA Notes, which is free to people from developing countries.

Mailing Lists

11. Community Research Network List (CRN-List) [email protected]

This list, run by the Loka Institute on-line at http://www.loka.org/crn/crnlist.htm is intended to connect people interested in community-based research. Discussion topics include the establishment of new centers, research methodologies, funding strategies and resource sharing, conference announcements (and sometimes planning), and democratic research and science. Subscribers to the listserv include practitioners of community-based research working at established community research centers (both university-based and non-profit, community-based organizations), and individuals who are either praciticing or would like to practice community-based research. Subscribe by

E-5 CBNRM Social Science Resource Kit Institutional Analysis Resource Book sending a message to the above e-mail address with a blank subject line and these words on one line in the body text of the message: subscribe crn-list

12. Partnerships [email protected]

This is a new list discussing what works and what doesn't in 'real world' and online partnerships, and the processes of participation and collaboration. For more information visit Partnerships On-line at http://www.partnerships.org.uk/part/ Subscription Information: Send e-mail to the above address; in the body of the message, type SUBSCRIBE PARTNERSHIPS

13. CM-Forum

This is an open platform, intended to facilitate information exchange and discussions on collaborative management (CM) of natural resources. It is managed by the Social Policy Group (SPG) at IUCN. Practitioners are invited to present CM initiatives, share lessons learned, and discuss issues related to participatory natural resource management with colleagues around the world. Subscription Information: Send e-mail to [email protected]; in the body of the message, type SUBSCRIBE CM-Forum.

14. AgREN E-mail Conference [email protected]

This electronic network was established early in 1996 to help increase the degree of interaction among AgREN members. 'Discussion' within the conference is structured around the issues raised in the most recent set of Network papers received by members. It enables members to comment on previous papers, contribute their own experiences and advance the discussion in key areas. The proceedings of the conference are summarised in the subsequent AgREN Newsletter for the benefit of those who do not have access to electronic communication systems. If you are interested in joining the conference, send an e-mail to the above address. NOTES