Society and Non-timber Products in Tropical Asia edited by Jefferson Fox

EAST-WEST CENTER OCCASIONAL PAPERS Environment Series

No. 19, August 1995

EAST-WEST CENTER The U.S. Congress established the East-West Center in 1960 to foster mutual understanding and cooperation among the governments and peoples of Asia and the Pacific region, including ihe United States. Principal funding for the Center comes from the U.S. government, with additional support provided by private agencies, individuals, and corporations, and more than zo Asian and Pacific governments.

The Center promotes responsible development, long-term stability, and human dignity for all people in the region and helps prepare the United States for constructive involvement in Asia and the Pacific.

The Program on Environment was established in October 1977 as a unit of the East-West Center. It engages in research to improve management of renewable resources and the environment in Asia and the Pacific. Work of the program is organized within five areas: biodiversity conser• vation, environmental governance, environmental risk and develop• ment, renewable resources management, and urban environmental management.

East-West Center Occasional Papers: Environment Series facilitates early dissemination of the Center's research findings and policy-relevant reports on the environment of Asia, the Pacific, and the United States. All manuscripts are peer-reviewed. This subseries continues the series Occasional Papers of the Program on Environment, which was published at the Center from 1987 to 1993.

East-West Center 1777 East-West Road Honolulu, Hawaii 96848, USA Telephone: |8o8) 944-7145 Facsimile: (808) 944-7376 EAST-WEST CENTER Society and Non-timber Forest Products in Tropical Asia

edited by Jefferson Fox

EAST-WEST CENTER OCCASIONAL PAPERS Environment Series No. 19, August 1995 About the Volume Editor lefferson Fox is a fellow in the Program on Environment at the East-West Center. He is also the program area coordinator of Renewable Resources Management. He has worked on designing and implementing joint community/state resource management plans in Asia for more than a decade. Telephone: (808) 944-7248 Facsimile: (808) 944-7298

LIBRARY OF CONGRESS CATALOCING-IN-PUBUCATION DATA Society and non-timber forest products in tropical Asia / edited by Jefferson Fox. p. CM. — (East-West Center occasional papers. Environment

series ; no. 19) "August 1995" Papers presented for a panel at the xvii Pacific Science Congress held in Honolulu, Hawaii, May 27-June 2, 1991. Includes bibliographical references.

ISBN O-86638-169-4 1. Forest products—Asia—Congresses. 2. — Social aspects—Asia—Congresses. 3. Forest policy—Asia— Congresses. 4. Forest products industry—Asia—Congresses. 1. Fox, Jefferson, 1951- . 11. Pacific Science Congress (17th : 1991 : Honolulu, Hawaii) in. Series. SD543.3.A78S64 1995 338.r74987'09509,3—dcio 94-4859° CIP

© 1995 by East-West Center. All rights reserved. Printed in the United States of America. CONTENTS

Figures and Tables v

Foreword vn

Preface ix

Introduction xi by Jefferson Fox

PART ONE

COLLECTING, PROCESSING, AND MARKETING:

VARIATION AND CHANGE

Chapter i Creating the Forest: Dayak Resource Management in Kalimantan 3 by Christine Padoch

Chapter 2 Bomean Forest Trade in Historical and Regional Perspective: The Case of Penan Hunter-Gatherers of Sarawak 13 by J. Peter Brosius

Chapter 3 Sal Leaf Plate Processing and Marketing in West Bengal 27 by Rashmi Pachauri Raj an

Chapter 4 Non-timber Forest Products in a Nepali Village in 1980 and 1990 37 by Jefferson Fox iv Society and Non-timber Forest Products in Tropical Asia

PART TWO

LOCAL INSTITUTIONS AND GOVERNMENT POLICIES:

METHODS OF MANAGING (OR MISMANAGING) NON-TIMBER

FOREST PRODUCTS

Chapter 5 The Impact of Cultivation on Peasant-State Relations in Development: The Transition from Native Forest Latexes to the Introduced Hevea in Kalimantan 55 by Michael R. Dove

Chapter 6 Extraction Interactions: Tropical Timbers in West Kalimantan, Indonesia 73 by Nancy Lee Peluso

Chapter 7 Management for Forest Conservation in Indonesian National Parks and Preserve Buffer Zones 97 by Stephen E Siebert

Chapter 8 Fiber Grass from Forest Land: A Case from North India 109 by Mark Poffenberger and Madhu Sarin

Conclusions 131 by D. G. Donovan

Contributors 143 FIGURES AND TABLES

Figures

5.1 The Kantu' territory in Kalimantan, Indonesia 57

8.1 HRMS grass lease areas in 1991 in the Pinjore-Morni Forest Division, Haryana 112

8.2 Kahinwala village and the HRMS grass lease area 115

Tables

3.1 Sal forest areas in Southwest Bengal 30

3.2 Cost breakdown of mechanized sal thai making 33

4.1 Firewood consumption 43

7.1 Frequency of Calamus exilis plants and canes in lower montane of Kerinci-Seblat National Park 101

7.2 Soil characteristics in lower montane forests of Kerinci-Seblat National Park 102

It could be difficult to overstate the importance of non-tim• ber forest products (NTFPs). For rural people in the region, such products commonly contribute to meeting food and other basic needs, are a source of input into the agricultural system, help households control exposure to risk of various kinds, and often constitute an integral part of the habitat and of the social and cultural structure of those living within that environment. Very large numbers of households also generate some of their income from selling NTFPs. A better understanding of the magnitude and nature of the role of NTFPs is therefore central to making decisions about forest management that adequately reflect society's de• mands upon the forest resource. However, even a cursory examination of the information available demonstrates the diversity of uses, needs, and patterns of involvement that oc• cur and the extent to which these are often changing. In some situations users are becoming more dependent on NTFPs; in others they are moving away from dependence on them. With regard to products overall in a given situation, some ac• tivities may be expanding while others are declining. Despite their importance, until recently the amount of research directed at synthesizing and interpreting these di• verse and evolving experiences has been limited. This vol• ume brings together papers that report on recent work that does address such issues. Covering a range of different situa• tions, they supplement the existing literature that helps ex• tend our knowledge and understanding of the human and social dimensions of NTFP use and management.

J. E. Michael Arnold Oxford Institute University of Oxford Oxford, United Kingdom

PREFACE

The papers presented in this volume were prepared for a panel at the XVII Pacific Science Congress held in Honolulu, Ha• waii, 27 May-2 June 1991. The panel focused on the collec• tion, utilization, and marketing of non-timber forest prod• ucts (NTFPs| in tropical Asia, how these patterns change through space and time, and the role of local institutions and government policies in managing these products. The devel• opment of NTFPs is a complex and multidimensional un• dertaking, contingent on the ecological characteristics of the plant species, the social and political context of local and national policies, and the scale of collection and marketing. Much of the NTFP literature is based on experiences from Latin America, but the literature on Asian experiences is sparse, specifically with regard to defining the major issues. These papers bring Asian perspectives on NTFPs into clearer focus. They offer details learned from case studies, and an• chor local-level realities learned through years of fieldwork with theoretical concerns. In compiling this volume, I have benefited from the comments, suggestions, and support of numerous people and organizations. The impetus for the panel came out of discus• sions with Nancy Peluso and Christine Padoch. Leslie Sponsel provided useful comments, and Erin Moore provided invalu• able editorial assistance on the original manuscript. At the East-West Center the volume benefited from comments by Michael Dove and Deanna Donovan, while Helen Takeuchi provided professional editorial assistance.. I would particu• larly like to thank the East-West Center and the Ford Foun• dation for funding the participation of some of the authors. Finally, I owe a considerable debt to the authors, who not only stimulated the thinking of everyone at the panel but also made the publication of this volume possible. x Society and Non-timber Forest Products in Tropical Asia

Rather than focus on any single element in depth, the papers address the many problems associated with NTFP development. The complexity of the issues and the site speci• ficity of the analysis signify that no precise solutions can be offered to the various problems raised. I hope that these pa• pers will supplement the sparse literature on NTFPs in Asia, contribute to a broader understanding of these products in the region, and relate to issues of sustainable development wherever forests are threatened.

Jefferson Fox Honolulu, 1995 Jefferson Fox

Introduction

Asia, after Latin America, is the location of the world's sec• ond largest block of tropical forests. Its rain forests are among the most complex and species-rich ecosystems in the world, harboring no less than 25,000 species of flowering plants, or 10 percent of the world's flora, and an even greater diversity of animal species (Collins, Sayer, and Whitmore 1991}. Non- timber forest products |NTFPs) from these forests have pro• vided forest-dwelling people and indigenous entrepreneurs with an important source of income and goods for nearly two thousand years (Wolters 1967). Only recently, however, have scientists begun to recognize that NTFPs are highly signifi• cant to rural and national economies and that these products are of greatest economic value to the rural population living in and near forests (deBeer and McDermott 1989; FAO 1991; Malhotra, Deb, and Dutta 1991}. These and other authors (e.g., Peters, Gentry, and Mendelsohn 1989; Wickens 1991; Panayotou and Ashton 1992] conclude that NTFPs represent a sustainable form of tropical forest exploitation and that there is great economic potential inherent in conserving and en• hancing forest resources through the marketing and local processing of these products. NTFPs are defined both by what they are and by how they are managed and harvested. They encompass all bio• logical materials that are extracted from natural forests for human use. These include foods, medicinal and culinary herbs, , latexes, dyes, fodder and fiber grasses, wildlife (products and live animals), fuelwood, rattan, , and smallwood. Methods of harvest include the collection of natu• ral products from the forest, and in some cases the cultiva• tion, generally on a small scale, of products that originally XII Society and Non-timber Forest Products in Tropical Asia grew naturally. Thus the definition includes harvested for local use for housing, construction, or export from the region of origin but not wood harvested by large-scale timber interests. Similarly, rattan grown in swiddens is included whereas that grown in large is excluded (deBeer and McDermott 1989; Peluso 1989). In rural households, NTFPs provide food security in the form of staples, nutritional supplements, such as snacks and side-dishes, and buffers against seasonal and emergency short• ages; health care in the form of ingredients in traditional medicines; raw materials for construction and implements for household use (i.e., agriculture, fishing, hunting) or small- scale processing and enterprises,- and finally, income and employment from the collection, trade, or pro• cessing and manufacture of NTFPs (Falconer 1988). In national economies, NTFPs provide income and materials for direct consumption or for use in large and small industries; employment in collection, trade, processing, and manufacture; and much-needed foreign exchange through export of raw materials and processed or finished products (Godoy and Lubowski 1992). Valuation of NTFPs, however, is plagued by inadequate measurements of costs, quantities extracted, and prices. Although researchers have produced important case studies, the results of different studies can• not be directly compared because different methods have been used (Godoy and Lubowski 1992). Estimates of the economic value of NTFPs are conservative because forests also pro• duce benefits such as biological diversity and environmental services (Panayotou and Ashton 1992). Despite growing interest worldwide in NTFPs and their economic potential, there is little information on questions such as how these products are managed, collected, used, or marketed (Godoy and Bawa 1993); how patterns of exploita• tion are changing in response to both internal and external pressures (May 1991); or how NTFP collection patterns af• fect relations within a community, between communities, or between communities and outside groups (i.e., traders, Introduction XIII

middlemen, and bureaucrats) or the state (Peluso 1989; Dove 1993). These important questions need to be examined be• fore NTFPs can be developed as a means of conserving forest resources. The papers in this volume seek to provide insights into these questions. The papers are organized in terms of the collection, use, and marketing of NTFPs and how these patterns change through space and time, and the role of local institutions and government policies in managing these prod• ucts. The specific readings are briefly discussed below.

COLLECTING, Christine Padoch describes how forests and agroforests are PROCESSING, managed by a Dayak group in West Kalimantan, stressing

AND MARKETING: how variability and change characterize these managed plots. VARIATION Her comments provide insights into why traditional forest AND CHANGE management patterns have been virtually invisible to most researchers and government officials. Padoch concludes that only when diversity is valued as a desirable characteristic in and forestry will traditional forest management practices be appreciated., But for now these woodlands are too managed for those interested in conservation and not managed enough for those interested in production. Peter Brosius analyzes patterns of collecting and trad• ing forest products by Penan hunter-gatherers in Sarawak, East Malaysia, and how these patterns vary through time and space. Brosius recognizes a trend away from extraction alone to manufacture, as is the case with the production of rattan products. Concomitant with this has been an increase in women's role in producing items for trade. Whereas trade based on the extraction of forest resources—whether cam• phor, jelutong, or damar—was primarily the province of men, the contemporary trade economy is based largely on the con• tribution of women. Rashmi Pachauri Rajan writes about the production, marketing, and sale of sal thai (leaf plates) in the Midnapore district of West Bengal, India. Rajan is particularly concerned about the economic effects the production process has on xiv Society and Non-timber Forest Products in Tropical Asia

local communities, and with the role of the West Bengal For• est Department in the production and marketing of this NTFP. Rajan concludes that to improve the production and market• ing system for sal plates the West Bengal Forest Department needs to "internalize" the experience it has had in joint man• agement systems over the past few years. Central to this pro• cess will be training middle- and field-level officers to better appreciate and understand NTFP systems so that they may better integrate and extend these systems in the long term. Jeff Fox discusses forest-use practices in a village in the hills of central Nepal in 1980 and again in 1990. In 1980, the forests of this village were in poor condition, and the major causes of this degradation were the grazing of and fodder col• lection required for the large livestock population. In 1990 the forests were in much better condition. The paper docu• ments differences in forest conditions and forest products, and seeks an explanation for these changes. Fox's paper brings to mind Boserup's hypothesis that population growth does not necessarily lead to a downward spiral of.land degradation and starvation. Innovations that support more sustainable forest management systems can be triggered by a number of factors including population growth and perceived environ• mental risk. In this case changes in forest tenure, introduc• tion of a new road, and forest management inputs from out• side sources led to a new forest management paradigm.

LOCAL The readings in this section of the volume are concerned INSTITUTIONS with how local people and governments manage or misman-

AND GOVERNMENT age NTFPs. Many of these readings are based in common POLICIES: property theory (e.g., see Ostrom 1990; Poffenberger 1990} METHODS and argue that the separation of ownership (government) from OF MANAGING (OR use (local community) has resulted in open-access situations MISMANAGING) with resulting abuse and exploitation of natural resources. NTFPS These papers argue that with proper conditions, local com• munities can be the best manager of land and water resources. Michael Doye argues that the management of NTFPs is primarily a political challenge involving relations between Introduction xv forest peoples and the state. Dove uses an historical analysis of the transition among peasants and tribesmen in Indonesia from gathering native forest rubbers to cultivating Para rub• ber [) to make observations about the de• velopment of NTFPs and to comment on contemporary peas• ant-state relations. Godoy and Bawa (1993) hypothesize that NTFPs become domesticated following an increase in extrac• tion costs caused by overharvesting, combined with contin• ued demand (Homma 1992). Dove, however, suggests that one of the main reasons tribesmen of Kalimantan planted and maintained Para rubber was not the degradation of na• tive rubbers but because of the tenurial benefits associated with the planting of Para. Indonesian government authori• ties view the planting of Para rubber as establishing rights not just to the but to the land under them. Nancy Peluso examines the indirect effects of commer• cial logging on the production, trade, and management of ironwood [Eusideioxylon zwageri) in a village in West Kalimantan. Peluso demonstrates that the main driving forces behind the destruction of ironwood in this village are com• mercial opportunity and the development of a transporta• tion infrastructure. She convincingly argues that while the traditional management system was not adapted to the con• straints and opportunities of the modern economic system, it could be adapted to modern conditions. To do this, she continues, the Indonesian government would need to recog• nize the existence of a community system for controlling access to ironwood within locally defined territories and to support this system in procedures of conflict management. Without such recognition, this study foreshadows the types of problems that will develop with strategies for marketing NTFPs for the benefit of local people. Steve Siebert provides a view of rattan management from a technical rather than a social perspective; yet, like others, he reaches conclusions with strong social and politi• cal implications. Siebert reviews the abundance, site prefer• ences, and cultivation potential of rattan [Calamus spp.) in two national parks in Indonesia. He concludes that these rat- xvi Society and Non-timber Forest Products in Tropical Asia tan species exhibit promise for sustainable harvesting because plants occur at high population densities, are available for harvest on a year-round basis, have the potential to be re• peatedly harvested without adversely affecting other flora and fauna, may be readily sold through well-established trading networks, and provide local people with an economic alter• native to . But Siebert further concludes that the potential for local management of rattan resources faces major social and political constraints—specifically, ill- defined resource access and tenure arrangements, absence of local management organization, and weak political and eco• nomic power of rattan collectors and villages vis-a-vis out• side commercial interests. The final case study by Poffenberger and Sarin explores the system of joint management developed by the Haryana Forest Department and local communities for managing for• est grasslands. Over the past decade, the Haryana Forest De• partment has attempted to overcome legal, procedural, and organizational barriers as well as the vested interests of pow• erful economic groups to devise participatory management systems for reserve forest lands in the Shivalik Hills. The department has struggled to identify appropriate grass and forest management technologies, and to open communica• tion channels with hill communities in order to involve them in collaborative efforts for regenerating highly eroded water• sheds. The department uses fiber-grass leases as incentives for encouraging community participation. Poffenberger and Sarin point out both the benefits and the problems of this system. In her concluding comments, Donovan suggests these papers illustrate that the diversity of NTFPs by type, techno• logical requirements, institutional arrangements, and social and cultural traditions is so great that no single case can be expected to serve as a universal paradigm. Each product must be considered in proper context (i.e., not only for its economic potential but with careful attention to its role in the ecologi• cal and social systems within which it has developed). Introduction xvn

CONCLUSIONS These papers highlight several lessons regarding the sustain• able collection, use, and marketing of NTFPs by rural com• munities. First, the ways in which NTFPs are collected, used, and marketed change constantly and in different ways in re• sponse to variables both internal and external to the com• munity. Population growth, for example, can lead to and the depletion of resources or, as the paper by Fox points out, can lead to innovative methods for managing scarce resources. Likewise, as Brosius points out, falling prices or stabilizing prices despite inflation lead to diminished in• terest in NTFPs. High prices, however, as in Peluso's study, lead to incredible pressures for uncontrolled exploitation of these products. Another factor that influences NTFP patterns is transportation infrastructure, particularly roads. Roads can lead to increased forest degradation and less sustainable NTFPs as pointed out by Peluso. But roads can also bring in external sources of fertilizers and other inputs and lead to more sustainable management systems as demonstrated by Fox.

Second, the management of NTFPs is foremost a po• litical challenge involving relations between forest peoples and the state. Since the end of World War II, South and South• east Asian countries have increasingly invested control of the region's forest resources in centralized resource manage• ment agencies. Land legislation either ignored or gave little recognition to the customary rights of long-term occupants or indigenous communities. Today, the acceleration of de• indicates that these centralized agencies are fail• ing to manage both timber and non-timber forest resources in a sustainable manner. While many scientists argue that environmental problems in South and Southeast Asia are the result of population growth and the commercialization of natural resources, the most significant factor may be the fail-, ure of existing institutional arrangements to rationally man• age scarce resources. Every author in this volume argues that forest management must explicitly recognize and support the rights of local people to manage their own forest resources. xvm Society and Non-timber Forest Products in Tropical Asia

As the papers by Peluso and Siebert demonstrate, bad na• tional policies lead to forest degradation and to rapid deple• tion of NTFPs. As the papers by Fox and Dove point out, however, good and even inadvertent policies can lead to sus• tainable management of these products. Third, although community participation in managing NTFPs may be a necessary condition for sustainable man• agement, it is not sufficient. Most of these authors demon• strate that state and national governments play important roles in community-based management systems: enforcing a community's right to forest resources against outside in• terests; guarding the interests of the state against short-term profit takers,- conducting research on , forest re• generation, and other topics of interest to community-based management; and providing technical support to local com• munities in developing processing and marketing systems. Finally, developing sustainable management of NTFPs is a learning process. As Poffenberger and Sarin point out, both government bureaucracies and local organizations need to be involved in a continuous process of research and ex• perimentation. As for bureaucracy's viewpoint, the great risk is to place too much emphasis on standard solutions for all situations. The only way to avoid this is to treat each in• stance in developing NTFPs as a learning situation. The im• plications of the various case studies presented here are that research must be built on observation, guided interviews, timeliness, informed interpretation, and attention to the pro• cesses of pragmatic cooperation and reiterative feedback. These case studies also demonstrate that much research still is needed to determine the effects on NTFPs of different ways of structuring government agencies, of different incentives for soliciting participation, and of different pricing and land management policies in NTFP development. Introduction xix

Collins, N. M., J. A. Saver, and M. forest products in a village C. Whitmore. 1991. The conserva• economy: A household survey in tion atlas of tropical forests: Asia Jamboni Range, Midnapore and the Pacific. New York: Simon District, West Bengal. Working and Schuster. paper, Indian Institute of Bio- social Research and Development, deBeer, J. H., and M. J. McDer- Calcutta. mott. 1989. The economic value of non-timber forest products in May, P. H. 1991. Building Southeast Asia. Netherlands institutions and markets for non- Committee for 1UCN, Amster• wood forest products from the dam. Brazilian Amazon. 42(165): 9-16. Dove, M. R. 1993. A revisionist view of tropical and Ostrom, E. 1990. Governing the development. Environmental commons: The evolution of Conservation 20(1): 17-24, 56. institutions for collective action. Cambridge: Cambridge University Falconer, J. 1988. The importance Press. of non-timber forest products in rural economies: A review of the Panayotou, T., and P. Ashton. literature. FAO, Rome. Unpub• 1992. Not by timber alone: lished manuscript. Sustaining tropical forests through multiple use manage• FAO. 1991. Non-wood forest ment. Covelo, Ca.: Island Press. products: The way ahead. FAO Forestry Paper No. 98. Rome. Peluso, N. 1989. Consultancy on non-timber forest products Godoy, R. A., and K. S. Bawa. relevant to local communities, in 1993. The economic value and particular those practicing shifting sustainable harvest of plants and cultivation. FAO document UTF/ animals from the tropical forest: INS/o6s/Ins:Forestry Studies, Assumptions, hypotheses, and Jakarta. methods. Economic Botany 47(3): 115-19. Peters, C, H. Gentry, and R. O. Mendelsohn. 1989. Valuation of an Godoy, R., and R. Lubowski. 1991. Amazonian . Nature Guidelines for the economic 339()une): 655-56. valuation of non-timber tropical forest products. Current Anthro• Poffenberger, Mark, ed. 1990. pology 33(4): 413-33- Keepers of the forest: Land management alternatives in Homma, A. K. O. 1992. The Southeast Asia. West Hartford, dynamics of extraction in Conn.: Kumarian. Amazonia: A historical perspec• tive. In Non-timber products from Wickens, G. E. 1991. Management tropical forests: Evaluation of a issues for development of non- conservation and development timber forest products. Unasylva strategy, ed. D. C. Nepstad and S. 42(165): 3-8- Schwartzman, 23-32. Advances in Wolters, O. 1967. Early Indone• Economic Botany No. 9. Bronx: sian commerce. Ithaca, N.Y.: New York Botanical Garden. Cornell University Press. Malhotra, K. C, D. Deb, and M. Dutta. 1991. Role of non-timber

PART ONE

Collecting, Processing, and Marketing: Variation and Change

CHAPTER 1

Christine Padoch

Creating the Forest: Dayak Resource Management in Kalimantan

Throughout the past decade, many reports have been pub• lished on the sophisticated management of tropical forest and agroforest vegetation by traditional peoples. Descriptions of such systems written years go, and little appreciated when they were first published (e.g., Conklin 1957; Gordon 1969), have also recently attracted attention. Although reports from Latin America (Alcorn 1984; Gomez-Pompa, Flores, and Sosa 1987; Denevan and Padoch 1988; Hecht 1982; Posey 1984, 1985) have been most prominently cited, many interesting Southeast Asian examples are also available (Kunstadter, Chapman, and Sabhasri 1978; Michon 1985; Chin 1985; Fox 1977; Torquebiau 1984; Wiersum 1982). Information on for• est management and traditional agroforestry has served to radically alter our understanding of traditional swidden sys• tems and of forest use. It has also led us to question some assumptions about the "naturalness" of many areas of tropi• cal forest.

This paper will briefly describe the patterns of manag• ing forests and agroforests employed by a Dayak group in West Kalimantan, dwelling particularly on the variability and change that characterize these managed plots. It will also discuss why management of these forests has been virtually invisible to most researchers and government officials.. Fi• nally, comments are offered as to why it is important to rec• ognize that much of the vegetation of West Kalimantan, as well as that of many other tropical forest areas, has been managed to a significant degree by the people who live in and near the forest. 4 Society and Non-timber Forest Products in Tropical Asia

TRADITIONAL Although descriptions of Dayak resource management in

FOREST Kalimantan have often mentioned the use of forest products MANAGEMENT and the planting of tree crops, especially rubber, to date only IN TAE a few works have dealt specifically with the management of forests and agroforests (Chin 1985; Morris 1953; Weinstock 1983). Dayaks are often still popularly perceived not as care• ful managers of resources, but as highly mobile, aggressive groups of shifting cultivators living at low population densi• ties. Perhaps this misconception persists because of promi• nent writings about a few Dayak groups, particularly the Iban of Sarawak (Freeman 1970; Sandin 1967; Sutlive 1978; Padoch 1982). Although such a general description may indeed still characterize a few limited areas of the island, it fails to de• scribe most Dayaks of West Kalimantan. The populations of Sanggau district and specifically of Balai subdistrict in West Kalimantan depart particularly from this traditional model.

Located just slightly north of the Kapuas River, the Balai subdistrict (kecematan) comprises an area of 396 km2. Its population (in 1989) of 21,206 mostly belong to a previously undescribed tribal group, the Tara'n Dayaks. The subdistrict as a whole has a density of 54 people/km2; the population density of some of the villages is substantially higher. The territory of Tae—a village of five hamlets, their fields, and forests—covers a total area of almost 16 km2 and supports about 88 people/km1. While mature forests in the Balai area are largely con• fined to hilltops too steep to cultivate, much of the area ap• pears wooded. Patches of younger and older vegetation on hillsides testify to the continuing practice of swidden farm• ing in the subdistrict..In the Tae hamlet (one of the five that make up the village of the same name), and probably in other parts of the subdistrict as well, swidden cultivation is wan• ing and intensive wet-rice farming is assuming greater im• portance. In March 1990, in Tae hamlet, only nineteen of the thirty-seven households that make up the village practiced swidden farming. About one-half relied only on irrigated or rain-fed permanent fields for their rice crop,- some households employed both techniques. As Tara'n farmers become less Creating the Forest 5 dependent on swiddening, and extend and intensify their use of local variants of irrigated and rain-fed permanent fields, the hillsides are being converted increasingly to managed forests, forest-gardens, or agroforests. The Tara'n distinguish three principal types of man• aged forest vegetation. The distinctions signal differences in stand origin, management priorities, and rights of access and inheritance. The three types are distinguished in the Indone• sian language as tembawang, tanab adat, and tanah usaha.

TEMBAWANC (FOREST-GARDENS)

The term tembawang refers to a former house site. These often extensive areas of forest-garden in Tae apparently origi• nated as the mixed fruit gardens that commonly surround longhouses, individual houses, or temporary farm huts and work areas. Herds of pigs and flocks of fowl roam freely throughout most Tara'n villages. Thus, the planting of an• nual crops close to communities is not feasible. Tree crops, therefore, often are found in a broad swathe around both principal and subsidiary dwellings. Tae residents report that their village sites are relocated about every 20 years, often by moving only a few hundred meters away. Moves are necessi• tated, according to several informants, when the planted fruit trees have grown too tall and shade the front yard. The mul• tipurpose yards then become useless for their most impor• tant function—drying rice in the sun. Rather than cut the fruit trees, however, residents prefer to rebuild their houses elsewhere. Productive fruit trees are rarely, if ever, cut. Several centuries of residence in the region, of reloca• tion and planting of gardens have formed a landscape dotted with tembawang that mark where longhouses, farm huts, and or rubber-processing huts once stood. Each tembawang forest-garden is about 10 hectares and extends hundreds of hectares within the territory of the larger village of Tae. They contain not only planted species but also trees that grow spontaneously among and around planted trees. Management of tembawang over the years includes 6 Society and Non-timber Forest Products in Tropical Asia occasional interplanting of economically important species such as fruit, rubber, construction timbers, and rattan and medicinal fibers. Tara'n informants also cite the most casual and unintentional of planting techniques as important in tembawang formation. They believe that many trees sprout from seeds discarded by people eating fruits in the forest- garden. Most tembawang also receive a slash-weeding, par• ticularly around and tengkawang (illipe ) (Shoiea spp.) during harvest season, so that ripe, fallen fruits may be found and harvested with ease. During such weedings, eco• nomically valuable species are spared while others are cut down. Clearing around tengkawang trees is often especially thorough, with controlled burns sometimes used to allow small fruits to be found easily. In most tembawang it is diffi• cult to know with any certainty whether any but a few se• lected individual trees were definitely planted or whether they were volunteers selected during weedings.

TANAH ADAT (FOREST RESERVES)

The second type of managed forest is tanah adat, or land preserved by customary law. While these forested areas of• ten appear similar to tembawang, they were created in to• tally different ways. Tanah adat are usually areas of forest that have not been cleared, at least not within the memory of current villagers. Tanah adat vegetation has been, how• ever, heavily managed. Some tree species, especially fruits, are occasionally planted among the naturally occurring ones, and other trees are removed. Selective clearing of underbrush also takes place, especially just prior to fruit harvests. Tae hamlet has four tanah adat sites, totaling about 12-13 hectares. Current residents insist that these plots were set aside many generations ago. Two of the tanah adat plots in Tae include village cemeteries. Fruit trees, especially durian, which are planted to commemorate the dead, are quite common in these plots. Areas of tanah adat also include sites specifically preserved as extractive reserves for the entire community. Creating the Forest 7

CYCLIC TANAH USAHA (ENTERPRISE PLOTS)

The third category of managed forests in Tae is often simply called gata (rubber) by local residents, or tanah usaha (en• terprise plots). Most of these areas are planted to highly mar• ketable species. In Tae, rubber [Hevea brasiliensis) indeed is by far the most important, although these plots are often very diverse. All households in the Tae hamlet have substantial productive rubber gardens. Since the Amazonian rubber spe• cies was acquired by Tara'n farmers in the late 1930s or early 1940s, rubber has provided an important supplement to their annual crop production. (Rubber seeds at first were stolen from nearby government rubber estates, as Dayaks at the time were forbidden to plant rubber.) Today the minimum potential daily production of rubber per household in Tae is 3 kg; many household rubber gardens produce 6 to 8 kg per day. Some villagers have also planted limited areas of cacao and other commercial fruits. These enterprise plots, or agroforests, differ from the two previous types in that they are usually parts of cyclic swidden-fallow agroforestry sys• tems, wherein swidden plots dominated by annual crops are gradually transformed into perennial crops. Mixed stands of tree crops, usually rubber plus fruit trees, are planted in the field as the hill rice is ripening. Often after about thirty years, when rubber trees have passed their prime production years and as the pressure for rice land intensifies, these plots are felled for swiddens and a cycle is started anew.

The production of rubber in cyclic tanah usaha plots is soon to be supplemented in Tae with production from improved rubber varieties planted under a government-spon• sored planting program. These new rubber gardens differ sig• nificantly from the more traditional, diverse plots. The government program requires hillsides planted to the new rubber to be terraced; traditional plantings, on the other hand, left the soil largely intact. Until recently, interplanting of other crops among the new rubber seedlings was forbidden. Reportedly, rules on interplanting have of late been relaxed. The rubber plantations are fertilized with chemical fertiliz- 8 Society and Non-timber Forest Products in Tropical Asia

ers; the traditional system used no commercial chemicals and required less labor input. Most households in Tae ham• let have received credit as participants of the planting pro• gram, although gardens have not yet been designated. Apart from the three types of managed vegetation out• lined earlier, many other kinds of forest-gardens, managed forests, and agroforests can be found in and around Tae. Throughout the village, small specialized plots are found, often making use of particular edaphic conditions such as poorly drained soils or river banks. Among these are associ• ates of sago palms, pandanus, and other useful species that do well in waterlogged soils.

MANAGEMENT The three types of managed forest—tembawang, tanah adat,

PRIORITIES and tanah usaha—differ not only in the way they were cre• ated but also to some degree in their subsequent manage• ment. Like many similar sites in other tropical areas (Alcorn 1984; Denevan and Padoch 1988; Padoch and de Jong 1987), all are managed quite casually, with occasional slash-weed• ing as the most important task. The timing and intensity of clearing, however, differ somewhat with the species that make up the plot. Areas rich in fruits such as durian and tengkawang are usually slash-weeded once a year to clear the ground where ripe fruits are likely to fall. Plots that in• clude rubber trees and sugar palms are cleared at least par• tially whenever vegetation limits access to tapping. Weeding, although frequently done to enhance ease of harvest, none• theless has the effect of specifically removing undesirable species and promoting useful ones. Casual management, such as removing vines or other competing plants near an eco• nomic plant—when and if it is noticed—may also play an important role in traditional management systems. Not only are management schedules variable for the three types of managed stands, but they are also distinguished by different management priorities. Tanah usaha, or enter• prise plots as their name suggests, are largely cash-producing areas. Dominated by rubber in Tae and by other commercial Creating the Forest 9 crops in other villages, the composition and even existence of these plots change in response to market opportunities. As mentioned, rubber entered these regions only in the early 1940s. Structure and species diversity in tanah usaha vary greatly from plot to plot. Areas closer to the village are often far more diverse in species, resembling tembawang in the number and kinds of fruit trees present. Those too far away to be conveniently harvested contain fewer fruit trees. Tembawang, on the other hand, are by definition rich in fruits. Most include a variety of the familiar domesticated fruits of Southeast Asia (e.g., rambutan, langsat, mangosteen, and always durian). Tembawang are also obviously managed for diversity. One io-by-200-meter transect through a tembawang revealed that among 224 individual trees or stands of bamboo 5 centimeter dbh and greater, 44 species were represented. Thirty of the species had edible fruits or shoots. Many of these were semidomesticated or wild rela• tives of the well-known market fruits,- especially common are relatives of mango and rambutan. Tara'n forest managers claim to have occasionally planted up to seventy-four differ• ent fruit species. They identified more than 100 species of edible fruit in managed forests. Many fruit trees also have value as construction timbers and are also used medicinally. Although tanah usaha have been singled out as com• mercially oriented plots, the other types also contain prod• ucts that are sold. Much, if not most, tengkawang, an important "forest product" export of West Kalimantan, are harvested from managed tembawang. Except for tengkawang, for the past ten years Tae villagers have been marketing durian fruits in quantity. In 1991 on just one day (February 1) at the height of the durian season, more than 10,000 fruits were brought to market from four of Tae's five hamlets. Even at the minimal price of about 15 cents each, the income many families earned throughout the month-long season was sub• stantial. Tembawang also contain species specifically managed for their quality as construction materials. The tanah adat reserves, however, are usually cited as the areas from which io Society and Non-timber Forest Products in Tropical Asia

community members get most of the timbers needed to build houses. Many households in Tae have built new houses within the past five years so the reserves are quite depleted. Nevertheless along a io-by-200-meter transect through a tanah adat in Tae were found 55 species among 229 indi• vidual trees, with 30 of the species pointed out as particu• larly good construction wood.

CONCLUSIONS If the managed woodlands of Tae—and of many other com• munities in the forested tropics—are indeed large, diverse, well-managed, and profitable, why have they attracted so little comment and investigation in the past? The reasons are sev• eral. The apparent casualness of management is doubtless a factor explaining the "invisibility" of traditional forest man• agement. None of the or agronomists that staff pro• vincial or regional government offices acknowledge that what Tae residents do should be called forest management. Not even Tae villagers themselves would assert they manage for• ests in any technical sense. Perhaps the most important factor making traditional management invisible is the appearance of the forest. Tae's woodlands, and especially the tembawang, do not look "man• aged." The plantings are not even-aged, not evenly spaced, and appear haphazard. The most important factor in their invisibility is doubtlessly that the traditional stands are ex• tremely diverse. Spontaneous vegetation is not only toler• ated but encouraged in plots. Only when that diversity is valued as a desirable characteristic in agroforestry and for• estry will the woodlands of Tae be appreciated. But for now the Tae woodlands are far too managed for those interested in conservation and not managed enough for those interested in production. They are dismissed as primitive by many, and they are threatened by encroaching monoculture plantations of oil palm and improved rubber. In Tae the forest areas, managed or otherwise, are quite small; thus, no commercial timber concessionaire would likely seek the right to log them. But forests used by Dayaks Creating the Forest n

in other parts of Kalimantan, such as the village lands of traditional forest peoples in many tropical areas, are being exploited or threatened by commercial concerns. The recog• nition that many of those forests are managed and already governed by laws assigning access to particular resources to particular groups is important if these peoples are to gain control over their territories. When forest resources are deemed to belong to the state and only agricultural or man• aged resources can be claimed by the people, it is important to recognize and focus on the management rather than on the extraction of forest resources.

Note: A revised version of this paper was published in Fo• rum for Applied Research and Public Policy 7(4): 11, 1992. Reproduced with permission.

REFERENCES Alcorn, J. B. 1984. Huastec Maya man-made tropical forest of the ethnobotany. Austin: University Maya. Interciencia 12:10-15. of Texas Press. Gordon, B. L. 1969. Anthropogeog- Chin, S. C. 1985. Agriculture and raphy and rainforest ecology in resource utilization in a lowland Bocas de Toro Province, Panama. rainforest Kenyah community. Office of Naval Research Report, Sarawak Museum Journal Special Department of Geography, Monograph No. 4, Kuching. University of California, Berkeley.

Conlclin, H. C. 1957. Hanunoo Hecht, S. B. 1982. Agroforestry in agriculture. Rome: FAO. the Amazon Basin: Practice, theory, and limits of a promising Denevan, W. M., and C. Padoch, land use. In Amazonia: Agricul• eds. 1988. Swidden-fallow ture and land use research, ed. S. agroforestry in the Peruvian B. Hecht, 331-71. Cali, Colombia: Amazon. Advances in Economic CIAT. Botany No. 5. Bronx: New York Botanical Garden. Kunstadter, P., E. C. Chapman, and S. Sabhasri, eds. 1978. Fox, J. J. 1977. Harvest of the Farmers in the forest: Economic palm: Ecological change in development and marginal eastern Indonesia. Cambridge, agriculture in northern Thailand. Mass.: Harvard University Press. Honolulu: University Press of Freeman, J. D. 1970. Report on the Hawaii. Iban. London: Athlone Press. Michon, G. 1985. De l'homme de Gomez-Pompa, A., J. S. Flores, and la foret au paysan de 1'arbre: V. Sosa. 1987. The "Pet-Kot": A Agroforesteries Indonesiennes. ii Society and Non-timber Forest Products in Tropical Asia

Ph.D. diss., University of . 1985. Indigenous Mont pel lier, France. management of tropical forest ecosystems: The case of the Morris, H. S. 1953. Report on a Kayapo Indians of the Brazilian Melanau sago-producing commu• Amazon. Agroforestry Systems nity in Sarawak. Her Majesty's 3:139-58. Stationery Office, London. Sandin, B. 1967. The sea Dayaks Padoch, C. 1982. Migration and of Borneo before white rajah rule. its alternatives among the Iban of East Lansing: Michigan State Sarawak. Instituut voor Tall-, University Press. Land-, en Vokenkunde, Verhande- lingen 98, Marinus Nijhoff, Sutlive, V. H., Jr. 1978. The Iban of Leiden. Sarawak. Arlington Heights, 111.: AMH Publishing. Padoch, C, and W. de Jong. 1987. Traditional agroforestry practices Torquebiau, E. 1984. Man-made of native and ribereno farmers in dipterocarp forest in Sumatra. the lowland Peruvian Amazon. In Agroforestry Systems 2:103-27. Agroforestry: Realities, possibili• Weinstock, J. A. 1983. Rattan: ties, and potentials, ed. H. L. Ecological balance in a Borneo Gholz, 179-94. Dordrecht, The rainforest swidden. Economic Netherlands: Martinus Nijhoff/Dr. Botany 37(1): 58-68. W. Junk Publishers. Wiersum, K. F. r982. Tree Posey, D. A. 1984. A preliminary gardening and taungya on Java: report on diversified management Examples of agroforestry of tropical forest by the Kayapo techniques in the humid tropics. Indians of the Brazilian Amazon. Agroforestry Systems 1:53-70. In Ethnobotany in the neotiopics, ed. C. T. Prance and J. Kallunki. Advances in Economic Botany No. 1. Bronx: New York Botanical Garden. CHAPTER 2

J. Peter Brosius

Bornean Forest Trade in Historical and Regional Perspective: The Case of Penan Hunter-Gatherers of Sarawak

Hunter-gatherers have long occupied a specific niche in the economies of Southeast Asia. These people have been a ma• jor source of forest products, which are traded downriver and shipped to the coast for consumption or export. Only in the past two to three decades have we come to fully appreciate the role of hunter-gatherers in these regional trade econo• mies. In that time, many studies have been carried out, from both upriver and downriver perspectives (Peterson 1977, 1978a, 1978b; Dunn 1975; Estioko-Griffin and Griffin 1975; Endicott 1974; Hutterer 1974, 1976, 1977; Benjamin 1973; Rambo 1979; Fox 1969). The focus of this paper is on the collection and trade of forest products by Penan hunter-gatherers in Sarawak, East Malaysia. Variations in two dimensions of this trade are par•

ticularly examined: (1 ] change, or variation through time; and (2) regional variation, or variation through space. With re• spect to the former, this paper will discuss the kinds of changes that have occurred in the types of commodities traded, as well as some of the factors that have led to these changes. With respect to variation in space, this paper will show that there is indeed considerable local-level variation in the types and volume of commodities traded. The assump• tion is that there is a high level of uniformity in patterns of trade in any particular region of Southeast Asia where we observe trade between foragers and farmers. This may well be true in many areas, but it is not the case in Sarawak, and we should be looking for such variation elsewhere as well. 14 Society and Non-timber Forest Products in Tropical Asia

Consideration of the trade issue in Borneo is particu• larly significant in light of Hoffman's The Punan: Hunters and Gatherers of Borneo (1986). My goal is not to address the ethnogenesis issue of Bornean hunter-gatherers, but some mention of Hoffman's work must be made. Hoffman con• tends that the existence of peoples such as Penan in Borneo is explained exclusively with reference to their role as pro• viders of forest products in local trade networks. Clearly, for Penan, trade has long played a key role. But Hoffman has taken this argument to absurd lengths, claiming that Penan/ Punan exist primarily to trade and that trade is their raison d'etre. The particulars of this debate will not be dealt with here except to say that Hoffman's contentions are informed by fragmentary and incorrect data (see Brosius 1988; Sellato 1988; Kaskija 1988). Trade is critically important to the Penan, and presumably has been for hundreds of years and probably longer,- yet trade alone does not explain their existence.

THE PENAN AND Throughout the interior of central Borneo are groups of OTHER BORNEAN hunter-gatherers known as the Penan or Punan. From 1984

HUNTER- to j^gy j conducted field research among one such group, GATHERERS the Penan Geng, who inhabit the Usun Apau plateau between the Balui and Baram rivers. The term Penan applies to two primary populations, which Needham (1972) called Eastern and Western Penan. The Western Penan, of which the Penan Geng are included, are all those in the Belaga district, as well as communities in the Silat River watershed and at Long Beku. Eastern Penan are found in the Baram district, mostly to the north and east of the Baram River. Sago starch, derived from the palm Eugeissona utilis, has traditionally been the primary source of carbohydrates for Penan. The meat and fat provided by the bearded pig makes up the other major part of the Penan diet. One aspect of Penan subsistence requires emphasis: whereas most Southeast Asian hunter-gatherers are involved in trade for forest products in order to supplement their diets with carbohydrates, Penan Bornean Forest Trade in Historical and Regional Perspective 15

trade only to acquire tobacco, salt, and manufactured goods. With respect to food, Penan are wholly self-sufficient and do not require supplements from trade with longhouse peoples. With the strong encouragement of government officials during the 1950s and 1960s, Penan began to settle in increas• ing numbers and to take up agriculture. Today only about 5 percent remain fully nomadic. The rest live along a con• tinuum of sedentariness and rely on cultivated foods.

THE GENERAL Contrary to Hoffman's amazement, Penan have long been NATURE OF recognized as suppliers of forest products to regional trade TRADE RELATIONS networks. What I would like to do here is consider the na• ture of these networks and of Penan trade relations with their longhouse trade partners. The relationship between longhouse peoples and neigh• boring Penan was and is essentially paternalistic. In describ• ing these relationships, Penan employ the term tagung, a word that implies caring for the physical needs of another— feeding, clothing, and taking general responsibility. Tagung connotes dependence. Penan depended on longhouse peoples, particularly aris• tocrats, to act as mediators or conduits by which both infor• mation and material goods from the outside world reached them. This relationship was a variation of what has been called a patron/client relationship, which involved a series of reciprocal debts and obligations. Where this relationship differed was in the fact that Penan always had the option of dissolving the relationship simply by moving to another wa• tershed. Penan trade with longhouse peoples was therefore something of a seller's market. There was always some com• petition between longhouses for monopolizing relationships with Penan {e.g., the headmen of various longhouses would invite a Penan band to remain in the headwaters of nearby rivers). Hoffman gives the impression that relationships be• tween nomadic and sedentary communities are long-lasting. In reality, the length of time that Penan maintain relation- 16 Society and Non-timber Forest Products in Tropical Asia ships with particular longhouse communities is highly vari• able. Such relationships may endure for several generations, but they are equally likely to be intermittent or severed after relatively brief periods. The Penan Geng, over the past cen• tury, have variously had close relationships with Seping, Sebop, Lirong, Lahanan, Sambop, Uma Pawa, Kayan, and oth• ers. The primary determinant of just how long such relation• ships persist seems to be a matter of how well Penan get along with their trade partners and how well aristocrats/pa• trons treat Penan in their vicinity. If close and friendly rela• tions are maintained, Penan may be willing to go to some length to maintain such relationships, even to the extent of moving great distances to follow a longhouse that has moved. In other cases, relationships are of short duration and may be severed because of Penan dissatisfaction with trading part• ners, particularly if Penan feel they are being cheated or pressed too hard to repay debts.

One of the major features of the Penan/longhouse rela• tionship was indebtedness. Penan were (and are| always in• debted to longhouse people, never vice versa. When traders visit Penan, they bring many goods, which are often disbursed on the promise of later returns from Penan, creating indebt• edness. Often, visiting traders extract products of previous promises. Maintaining contact with nomadic Penan was some• thing of a specialized enterprise. Although longhouse peoples may know well-traveled routes in the forest, they lack the detailed knowledge of the interior landscape that Penan have. However, in every longhouse there are men with a knowl• edge of interior rivers, river names, and the trails connecting various places—these are the specialists in contacting Penan. When visiting longhouses or when being visited by longhouse peoples, Penan would inform their trade partners where they were likely to be found in subsequent months. The trade in products collected or manufactured by Penan generally occurs in one of two ways. Either Penan transport goods to longhouses and market centers such as Belaga, or longhouse traders travel to Penan. For goods trans- Bornean Forest Trade in Historical and Regional Perspective 17

ported to them, Penan get very little in return. A mat may bring only two sarong. An ajat basket may bring a small amount of tobacco, worth perhaps US$i or US$2 in Belaga. Clearly, for the Penan, this type of trade is not profitable. As Hoffman describes them, hunter-gatherers such as Penan are forever in search of trade goods. This certainly is not the case in Sarawak. Rather, trade-related collection and production usually occur in bursts of activity in anticipation of a trading trip. When such a trip is anticipated, several fami• lies move to an area where, for instance, rattan is plentiful. Once sufficient rattan has been collected, or mats and bas• kets produced, the group returns to its base camp or commu• nity.

CHANGE Contemporary Penan have a rich oral historical tradition, and

THROUGHTIME a theme that appears repeatedly in their accounts are descrip• tions of the production of forest products for trade. What is clear from these accounts is that the range of forest products collected has changed considerably through time. Three commodities occur repeatedly in the oldest his• torical accounts: , belian (ironwood) shingles, and rhinoceros horn. Camphor, crystals of which were extracted from trees of the genus Dryobalanops, has not been produced by Penan since the early part of this century, and only the oldest informants recall having ever produced it. It seems to have been a time-consuming enterprise, particularly locat• ing trees that had a sufficient store of crystals. Penan gave it up because, they explain, it simply was not worth it. They likewise gave up belian shingles, which are produced from wood of the species Eusideioxylon zwageri. The Penan Geng produced belian shingles in the earlier part of this century near the Belaga River. Production, however, was difficult, time-consuming, and unprofitable that it was no longer worthwhile to engage in this activity. Apparently, from Penan descriptions, inflation was occurring or, as they explain it, the ringgit (Malaysian money) became "less heavy." Yet they were not paid more. Finally, there was rhinoceros horn. This 18 Society and Non-timber Forest Products in Tropical Asia commodity was a profitable enterprise, but became unten• able as the rhinoceros [Diceiorhinus sumatrensis) was de• pleted. This depletion appears not to have been brought about by Penan. In the late nineteenth and early twentieth centu• ries, Iban men on bejalai traveled extensively in the Usun Apau and elsewhere in the Ulu Balui, in part to search for rhinoceros horn. They were apparently present in sufficient numbers to drive the rhinoceros to extinction. The last rhi• noceros disappeared from the area in the 1930s. One product considerably important to Penan up through the mid-twentieth century was damai, the of the massive tree Agathis borneensis. This resin, used in manufacturing vamish, made up a considerable percentage of the total trade volume. Collecting damai was primarily men's activity. As it is explained today, several households would establish forest camps in areas where damai was found and spend several weeks or months collecting it. The task of collecting damai is not all that time-consuming, but the job of transporting it to market is exceedingly difficult. In bring• ing it down to trade, groups of men would carry baskets weigh• ing up to 45 kilograms (100 pounds). The trip to where it could be loaded onto boats required two to five days of ardu• ous overland hiking. Today, because prices are so low, damai has ceased to become a significant item of trade. Finally, a product that was important up until about i960 was jelutong [ketipe], a hard latex derived from trees of the genus Dyera. Although still used extensively by Penan, it is now almost never produced for trade. As with other prod• ucts for which collection has been abandoned, Penan claim that the price is no longer high enough to justify collection. Increasingly in the current century, Penan have become known for producing fine woven rattan mats and baskets. Trade in these items continues to be the primary source of cash and manufactured goods for Penan in the Belaga district today. One reason the Penan can produce such high-quality rattan products, in addition to their skill, is that they have access to the best qualities of rattan in the extensive areas of primary forest that they occupy. Bornean Forest Trade in Historical and Regional Perspective 19

The collecting, processing, and weaving of rattan is a lengthy process to which the Penan devote a great deal of time. Collecting is the least time-consuming part. Penan men may make brief, one- to three-day trips into the forest, trans• porting unprocessed lengths of rattan back to the settlement. Alternatively, several family groups may establish a satellite camp that requires one to two days of travel from the main settlement, remaining at the camp to collect and process rat• tan. Processing involves splitting individual lengths of rat• tan into strips and shaving these down to the width and fi• nesse necessary for weaving. After drying, large bunches of strips are transported to the main settlement for weaving over the next several months. Among Penan, women are the exclusive processors and weavers of rattan, and as such, they play a key role in the Penan trade economy. Women vary greatly in their skills at weaving. Most reach a skill level where they can weave de• signs freehand without reference to another basket, and they are competent at producing plain or cross-hatched mats. Only about 20 percent of Penan women reach a point where they are able to improvise designs. The time it takes to weave a mat or basket varies de• pending on size, complexity of design, and skill of the weaver. Just preparing the rattan for a large mat takes a minimum of one week for one woman. As for the actual process of weav• ing, if done full time, an ajat basket takes about two days, whereas a mat takes perhaps two weeks. Actually, however, mats and baskets are rarely worked on full time. Considering the time it takes to produce various rattan items for trade, the return to Penan is extremely low. They generally receive US$4 for a basket, although this amount may vary from US$3 to US$5 for variations in size and qual• ity. They generally receive US$24 for a plain mat, US$30- $32 for a cross-hatched mat, and US$38-$40 for a mat with complex design. Finally, one important commodity from the 1970s is the kayu gaharu. This aromatic wood, derived from the ge• nus Aquilaria, still provides the Penan some cash income, io Society and Non-timber Forest Products in Tropical Asia but not as much as in the 1970s. During the 1970s and early 1980s the garu economy boomed throughout Sarawak due to increased demand and rising prices. During this time, both Penan and their longhouse neighbors scoured the forests. Many hundreds of dollars could be made on a single trip to the forest, which usually lasted several weeks. The Penan describe this period in glowing terms, saying they were much richer then than they are now. Today the gam trade has de• creased considerably. Not only have prices declined but garu itself has been drastically depleted. Not only the commodities and the volume of those commodities have changed through time, but patterns of trade have changed as well. Today traditional relationships between Penan and longhouse peoples are breaking down for numer• ous reasons. One factor in the transformation of traditional relationships has been the expansion of the cash economy and the extension of government institutions, which has undermined the authority of longhouse aristocrats. What appears to be the most important factor in the breakdown of traditional trade networks is the appearance of logging companies in the area. The simple physical exten• sion of logging roads makes access to market centers such as Belaga much easier for Penan and allows them to bypass their former longhouse trade partners. Penan simply prefer to take the easiest route to market and like the access this gives them to a wider range of goods. As they have traveled more frequently to Belaga, Penan have developed increasingly close ties with towkays (traders) there. The significance of these relationships is destined to grow, just as the role of longhouse aristocrats/traders will shrink. New sources of income available to both Penan and longhouse peoples have had some impact on the regional trad• ing economy as well. The opportunity to sell meat and fish to towkays traveling upriver with refrigerated boats is one such new source. As logging continues to expand in the area, large numbers of men seek employment in this industry. For longhouse people, the income derived from trade with Penan is not as crucial as in the past, when it was the only in Bomean Forest Trade in Historical and Regional Perspective 21

town. This is not to say that trade in forest products is be- coming.any less important as a source of income to Penan. The current disruption of traditional trade relationships is of no great consequence for Penan. They simply go to wher• ever they can most profitably and easily trade their products. They are not particularly bound by affective ties to their tra• ditional longhouse trade partners and are not in the least bothered by the gradual dissolution of these relationships. In the past, Penan bands have frequently dissolved relationships with particular longhouse communities and initiated new relationships. The transformations occurring today are merely a variation of this older pattern.

VARIATION Regional variations in trade seem to have been largely ig-

IN SPACE nored in studies of the Penan, perhaps because they are in• visible from a downriver perspective. Among contemporary Penan, patterns of trade vary considerably in different com• munities and watersheds. The most obvious factor producing local-level variation is the distribution of particular types of resources. There is a considerable degree of microenvironmental variation in the forests of Borneo, a fact Penan are very well aware of. Some plants may be abundant in one watershed but absent in an• other. This is as true for trade resources as for other plant and animal species. Three examples serve to illustrate this point: bezoar stones, damai, and belian wood. Bezoar stones, extracted from the gall bladder of mon• keys and porcupines, still play some part in Penan trade and continue to realize very high prices. The stones, however, are not found everywhere in the Belaga district but only in certain watersheds. For instance, they are never found near the Seping River, but do occur in the headwaters of the Linau and Balui rivers where the Punan Busang occasionally pro• cure them. What accounts for this uneven distribution has not been determined. The distribution pattern of Agathis boineensis, the source of damar, is even more localized, occurring in dense 2i Society and Non-timber Forest Products in Tropical Asia concentrations only in certain areas. According to both Penan and the forest ecologist Richards, whose classic 1936 study of the flora of Mount Dulit was carried out on the northern margins of Penan Geng territory, great concentrations of Agathis exist in the upper Koyan River, a tributary of the Belaga River. Richards reports that in one sample plot this species comprised 35.2 percent of all trees over 406 millime• ters (16 inches) in diameter. Dense stands are also said by Penan to exist near the western edge of the Usuan Apau East• ern Tableland. As noted, when Agathis was still an impor• tant trade item, Penan would travel some distance to estab• lish camps near Agathis stands. Finally, with respect to belian wood, this species is very localized in occurrence in the Belaga district. Although it also grows elsewhere in lower elevations of the district, its greatest concentration is in the middle reaches of the Belaga River. A second factor influencing variation in patterns of trade is the population density of foragers in different watersheds. This affects not so much the types of commodities traded, but trade volume. In the Belaga district today, there exists marked variation in volume of trade and in the frequency with which traders have contact with Penan. The outcome here is that rivers with a higher population of Penan attract traders. This creates a sort of feedback loop by which, be• cause of frequent trader visits, the Penan in these rivers pro• duce more rattan products. The motivation for trading with more populous rivers is clear, since by virtue of the higher trade, volume traders make higher profits. The river with the highest trade volume seems to be the Linau River (population 439), which has three Penan settlements, and the Punan Busang settlement of Long Lidem. All these settlements in the Linau River produce large quan• tities of woven rattan goods. The frequency with which trad• ers visit this river is remarkable—when river levels are suit• able, traders visit weekly. Each trade expedition usually brings a minimum of fifty ajat baskets and twenty mats. In the Seping River (population 150), by contrast, only Bornean Forest Trade in Historical and Regional Perspective 23

small numbers of these items are produced, and it is infre• quently visited by traders. This area was visited by traders only twice in the three years I was there. When one sees how little traders coming to the Seping are able to procure—five or six baskets and, if they are lucky, one or two mats—it is clear why they do not come. The goods procured are not enough to make the long and difficult three-day hike worth• while. Finally, just as the acquisition of certain skills has pro• duced changes in patterns of trade through time, so too does this produce variation in patterns of trade on a regional level. Two trade commodities may serve to illustrate this, namely, woven rattan products and parang (a type of knife) manufac• ture. The weaving of high-quality mats and baskets with elaborate design is largely a Western Penan phenomenon. The Eastern Penan style of weaving involves only black and white banded designs and the workmanship is generally not nearly as fine. Not only is Eastern Penan weaving technically less developed than that of Western Penan but, as a whole, the volume of items produced is much lower. The Western Penan are also skilled metalsmiths and today produce large numbers of parangs. They use bellows to soften the metal, which is then pounded into shape and filed to its final form. The production of a single blade takes about one day. In the Belaga district, the parangs that Penan produce are mostly for personal use and are not extensively traded. This practice contrasts markedly with Western Penan communities in the Baram district. Penan in the Silat Raver and at Long Beku have essentially specialized in producing parangs in large numbers. Thus, they are probably the main suppliers of parangs in the Baram district.

CONCLUSIONS The preceding discussion considered changes through time in patterns of trade, as well as contemporary variation on a regional level. Let me now summarize some of the major factors that have led to changes in the types and quantities 24 Society and Non-timber Forest Products in Tropical Asia of forest products traded by Penan. Apparently, from Penan statements, the most important factor influencing this trade has been the market value of various commodities—either falling prices or stable prices in the face of inflation. A sec• ond factor has been resource depletion, as occurred in the cases of rhinoceros horn and garu. Finally, the acquisition of new skills has been a factor. The weaving of mats and bas• kets, particularly, is a skill that Penan began to learn only toward the end of the past century. One factor that does not seem to have been of any significance to transition in the types or quantities of trade items has been settlement and the adoption of agriculture by the Penan. With respect to a general direction of development, we can recognize a trend away from extraction alone to manu• facture, as with the production of rattan products. Concomi• tant with this has been an increase in the role of women in producing items for trade. Whereas trade based on the ex• traction of forest resources—whether camphor, jelutong, or damar—was primarily the province of men, the contempo• rary trade economy is based largely on the contribution of women. With respect to these patterns of historical change, the main shortcoming in the material presented here is that it represents but a brief slice of time in the trading history of the Penan. What is lacking is real-time depth, of the sort that can only be provided by archaeological investigation. With respect to spatial variation, it has primarily been my goal here simply to point out that this exists and to ex• amine some of the factors that contribute to this variation. As to the relevance of such variation, one is that it undoubt• edly had some effect on regional settlement patterns. One of the primary determinants of longhouse locations at the mouths of major rivers was the control or monitoring of trade with hunter-gatherers. Obviously, from the perspective of trade with hunter-gatherers, some locations are more suit• able than others. Unfortunately, we know very little about how settlement decisions of longhouse peoples are actually made. Bornean Forest Trade in Historical and Regional Perspective 15

There are undoubtedly forms of spatial variation not visible on the scale I am examining here. Perhaps among the most important is the overall pattern of distribution of hunter- gatherers in Borneo. A look at a distributional map of hunter- gatherers in Borneo shows them to occur in an east-west band through the middle of the island; they are absent from Sabah and from the southern part of Kalimantan. Certainly there are (or were) abundant forests in those areas where no hunter- gatherers are now found, as well as a number of important entrepdt trade centers. Two possible explanations account for this distribution of hunter-gatherers. One is in the distri• bution of the sago palm in Borneo and major source of food for the Penan. At present not enough is known about the distribution of this species to make a definitive statement on such a relationship. The other possibility, suggested by Bernard Sellato (1988), relates to the distribution of Kayan, Kenyah, and Kajang peoples, whose particular form of rela• tionship with hunter-gatherers has served to maintain these groups as distinct populations. Ethnic groups such as the Iban compete directly with hunter-gatherers (e.g., the Penan) in their forms of extraction and may absorb these distinct popu• lations in their patterns of marriage. However, the Kayan, Kenyah, and Kajang peoples do not normally engage in ex• tracting forest products, nor do they absorb hunter-gatherers through intermarriage. These, however, are only suggestions and await testing in future work.

REFERENCES Benjamin, G. 1973. Introduction. resource utilization in modern In Among the forest dwarfs of and ancient Malaya. Monographs Malaya, ed. P. Schebesta, 131-53. of the Malaysian Branch of the Kuala Lumpur: Oxford University Royal Asiatic Society No. 5, Kuala Press. Lumpur.

Brosius, J. P. 19S8. A separate Endicott, K. 1974. Batek Negrito reality: Comments on Hoffman's economy and social organization. The Punan: Hunters and Ph.D. diss., Department of Gatherers of Borneo. Borneo Anthropology, Harvard University. Research Bulletin 20(2): 81-106. Estioko-Griffin, A., and P. B. Dunn, F. L. 1975. Rainforest Griffin. 1975. The Ebuked Agta of collectors and traders: A study of northeastern Luzon. Philippine 16 Society and Non-timber Forest Products in Tropical Asia

Quarterly of Culture and Society Peterson, J. T. 1977. Ecotones and 3(4): 237-44. exchange in northern Luzon. In Economic exchange and social Fox, R. G. 1969. Professional interaction in Southeast Asia: primitives: Hunters and gatherers Perspectives from prehistory, of nuclear South Asia. Man in history, and ethnography, ed. K. L. India 48:139-60. Hutterer, 55-71. Ann Arbor: Hoffman, C. 1986. The Punan: Michigan Center for South and Hunters and gatherers of Borneo. Southeast Asian Studies. Ann Arbor: UMI Research Press. . 1978a. The ecology of Hutterer, K. L. 1974. The social boundaries. Urbana: evolution of Philippine lowland University of Illinois Press. societies. Mankind 9:187-99. . 1978b. Hunter-gatherer/ . 1976. An evolutionary farmer exchange. American approach to the Southeast Asian Anthropologist 8o|i): 335-51. cultural sequence. Current Rambo, A. T. 1979. Human Anthropology i7|2|: 221-42. ecology of the Orang Asli: A , ed. 1977- Economic review of research on the exchange and social interaction in environmental relations of the Southeast Asia: Perspectives from aborigines of peninsular Malaysia. prehistory, history, and ethnogra• Federations Museum Journal phy. Paper No. 3. Ann Arbor: 24:41-71. Michigan Center for South and Richards, P. W. 1936. Ecological Southeast Asian Studies. observations on the rain forest of Kaskija, L. 1988. Carl Hoffman Mount Dulit, Sarawak. Parts 1 and and the Punan of Borneo. Borneo 2. Journal of Ecology 24:1-37, Research Bulletin 10(2): 121-29. 340-60.

Needham, R. 1972. Penan-Punan. Sellato, B. J. L. 1988. The nomads In Vol. 1, Ethnic groups of insular of Borneo: Hoffman and "devolu• Southeast Asia, ed. F. M. Lebar, tion." Borneo Research Bulletin 55-69. New Haven: HRAF Press. io{iY 106-1 r. CHAPTER 3

Rash mi Pachauri Rajan

Sal Leaf Plate Processing and Marketing in West Bengal

The West Bengal State Forest Department (WBFDJ has used its social forestry program to bring about a shift from a reli• ance on custodial policing and -controlled timber pro• duction systems to collaborative management with local vil• lages and communities. This transition is based on the rec• ognition that the state's shrinking forests can only be saved through cooperative management response to pressures from the increasing population of the state.

HISTORICAL A pilot project was started in the Arabari Range in the 1970s. DEVELOPMENT The project demonstrated that if villagers protected the natu• OF IOINT FOREST ral sal forests of that region from fuelwood cutting, the trees MANAGEMENT had the ability, through hardy rootstocks, to regenerate rap• SYSTEMS idly. The Forest Department would give the villagers 25 per• IN BENGAL cent of all revenue generated from the sale of fuelwood and timber in return for protecting the forests. The Arabari experience turned out to be a major suc• cess story; the returns on investment were impressive. Rs 1,000,000 were spent on forest-related work in this area in the fourteen years that the project was monitored, and by 1986 the value of fuelwood and timber rose from almost noth• ing to Rs 12,600,000 (Poffenberger 1990). Under the project, new job opportunities were created, forester-villager relations improved, natural forests recovered rapidly, and Forest De• partment revenues increased. Encouraged by its success, se• nior forest officials advocated extending the project and its 28 Society and Non-timber Forest Products in Tropical Asia

approach to other similar ecological zones throughout South• west Bengal. By year-end 1989, an estimated 1,500 commu• nity-based committees (FPC) had been formed to protect 152,000 hectares of forest land. Of this, 100,000 hectares was regenerating sal [Shoiea xobusta) 5 to 10 meters high. Productive companion species flourished, resulting in increased income to collectors, especially women (Poffenberger 1990). Other communities, concerned about the rapid degra• dation of forests, began to form FPCs after seeing the success of these programs. Neighboring communities were often motivated to become involved as well and were encouraged to form committees by the Forest Department range and beat officers. Consequently, the Forest Department now has nearly 2,000 registered FPCs and hundreds more are being formed. This has led, besides the aforementioned improvements, to a remarkable transformation in the natural forest landscape with a minimal capital investment from the Forest Depart• ment. Income estimates indicate that wood forest products generate substantially more cash (25 percent) to forest dwell• ers than the amount received from pole sales. Forest dwell• ers are further attracted to non-timber forest products (NTFPs) because they are available on a continuous or at least a sea• sonal basis. The West Bengal Forest Department (WBFD) is a pioneer in this new approach to joint management for forest protection. But as the project is relatively new, WBFD is still developing and perfecting joint forest management systems.

PRODUCTION The objective of this undertaking was to gain insight into AND MARKETING the role that NTFPs play in the local economy of the regions POTENTIAL studied. The production, marketing, and sale of sal thai plates OF SAL THAI (leaf plates) in the Midnapore district of West Bengal were PLATES investigated with respect to the economic effects the pro• duction process had on the local communities, as well as the role the Forest Department played in the process. Sal Leaf Plate Processing and Marketing in West Bengal 29

METHODOLOGY

Background information was first collected through indi• vidual discussions and analyses of articles and papers. Two trips were made to Jhargram in Midnapore, where lengthy discussions were held with concerned forest officers, retail• ers, and wholesalers regarding the sale and marketing of sal thai plates. A number of villages were visited, and villagers involved in producing thals and plates were interviewed to gather information on production. Semi-structured inter• views, including questionnaires and individual and group discussions, were used to collect primary data. Two sal plate- making private units were also visited, and lengthy discus• sions were held with the proprietors regarding costs, produc• tion methods, marketing channels, and profits. Wholesalers and retailers in various bazaars in Calcutta were also visited and informally interviewed since Calcutta is one of the final markets for the end product. Other related studies were ex• amined as a base for background information and data.

BACKGROUND

Sal [Shorea robusta) is a monsoon forest tree prevalent in the forests of India, Pakistan, and Bangladesh, and second only to as an important timber tree in South Asia. In India it grows mainly in Bihar, Orissa, Madhya Pradesh, West Ben• gal, Assam, and Uttar Pradesh states (i.e., in two distinct ecozones—the sub-Himalayan region north of the Gangetic Plain, and eastern India south of the Ganges). The tree grows to about 40 meters with a girth of approximately 3.5 meters (and as much as 7 meters in favorable conditions). It starts flowering after twenty-five years. Sal provides a number of useful products besides tim• ber, including oilseeds, oleoresin, and leaves. The leaves are made into leaf plates and thals, which are used mainly for ceremonial occasions and in sweet shops all over India. Leaves 30 Society and Non-timber Forest Products in Tropical Asia are collected when coppice regrowth is thick and until the tree is about three years old. In the third year, most of the coppice growth is removed, and only two to three larger shoots are left to grow into poles (timber). At this stage leaves are not collected because very little coppice is available, and the leaves on the growing stems are inaccessible. In its tenth year, the tree is cut back to allow it to resprout, and the cycle begins once more. According to a recent study conducted in Chingara and Khas Jungal of West Bengal, 5 to 10 percent of the total leaves are regularly harvested (between April and June and then again between August and October). Insect consumption of the total leaf production of sal forests in this area is rarely greater than 9 percent. About 8 percent of the leaves are lost (leaf-drop) just before the rains. These natural leaf losses may prove to be more detrimental to the forest than from harvesting by villagers. Therefore, the study recommends taking "neces• sary measures against excessive attack (of insects), rather than against the harvesting of trees by villagers for ensuring pro• tection of sal forests." Information and data for this study have been collected from Southwest Bengal (in particular, from Jhargram in Midnapore district), forest officers, wholesale and retail mar• kets, and villagers involved in producing sal plates and thals. A district-by-district breakdown of the area in Southwest Bengal covered by sal forests is listed in Table 3.1. The average area of sal as a percentage of the total for-

Table 3.1 Sal forest areas in Southwest Bengal

Sal forest Total forest District (ha) Area (ha) As % of total

East Midnapore 81,082 45,000 55.5 West Midnapore 89,108 70,000 78.6 North Bank lira 73,688 44,910 60.9 South Bankura 66,042 30,263 45.8 Purulia 92,254 47,049 51.0

Total 402,174 237,222 58.9

Source: Government of West Bengal {1988). Sal Leaf Plate Processing and Marketing in West Bengal 31 est area in the three districts (see Table 3.1) is about 59 per• cent; West Midnapore has the largest percentage. In West Bengal approximately 90 percent of forest lands is dominated by sal trees. Sal plates are made by stitching together 8-9 sal leaves with small tillies (usually teak sticks) to form a circu• lar plate. Plates are sold in bundles of 1,000 (although nor• mally a bundle contains no more than 600-650 plates). Sal plates are what have traditionally been used in marriages and other ceremonies. The processing of sal plates has recently been improved, resulting in a more elaborate plate called a sal thai. Thais are made by first placing a sheet of polyethyl• ene between two 5a/ plates. Then a mechanized press and heat are used to "weld" the three into a thai with a rim. The thai prevents leaks from between the leaves because of the polyethylene sheet as well as from the edge of the plate be• cause of the rim. Sal thals are sold in bundles of 100. Both sal plates and thals are available in 12-, 10-, 7-, and 5-1'nch diameters, as well as in the form of small katories (bowls).

Seasonality plays an important role in changing the demand for and supply of sal plates and thals, and prices fluc• tuate accordingly. Demand is highest during the marriage seasons (second half of January, first part of March, late April, early August to 15 November, and first half of December) and religious occasions, particularly during Durga (Kali) Pooja in October, and prices increase during these times. Prices decrease during the off-peak agricultural period, as supplies increase because villagers have more time available to col• lect leaves and make plates. The areas under discussion are mostly ones where one annual agricultural crop (paddy) is grown and where planting is carried out during the monsoons (June and July) and harvesting takes place in November and December.

PRODUCTION AND MARKETING CHANNELS

Collection of sal leaves involves no direct cost. Women of the household do most of the collecting and stitching of the 3i Society and Non-timber Forest Products in Tropical Asia leaves, which is done six to twelve months of the year, de• pending on labor availability. The leaves collected must be medium sized, undamaged, and mature, but not old. Since instruments are not used for picking leaves, only leaves within easy reach are collected from trees. Sal leaves are dried in the sun for one day and then stitched together with teak sticks to form round plates. The stitching is also done pri• marily by women. The plates are bundled and taken to village depots (also organized by individual village members) to be sold for ap• proximately Rs 8 a bundle. These depots serve as storage and stocking areas for plate bundles until contractors from Calcutta send trucks (usually twice a week) to buy the mate• rial. Depot owners sell the plates to the contractors for Rs 10-12 a bundle (these prices vary, depending on the prevail• ing supply and demand). The contractor pays a royalty of Rs 416 per truckload to the Forest Department. (Rs 200 is for the actual royalty; Rs 16, or 8 percent, is the sales tax,- and Rs 200 per 40 quintals, or one truckload, is the export fee.) In Calcutta, the contractors sell the plates to wholesalers at Rs 14-16 a bundle. The communities involved in sal plate-making live near sal forests and are not affluent in terms of landownership or other wealth. They have traditionally been involved in sal plate-making and are not proficient in any other traditional skills. The sal plate industry has a number of attributes that deserve special mention. This extremely labor-intensive in• dustry is very cost effective as the opportunity cost of labor involved is usually very low. The industry provides remu• nerative employment opportunities for women to work in their homes. The raw materials used [sal leaves and teak sticks for stitching) grow wild and have little opportunity cost to the villagers or environment. The sal plates, although not the thals, are entirely biodegradable. Also, the solar en• ergy used to dry leaves is free and renewable. Production can be made even more cost effective if demand for the product can be increased. Sal Leaf Plate Processing and Marketing in West Bengal 33

ECONOMICS OF SAL PLATE AND THAL MAKING

Sal Plate/Thai Making by Hand at the Village Level An average of ten to twelve person-days, which include the time spent plucking leaves and stitching plates, are required to make 1,000 plates. At Rs 8 per 1,000 paid by the depot middlemen, a sal platemaker would earn Rs 8 per 10 days or Rs 24 per person per month. Families could earn Rs 72 per month if three persons per family made sal plates. This en• amount is profit since there is no direct cost for collect• ing leaves.

Sal Plate/Thai Making with Machines The economics of mechanized sal thai making are quite dif• ferent compared to the economics of handmade sal thals (Table 3.2). Assuming that one person is employed per ma• chine and that one machine is owned by the manufacturer, the net return per 1,000 thals (based on variable costs only) = Rs 150 - Rs 77 = Rs 73, or approximately Rs 75. If we estimate that one person makes about 200 thals per day, it would take five days to make 1,000 thals. Thus, a net profit of Rs 75 is made every five days. Therefore, to break even or recover the investment cost of the machine (Rs 3,900), it would take approximately 260 days. After that, a steady net profit of as much as Rs 75 per 1,000 thals could be made.

Table 3.2 Cost breakdown of mechanized sal thai making

Cost Rs

Fixed Machine 3,900' Variable Labor @ Rs 20/1,000 20 Polyethylene @ Rs 24/kg {800 thals) 30 Electricity @ Rs 5/1,000 "5 Plates @ Rs 11/1,000 plates 22

Total 77 Price @Rs 15/1,000 thals 150

a. Data provided by AMI Engineering, manufacturers of tfai/-making machines, Station Road, Patna 800001. 34 Society and Non-timber Forest Products in Tropical Asia

The blade and the electric heating coil will have to be re• placed after manufacturing about 50,000 thals.

CONSTRAINTS AND As for market size and pricing, Khare (1990) notes that lim- OPPORTUNITIES jted access to markets, as well as a great dependence on gov• FOR NTFPS ernment and private intermediaries, has a direct bearing on AND SAL THALS NTFP prices. The cost of produce sold by the producer, whether to consumers or to intermediaries, seems to have no relation to the cost of labor, inputs, arid transportation. The following factors combine to depress prices of di• rect sales. First, very often the supply exceeds demand in localized markets. Second, traders control the market and dictate prices not only during the season when plates are in high demand but also often in the off season because of tradi• tional credit links with consumers. The third factor, which affects prices, is the selling of produce during the high-de• mand season and the purchasing of the same product during the lean season. Large-scale deforestation is another constraint or threat to the welfare of forest-based communities, particularly those dependent on NTFPs for their livelihood. New monoculture efforts yield very little, if anything, of NTFPs, long an important source of livelihood for rural people, par• ticularly to women, in these forest-based communities. About 75 percent of the species now being planted are for use by large industries. The remaining 25 percent has a rather insig• nificant relationship with the NTFPs currently collected. Trading of NTFPs was nationalized by the government to keep village industries adequately and continuously sup• plied with NTFPs and to ensure that primary collectors re• ceived remunerative prices. Some people contend that this nationalization has been done incrementally and is confined somewhat to the larger scale operations with more remu• nerative prices, resulting in maximum revenues for the state governments. People dependent on forests for their livelihood often Sal Leaf Plate Processing and Marketing in West Bengal 35 find the production of NTFPs more attractive than the sale of poles. This is evident in West Bengal, where a person can earn up to an average of Rs 8 per day from about 1 hectare of mixed sal forest after six years of regeneration. The average daily income from sal poles, after local communities are given the 25 percent share they are entitled to, would be about Rs 1.4 per hectare. The poor return on sal poles, along with the fact that NTFPs are generally continuously or at least sea• sonally available, makes NTFPs more attractive. The state government could convert this situation into an opportunity to further develop and generate production and marketing channels for these products (Poffenberger 1990]. The Forest Department should also seriously consider the actions that must be taken to compensate low-income families (the most disadvantaged groups and most dependent on the forests around them) who suffer losses because of steps taken dur• ing forest conservation and protection. NTFP harvesting and processing would create employment opportunities for the local communities, thereby ensuring that the forests are pro• tected during maturation of the new production system.

Sal forests have a multiproduct advantage compared to monoculture systems. This is a decided benefit for low-in• come families and women, as the opportunity cost of their time is low and they can therefore spend more time harvest• ing and processing NTFPs and capture more benefits. Also, the investment costs to the Forest Department of establish• ing sal forests are relatively low, and it can use this as an opportunity for encouraging further development in these communities. The Forest Department will need to "internalize" the experience it has had in joint management systems over the past few years. A key element of this process would be the training of middle- and field-level officers in the department to better appreciate and understand NTFP systems so that they may better integrate and extend these systems in forest protection and management over the long term. 36 Society and Non-timber Forest Products in Tropical Asia

REFERENCES Government of West Bengal. 1988. Poffenberger, Mark. 1990. Joint Project report on resuscitation of management of forest lands: sal forests of Southwest Bengal Experiences from South Asia. A through peoples' participation. Ford Foundation program Department of Forests. statement. New Delhi: The Ford Foundation. Khare, Arvind. r990. Small-scale forest enterprises in India with special reference to role of women. Wasteland News. November 1989-Ianuary 1990. CHAPTER 4

Jefferson Fox

Non-timber Forest Products in a Nepali Village in 1980 and 1990

Non-timber forest products (NTFPs) generally include foods, medicinals, resins, latexes, fodder and fiber grasses, wildlife, fuelwood, rattan, and other items. In the sal [Shoiea robusta) forests of northern India and Nepal, however, fodder and fuelwood make up the largest portion of the NTFPs collected. Malhotra et al. (1991), for example, found that in a typical village in West Bengal fodder and fuelwood account for ap• proximately 88 percent of the NTFP harvested an• nually. In 1980-811 studied the forest-use practices of a village in the hills of central Nepal (Fox 1983). Results indicated that the forests of this village were in poor condition and that the major causes of this degradation were the grazing of and fodder collection required for the large livestock popula• tion. In 19901 returned to this village to resurvey forest con• ditions and forest-use practices.. In 1990 the forests of this village were in much better condition than they had been in 1980. This paper documents differences in forest conditions and changes in the production and collection of NTFPs, and seeks to determine what factors triggered these changes.

BHOGTENI IN (980 Bhogteni is situated on the eastern ridge of the Daraundi AND 1990 watershed at about 1,200 meters. This elevation is a sub• tropical haven between the hot sultry climate of the river valley and the bitter cold winters farther up slope. It takes about an hour to walk to the village, which is north of Gorkha, the administrative center of the district. 38 Society and Non-timber Forest Products in Tropical Asia

POPULATION AND LAND RESOURCES

The 1981 population of the village was 653 living in 107 households with an average household size of 6.1 persons. By 1990 the population had grown to 835 living in 144 house• holds of 5.8 persons. The per year growth rate of the village is 2.46 percent, compared with a national rate of 2.62 per• cent (Chhetri, pers. com.). The number of households grew at a rate of 2.97 percent per year; this is faster than the popu• lation growth rate and suggests that family groups were break• ing off to form new households. In t98o, 62 percent of the population were Brahmin- Chhetri, 18 percent were Newari, 16 percent belonged to ser• vice castes (Kami, Damai, and Sarkai), and4 percent belonged to either Jogi or Magar caste groups. In 1990 these figures were 63 percent, 17 percent, 10 percent, and 10 percent, re• spectively. Villagers in Bhogteni own approximately 113 hectares (ha) of agriculture land and have access to 73 ha of Schima- Castanopsis scrubland and 39 ha of sal [Shorea robusta) for• est. In relation to land resources, the population density in 1981 was 290/km2, or in terms of cultivated lands and forest/ scrub/grazing lands, 577 and 583/km2. In 1990 these figures were 371, 738, and 745 persons/km2, respectively. The aver•

age farm size in 1980 was estimated to be 1.1 ha; in 1990 the average farm had shrunk to 0.8 ha.

LIVESTOCK

In 1980 the total livestock population of the village included 167 buffalo, 275 cattle, 303 goats, and 10 pigs. In 1990 these figures were 195, 242, 337, and 27, respectively. The total livestock population was approximately the same (390 vs. 398 livestock units); consequently, the amount of manure available for fertilizing village lands remained constant. In 1990 the number of grazing livestock (cattle and goats) was less than in 1980 (203 vs. 223) because farmers kept fewer cattle. In addition, the number of livestock kept per house- Non-timber Forest Products in a Nepali Village 39 hold in 1990 (3.13 animals] was significantly less than the number kept in 1980 (4.31 animals). More specifically, an average household kept significantly fewer female buffalo, cows, and oxen in 1990 than in 1980. However, because farms were smaller, the amount of manure available per farm re• mained approximately the same. In 1980 livestock ownership patterns were influenced by caste and farm size. Brahmin farmers owned significantly more cattle than non-Brahmin households, and Brahmins who owned large farms had more cattle than Brahmins with smaller farms. Among non-Brahmin households, farm size had a greater impact than caste on the number of livestock kept. Farmers who owned large farms kept more buffalo and cattle than farmers who owned small farms (Fox 1987, 172). By 1990 Brahmin farmers no longer kept more cows than their non-Brahmin neighbors. This trend was particu• larly strong among Brahmins with large farms who owned an average of 5.4 cows per household in 1980 and only 0.75 cow in 1990. In 1980 it was hypothesized that Brahmins, the highest caste in the Hindu hierarchy, kept cows for religious and not for economic reasons (Fox 1987). For whatever rea• son, Brahmins who kept more cows in 1980 were no longer doing so in 1990. Among all caste and farm size groups, farm• ers kept fewer cattle (cows, calves, and oxen) in 1990 than in r98o. In addition, farmers who owned large- and medium- size farms, as well as farmers from the Brahmin and Chhetri castes, kept fewer female buffalo in 1990 than in 1980.

TRANSPORTATION

Since 1980 a road has linked Gorkha with Kathmandu and Pokhara. Allan {1986, 191) suggests that the key factor in an entire mountain landscape is the construction of a new road. A road provides easy access to markets, opens opportunities for seasonal migration for cash labor, and facilitates diffu• sion of new food plants. Furthermore, construction of new roads in mountain landscapes integrates mountain regions into the lowland culture. 40 Society and Non-timber Forest Products in Tropical Asia

FOREST TENURE

Natural forests in the village are in the Schima-Castanopsis and sal categories (Stainton 1972). Schima-Castanopsis for• ests predominate at heights exceeding 1,000 meters. Most of these forests, however, were converted some time ago to ag• ricultural lands. Below 1,000 meters, sal forests predominate. Stainton maintains that sal is better able to withstand harsh treatment from lopping and than most other species. Because the objective of this study is to understand for• est-use practices (including the collection of NTFPs), forest lands must further be classified according to ownership and/ or control. Villagers perceive forest ownership patterns as (1) government owned [sarkari], (2] privately owned [vyatigat], and (3) community owned [sarvajanik). These categories are associated with the traditional rights of people to use these lands. In 1980 villagers recognized sal forests as government property, which somewhat limited their use of these forests. Schima-Castanopsis remnants, however, were considered community property and management was open access. But a few Schima-Castanopsis remnants have been regarded his• torically as private property.

The National Forestry Plan of 1976 provided a policy base for initiating community participation in managing for• est lands (NAFP1979). Under these rules, a village panchayat (local unit of government) could apply for up to 125 ha of severely degraded forest land, designated as panchayat for• est, and up to 500 ha of good forest, designated as panchayat- protected forest. These landmark regulations formally rec• ognized the rights of villagers to manage their own forest resources with technical assistance from the Department of Forestry. Gilmour and Fisher (1991, 16), however, estimate that less than 600 ha of forest land has actually been trans• ferred to the panchayats. By r990 villagers in Bhogteni were aware of the panchayat forest management schemes. Although villagers have never initiated any effort to assume official responsibil• ity for managing government forest lands, they established Non-timber Forest Products in a Nepali Village 41

several committees to protect local forests. These commit• tees allowed people to collect firewood, to gather leaves for roofing, and to graze livestock. Cutting of large trees was prohibited. These committees operated informally, and their activities were limited to forest protection. Committees were formed to manage portions of the sal forest, as well as most of the Schima-Castanopsis remnants.

FOREST RESOURCES

In 1980 private woodlots had 179 trees/ha, 92 percent of which were Schima wallichii and 8 percent miscellaneous species (primarily Castanopsis indica). In 1990 private woodlots had 489 trees/ha. Whereas in 1980 the basal area of an average tree was 562 cm2, by 1990 this figure had dropped to 200 cm2. Likewise, in 1980, communal forests had less than 26 trees/ ha, primarily Schima wallichii, with an average basal area of 474 cm1 per tree. By 1990 these forests had an average of more than 450 trees/ha, and the size of the average tree had decreased to 139 cm2. In 1980 government-owned sal forests had 746 trees/ha, 95 percent of which were sal. By 1990 this figure had increased over fourfold to 3,345 trees/ha. The size

2 of the average sal tree in 1980 was 232 cm per tree; by 1990 this figure was 63 cm2. The size of an average tree in 1990 decreased because the number of small trees increased tre• mendously. The total basal area (a product of the number of trees per hectare and mean basal area m2/ha), however, was higher in 1990 than in 1980 because of the great increase in the number of trees per hectare.

Wood volume was estimated using tables compiled by the Department of Forestry (1973). In 1980 timber volumes in Bhogteni ranged from 9 m3/ha on communal lands to 106 m3/ha in govern men t-owned lands,- in 1990 these figures were 46 and 188 m3/ha, respectively. Average mean annual incre• ments (tree growth) on communal and government-owned lands also increased tremendously in 1990 when compared with 1980-81 statistics. 42 Society and Non-timber Forest Products in Tropical Asia

DEMAND FOR NON-TIMBER FOREST PRODUCTS

The 1980 study suggested that villagers used their local for• ests for grazing livestock and for collecting fodder and fire• wood. Because the total livestock population in 1990 was approximately the same as the 1980 population and the num• ber of grazing livestock (cattle and goats) had actually de• creased, the demand for leaf fodder from the forest and for grazing remained constant or decreased during this period. Data were not collected in 1990 to make it possible to esti• mate the portion of fodder collected from the forest as op• posed to private land. Furthermore, the 1980 study indicated that tree fodder was collected from public lands on which livestock were grazed primarily at the end of the dry season. May and June (the 1990 research was conducted in December). This is a livestock feed-shortage period, a time that public lands serve as a critical source of tree fodder and a place for grazing live• stock until grass fodder becomes available. Although in 1990 it was obvious that farmers had greatly increased (three- to fourfold) the number of fodder trees grown on private land, it would be useful to quantify the value of public forests as sources of tree fodder and places to graze livestock during the dry season. In 1980 firewood consumption rates were measured using three methods, and the relative influence of farm size, family size, caste, and season of the year on firewood con• sumption rates was determined (Fox 1984). In 1990, rates were remeasured using two of the original three methods. For the weight study, firewood was weighed at each participating household. Firewood was weighed on four sepa• rate occasions in 1980 and on a single visit in December 1990. In the recall survey, participants were asked two questions: (1) how much wood do you burn daily in the cold season? (2) how much wood do you burn daily in the warm season? In 1980 and 1990, 76 and 130 households, respectively, partici• pated in this survey Table 4.1 shows firewood consumption rates according to year and methodology. No significant dif- Non-timber Forest Products in a Nepali Village 43 ferences were observed between average firewood consump• tion rates in 1980 and 1990. When the survey results are broken down by caste and farm size, no significant differences were observed in 1980 or in 1990. When broken down by family size, however, sig• nificant differences occurred within each group in both time periods (i.e., large families burn less wood per person than small families). These results indicate that firewood demand patterns have remained constant throughout the ten-year period.1 As for supply, the annual increment of wood available for fuel is the sum of wood growth and deadwood generation. For example, using 3 percent of total wood as an estimate of annual deadwood generation, the estimated annual firewood supply from all sources (private, common, and government forests) in 1980 was approximately 370,000 kg/year. By 1990

Table 4.1 Firewood consumption (m3/capita/year)

Methodology Weight Recall survey 1980 0.95 1.86 n - 100 n-76

1990 0.95 1.79 n - 13 n - 130

Caste Brahmin/Chhetri Newar Service 1980 (Recall survey) 3.2 3.1 2.9 n-76 (a) (a). (a) 1990 (Recall survey) 2.31 2.14 1.63 1.61 n - 130 (a) (a) (a) (a)

Family size (people) 9-20 5-8 1-4 1980 1.98 2.89 3.65 n-76 (a) (b) (c) 1990 1.26 1.75 2.79 n- 130 (a) (a) (b)

Farm size Large Medium Small (>1.25 ha) (0.75-1.25 ha) (<0.75 ha)

1980 2.64 3.14 3.07 n-76 (a) (a) (a) 1990 1.43 2.36 1.84 n- 130 (a) (a) (a)

Note: A mean followed by (a) is significantly different (p < 0.0s) from one marked (b) or (c) but is not significantly different from a mean marked (a). 44 Society and Non-timber Forest Products in Tropical Asia

firewood supply had grown to approximately 800,000 kg/year. Firewood demand, however, has only grown from 350,000 to 478,000 kg/year,- thus, firewood supplies are sufficient to meet demand.

CHRONOLOGY OF CHANGE

Between 1980 and 1990, a number of events occurred that affected non-timber forest resources in the village. The fol• lowing is a brief chronology of these changes. In 1976 designation of the panchayat-protected forest and panchayat-forest regulations were passed. Although these events occurred before the 1980-81 study, began, it was dur• ing the 1990s that the community began to act based on these regulations. In 1981 and 1982 the Resource Conservation and Utili• zation Project (RCUP), funded by USAID, conducted enrich• ment plantings on several areas of forest land. RCUP hired villagers for a four-year period to protect these plantations. During this time, grazing and collecting of firewood and fod• der were prohibited, but villagers were allowed to collect dry firewood. Once the forests were protected from grazing, na• tive species quickly regenerated and shaded out the enrich• ment plantings. The closure of the village's forest land to grazing livestock forced villagers to stall-feed their animals and may have led to the change in total livestock composi• tion (more buffalo and fewer cattle). Farmers also increased the number of fodder trees they grew on their private lands. In 1982 the road from Gorkha to Kathmandu and Pokhara was completed. This allowed villagers to import chemical fertilizers and thus maintain or increase per capita agricultural production without a simultaneous increase in the alternative source of nutrients—manure. In 1985 the RCUP guards were removed and the pro• tected forest lands turned over to the Department of Forestry. The department lacked the resources to manage these small pieces of forest and turned them over to the panchayat for management. This was an informal arrangement and nei- Non-timber Forest Products in a Nepali Village 45 ther the Department of Forestry nor the panchayat made any effort to complete the formal paperwork necessary to offi• cially record this land as panchayat or panchayat-protected forest. Patches of forest near poorer households were quickly degraded again. Patches near more wealthier households were protected by the local community. By 1990 villagers had established forest protection com• mittees to protect seven individual areas of forest land. These formal committees consisted of a chairperson, secretary, and designated household members, who reside around the spe• cific area of forest. Rules governing the management of these patches of forest vary by committees. Some committees al• low grazing and the cutting of grass and collection of dead- wood for fuel. Other committees forbid grazing. The com• mittees have the right to fine offenders but seldom do so. In 1990 the committee for the sal forest hired local farm• ers to thin the regenerated sal forest. Two factors lay behind the committee's decision to thin the forest—a management activity. First, a small population of leopards had returned to the regenerated sal forest and were perceived as a threat to children walking to school. The protection committee felt it essential to cut the vegetation near the trail for the safety of the children. Second, the panchayat system of government was overthrown by a people's movement. With its removal, the protection committee no longer had to share the pro• ceeds of the management activity (i.e., the ) with the panchayat leader. In December 1990 the sal forest committee was em• broiled in a controversy over how to divide the products of the forest- activity. Villagers were involved in heated arguments over who had rights to this wood (i.e., defining who is in and who is not in the forest protection committee). In general, any member of the ward in which the forest is located has a right to forest products. Although the commit• tees are heterogeneous and reflect the make-up of the user- group population, people from the service castes are inad• equately represented on the committees. These committees are still developing. With proper leadership and guidance, they 46 Society and Non-timber Forest Products in Tropical Asia

could become influential, stabilizing forces in the commu• nity.

TOWARD Bhogteni's continued rapid population growth rate and its A THEORY already high population density may have contributed to the OF CHANGE willingness of villagers to seek more efficient methods of managing forest lands (Boserup 1965; Brookfield 1984). But clearly other variables were at work as well. Perhaps the most significant factor affecting change in Bhogteni was the introduction of a new tenure regime for forest lands (through the recognition of panchayat forest and panchayat-protected forest in 1976). This change, which emanated from the efforts of the World Bank and the inter• national donor agencies that perceived a high risk of envi• ronmental degradation, led to a new method—community groups assuming responsibility for managing forest lands. Once villagers realized that protection committees are sanc• tioned by government officials, they quickly developed their own methods for managing local resources. The new tenure regime provided incentives for local people to divert resources to the management of forest lands. The fall of the panchayat system and the establishment of a more democratic govern• ment have also encouraged farmers to invest labor into man• aging forest lands through community programs. These changes have clearly Led to a more efficient management system that increases the productivity or quality of a unit of labor input. Another major change in Bhogteni was the introduc• tion of the road. This allowed villagers to import chemical fertilizers and thus maintain or increase per capita agricul• tural production without a simultaneous increase in live• stock. Although the changes induced by the road have in• creased the intensity with which agricultural lands are used, these changes have also reduced the workload of farmers (fewer livestock per capita means less labor). Farmers have invested this surplus labor in higher value livestock (buffalo), which they stall-feed, resulting in a more efficient forest

! Non-timber Forest Products in a Nepali Village 47

management system. In addition, the road has encouraged village youths to migrate to Kathmandu and to the tarai (southern portion of Nepal] for education and seasonal job opportunities. A third major change in Bhogteni has been the role of outsiders (the Resource Conservation and Utilization Project and the non-government organizations) in helping villagers adjust to environmental change. The effect of these efforts is less clear. When the Resource Conservation and Utilization Project closed forests to grazing, villagers were forced to stall- feed their livestock and to grow more fodder on their private lands. This effort, however, would not have been successful without the change in forest tenure policies (to encourage farmers to assume responsibility for forest management) and the construction of the new road. Together with these other changes, however, outsiders provided another surplus (e.g., job opportunities in the nursery, free seedlings) that villagers invested in adopting a more sustainable forest management system.

Another agent of change that may play a larger role in Bhogteni in the future are small non-government organiza• tions. Within the last several years, a women's development project and Save-the-Children have started working in Bhogteni and are supporting efforts. Ironi• cally, however, both organizations are also encouraging live• stock-raising projects. In Bhogteni the human dimensions of land-use change are more complex than might be suggested by theory. Dur• ing this ten-year period, the village has undergone both in• tensification (increased production on agricultural lands re• sulting from application of fertilizers) and innovation (im• proved management of forest lands resulting from new for• est policies). The new road was a factor in both the intensifi• cation and innovation processes, and it is not clear that ei• ther change could have occurred without it. In addition, per• ceived environmental risk was an important factor in the design and adoption of new forest policies. This was true both in terms of external actors such as the World Bank and other 48 Society and Non-timber Forest Products in Tropical Asia

donor agencies who pushed for the adoption of the new laws, and internal actors, the farmers who invested in organizing community management of forest lands. The role of popula• tion growth in this process is less obvious. While clearly neo- Malthusians were wrong, population growth did not lead to further land degradation; there is no evidence to confirm Boserup's hypothesis (1965) that population growth would lead to internal innovations for managing forest lands more efficiently. Forests, however, would not have been perceived as being at risk by either external or internal actors if it were not for the high population density and growth rate. It is important to remember that this is a village study and that these results cannot be generalized to the rest of Nepal. But when these results are considered along with Gilmour's work (1991) in the Jhiku Khola watershed, the evidence begins to indicate that despite continued popula• tion growth, more sustainable systems for the production and consumption of NTFPs are being established through• out the hills. The Malthusian theory of population growth and resource degradation is clearly a myth that needs to be slain before the human dimensions of environmental degra• dation in the Himalayas can be understood. Although the changes discussed are encouraging, the question remains, how sustainable are they? Rather than answer this question, I would like to pose three more ques• tions. The answers to these questions may determine the future of NTFPs in Bhogteni.

1. How stable are the forest management committees? To the extent these committees represent an outsider's view of how villagers should manage their forest re• sources, these committees will probably fail. Villagers will need to decide (implicitly or explicitly} the nature of the social arrangement, the nature of the NTFP har• vesting practices and controls to be applied, and who will be included as members of the user community. 2. What will be the direction of the government's forest management program? To the extent the government Non-timber Forest Products in a Nepali Village 49

supports user committees, these committees may be able to exert a positive influence on forest-use prac• tices. These committees are badly in need of informa• tion on productive methods of managing regenerated forests for non-timber products. The government could play a positive role in conducting research on this ques• tion and providing forest extension services to user com• mittees. 3. How stable is the "new" agroecosystem with its in• creased use of chemical fertilizers? Is the use of chemi• cal fertilizers sustainable in the Nepali context? If the answer to this question is no, farmers may soon be rais• ing more livestock than the forest can sustain.

In terms of the broader question of NTFPs and change, these results are simultaneously encouraging as well as dis• turbing. It is encouraging that population growth does not necessarily lead to a downward spiral of land degradation and starvation. Brookfield's theory (1984) that innovations that support more sustainable management systems can be trig• gered by a number of factors including population growth and perceived environmental risk appears to be correct. In this case changes in forest tenure, introduction of a new road, and forest management inputs from outside sources have led to a new forest management paradigm and to the sustainable production and collection of NTFPs. The new community forest management groups provide a more efficient manage• ment system in which the productivity or quality of a unit of labor has been increased. The disturbing aspect of these results is the suggestion that forest degradation must be fairly severe before perceived environmental risk causes forest management policies to be changed. Once forests degrade and few NTFPs of commer• cial value remain, it is less difficult to improve forest condi• tion (even in the face of population growth) by giving local people incentives for managing local forests. It is much more difficult to protect valuable NTFPs in sparsely populated undegraded environments like Kalimantan of Indonesia or $0 Society and Non-timber Forest Products in Tropical.Asia

the Amazon. We should recognize, however, that the imme• diate threat to these lands and products is not local people or population growth but bad forest management policies. For• est policies that provide incentives [rather then disincentives) for local people to manage forest resources must be imple• mented.

Acknowledgments—I would like to acknowledge the many people who assisted me in conducting this research and writ• ing this paper. In the field the assistance of Buddhi Rijal, Davesh Mani Tripathi, Ganesh Bahadur Karki, Jagannath Joshi, Krishna Prashad Acharya, Narayan Prashad Sharma, Shir Prashad Sharma, and Surya Kant Siqdel was invaluable. Drs. Ram Chhetri and Neeranjan Rajbhandari assisted with preparing the questionnaire, and data coding and analysis. Comments from Dr. Deanna Donovan and an anonymous reviewer were also helpful.

Portions of this article originally Himalayas. My study of firewood appeared in a different form in consumption (Fox 1984) is not Mountain Research and Develop• cited. Clearly, however, if ment, Vol. 13, No. 1, pp. 89-98. firewood consumption rates can © 1993 by The International be replicated at a ten-year interval Mountain Society and United and show no significant differ• Nations University; permission ences, we can conclude that our granted. method of documenting firewood consumption and total consump• 1. Thompson and Warburton tion rates is valid. The Himalayas (1985} cite Donovan's [1981} may face great uncertainty, but review of firewood consumption the use of firewood consumption rates (varying by a factor of 67) as rates is not a good example. proof of the magnitude of the uncertainties faced in the

Allan, N. J. R. 1986. Accessibility under population pressure. and altitudinal zonation models of London: Allen and Unwin. mountains. Mountain Research Brookfield, H. C. 1984. Intensifica• and Development 6(3); r83-94. tion revisited. Pacific Viewpoint Boserup, E. 1965. The conditions 25(1): 15-44. of agricultural growth: The Department of Forestry. rg73. economics of agrarian change Forest resources survey: Forest Non-timber Forest Products in a Nepali Village 51

statistics for the hill region. Gilmour, D. H., and R. J. Fisher. Kathmandu, Nepal. 1991. Villagers, forests, and foresters: The philosophy, process, Donovan, D. G. io8r. Firewood: and practice of community How much do we need? Institute forestry in Nepal. Kathmandu: of Current World Affairs, Hanover, Sahaygoi Press. New Hampshire. Malhotra, K. C, D. Deb, and M. Fox, J. M. 1983. Managing public Dutta. 1991. Role of non-timber lands in a subsistence economy: forest products in a village The perspective from a Nepali economy: A household survey in village. Ph.D. diss., University of Jamboni Range, Midnapore Wisconsin, Madison. District, West Bengal. Working . 1984. Firewood consump• paper, Indian Institute of Bio- tion in a Nepali village. Environ• social Research and Development, mental Management 8(3): 243-50. Calcutta.

. 1987. Livestock ownership NAFP. 1979. Nepal's national patterns in a Nepali village. forestry plan 1976 (2033). Mountain Research and Develop• Unofficial English translation, ment 7(2): 169-72. Nepal-Australia Forestry Project, Kathmandu, Nepal. Gilmour, D. H. 1991. Trends in forest resources and management Stainton, J. D. A. 1972. Forests of in the middle mountains of Nepal. Nepal New York: Haftner. In Soil fertility and erosion issues Thompson, M., and M. War- in the middle mountains of burton. 1985. Uncertainty on a Nepal, ed. P. Shah, H. Schreier, S. Himalayan scale. Mountain Brown, and K. Riley. Vancouver: Research and Development 5(2}: Department of Soil Science, "5-35- University of British Columbia.

Local Institutions and Government Policies: Methods of Managing (or Mismanaging) Non-timber Forest Products

CHAPTER 5

Michael R. Dove

The Impact of Cultivation on Peasant- State Relations in Forest Product Development: The Transition from Native Forest Latexes to the Introduced Hevea in Kalimantan

BACKGROUND With growing interest in developing non-timber forest prod• ucts, an outstanding historic example of such development— carried out without the support of government, and to some extent despite its opposition—merits attention. This example involves the transition among the inhabitants of Indonesia (and also Malaysia) early this century from gathering native forest,rubbers to tapping Para rubber [Hevea brasiliensis).1 A key distinction between tapping Para rubber and gath• ering forest rubbers is that the former falls within the realm of "agriculture" whereas the latter does not. The develop• ment of one from the other involved a transition from with• out agriculture to within agriculture, from nature to culture.2 The magnitude of this transition reflects the involvement of not just society and nature, but society, nature, and the state. The transition to Para rubber cultivation was a key point in the development of this peasantry's orientation toward the state. A historical analysis of this transition will contribute to more informed planning of the development of non-tim• ber forest products and to improved understanding of con• temporary peasant-state relations.

OUTLINE Following a brief description of the research locale and fo• OF PAPER cus, this analysis begins with the history of the Southeast Asian trade in forest products, the development of the rub- 56 Society and Non-timber Forest Products in Tropical Asia

ber trade in particular, and then the introduction of Para rub• ber. This is followed by a brief analysis of the similarities and differences between gathering forest rubbers and tapping Para rubber, especially with respect to swidden agriculture. Then the political-economic context of these developments is analyzed, focusing on state involvement in the forest rub• ber trade and the impact that this had on the adoption of Para rubber. The final section of the paper compares the po• litical-economy of Para rubber development in Southeast Asia with its development in South America.

RESEARCH LOCALE The data upon which this analysis is based were gathered during several periods of research in West and South Kalimantan, carried out between the years 1974 and 1984. The most detailed data were gathered during an extended stay with the Kantu', an Ibanic-speaking tribe of West Kalimantan (Figure 5.1 ).3 The Kantu' grow dry rice (as well as some swamp rice), maize, cassava, and a wide variety of nonrice cultigens in swiddens cut from both primary and . In addition to cultivating annual food crops in swiddens, the Kantu' cultivate several types of perennial cash crops in the fallowed swidden land. These include a variety of trees yielding edible fruits and oils, the pepper plant [Piper nigrum), and, especially, the Para rubber tree. For much of this century, Para rubber has been the Kantu's primary source of cash or tradable commodity used to obtain the ba• sic trade goods of salt, tobacco, clothing, and kerosene.

Kantu' are far from unique in their dependence on rub• ber, as Indonesia is the world's second largest rubber pro• ducer. Rubber is one of Indonesia's most important resources. It is a major source of household income for more than one million households, and it is the largest agricultural genera• tor of foreign exchange (CPIS 1993, 3; Government of Indo• nesia 1992, 340). Some of this rubber is produced on large plantations or estates that employ modem agricultural tech• nology, heavy capital investment, and a wage labor force. But The Impact of Cultivation on Peasant-State Relations 57

the bulk of Indonesia's rubber, three-quarters at the most recent count (Govern• ment of Indonesia 1992, 231-32), is produced in tiny gardens of a hectare or so, with century-old technology, by so-called "smallholders"4 like the Kantu'—ordi• nary farmers who produce rubber with household labor to meet part, and typically not the major part, of their household's income requirements. Perhaps most re- $8 Society and Non-timber Forest Products in Tropical Asia

markable of all, this three-quarters market share is the prod• uct of more than three-quarters of a century of direct compe• tition between smallholders and estates. Estates held a com• manding share of Indonesia's rubber production during the industry's early years in the second decade of this century, and they have steadily lost ground to the smallholders ever since (Booth 1988, 201). The historical success of rubber smallholders is due partly to the fact that Para rubber filled the same niche in the peasant or tribal economy previously filled by native forest latexes.

HISTORY Trade in non-timber forest products has a long history in the region and has retained considerable economic importance to this day (Dunn 1975; Peluso 1983). Evidence of the export of forest products from the western islands of Indonesia to China dates from the fifth century (Wolters 1967); Middle Eastern trade with the Malay Peninsula dates from A.D. 850; and European trade with Indonesia's islands dates from the fifteenth century (de Beer and McDermott 1989, 16). The antiquity and ubiquity of this trade is suggested by the im• portance of the goods that it supplied—in return for forest products—to people who otherwise lacked them, principally salt and iron (Lian 1988, 118).

FOREST RUBBERS

A major category of forest products throughout this history has been natural gums, resins (intra-regional trade in which may date back to Neolithic times [Dunn 1975, 120-37]), and rubbers.5 Rubber-gathering was formerly a central part of the tribal economy in Kalimantan. The contemporary Kantu' say that their ancestors first explored and settled their present territory in the Empanang river valley, not in search of fresh swidden territory, as our image of swidden cultivators would suggest, but in search of jangkang [Palaquium leiocarpum) and kubal (perhaps Ficus elastica).6 These are two of a wide The Impact of Cultivation on Peasant-State Relations 59

variety of indigenous, latex-producing trees and vines.7 They were (and are) used traditionally, within the local economy, for caulking and sealing purposes (e.g., of canoes (cf. Jessup and Vayda 1988, 16)). It was not desire for this use, however, but for trade that prompted the ancestors of the Kantu' to migrate to the Empanang valley in search of these native latexes. There are few data available on the history of the re• gional trade in forest rubbers. Large-scale trade with Europe dates from the 1840s and the discovery that gutta-percha—a generic name for latex from a number of different plants, but especially homPalaquium spp. (Burkill (1935] 1966, 2:1651)— could be used for insulating marine telegraph cables, among other purposes (Lindblad 1988, 14).8 A second boom in the European trade occurred during the first decade of this cen• tury, following the discovery that fire-resistant plates and tiles could be made from jelutong, again a generic term for latex from a number of different plants, but especially from Dyera spp. (Burkill [1935] 1966, 1:889; Lindblad 1988, 18). This boom was short-lived.9 By World War I interest in rub• bers was shifting from exploitation of native forest sources to cultivation of the introduced Hevea, while among forest products, attention was shifting from rubbers to (Cramb 1988, 109; Lindblad 1988, io2).10

PARA RUBBER

Large-scale international trade in Para rubber dates from de• velopment of the process for clothing in 1839 and development of the pneumatic tire for automobiles in 1888 (Purseglove 1968, 147). The trade was initially supplied by the tapping of naturally grown trees in the Amazon ba• sin." The rubber trees of Kalimantan and the rest of South• east Asia are descended from seedlings that the British gath• ered in Brazil in 1876 and planted in Ceylon in 1876 and Singapore in r877 (Purseglove 1968, 149).12 The first seed• lings from the Singapore trees arrived in Sarawak in 1882 6o Society and Non-timber Forest Products in Tropical Asia

(Tremeer 1964, 52). By 1908, the Sarawak government was distributing rubber seedlings to natives in the interior (Cramb 1988, iii; Tremeer 1964, 52}. On the Dutch side of the bor• der, in West Kalimantan [which today ranks third among In• donesian provinces in terms of rubber acreage (Effendi 1985, 108]), the introduction of the rubber tree occurred at about the same time, in 1909 (Uljee 1925, 74, cited in King 1988, 237). The Kantu', for whom communication was (and re• mains) easier across the border to Sarawak than to the dis• tant Kalimantan coast, say that they obtained their first rub• ber seedlings from Iban tribesmen in the Saribas drainage of Sarawak and from a Catholic mission downriver on the Kapuas, in the late 1920s. A majority of Kantu' households had planted some rubber by World War II, although house• holds with mature rubber were then still in the minority.

NATIVE RUBBERS The historic exploitation of indigenous latex-producing trees VERSUS and vines resembled in many respects the subsequently in- PARA RUBBER troduced system of rubber cultivation.13 The technology used to obtain the latex of most jelutong and some gutta-percha was much like that used to exploit rubber: tapping by means ofv-shaped incisions in the bark (Burkill [1935] 1966,1:892).14 Exploitation of forest rubbers could be carried out at a short notice, in response to market fluctuation, with minimal capi• tal investment or risk-taking, as is also the case with rubber (which is unique in this respect among export crops [Drake 1982, 294-95]):15 The labor requirements of both systems are relatively low (Cramb 1988, 112). These similarities facili• tated the adoption of rubber by the Kantu' and other groups; but this adoption would not have occurred with the speed and magnitude it did if there were not also significant differ• ences. One major difference between Para rubber and the na• tive rubbers involves complementarity with swidden culti• vation. The central act of the swidden cycle, clearing the for• est, clearly differentiates between the two. Whereas the na• tive rubbers are at risk whenever the natural forest is cleared, The Impact of Cultivation on Peasant-State Relations 61 it is cleared forest in which Para rubber is planted: the rub• ber seedlings are planted in newly cleared swiddens. As the habitat of the native rubbers is destroyed, the "habitat" of Para rubber is created. As more and more of Kalimantan's primary forest cover has been cleared, a shift of dependence from the native rubbers to Para rubber became inevitable. This shift was also promoted by the changes in the patterns of agriculture and settlement that have occurred over the past century.16 As a result of demographic and political con• straints, both swidden cycles and settlement patterns have become more sedentary, which favors Para rubber. Whereas the supply of natural rubbers within easy access of any given settlement eventually will become exhausted, the produc• tivity of a rubber grove is potentially open-ended (if natu• rally grown saplings are allowed to succeed the older genera• tions of trees); it is terminated only by migration from the area. While sedentariness is inimical to the continued ex• ploitation of natural forest rubbers, it is essential to the ex• ploitation of Para rubber.

A second important difference between Para rubber and the native rubbers involves recognition of ownership. Indi• vidual ownership of latex-producing forest trees was recog• nized under traditional tribal adat (law). The first person to tap a tree was judged to be its owner.17 These traditional jural principles were not respected by the outside world, however. Even today, valued and individually claimed forest trees are cut down with impunity by outsiders.18 Recognition of pro• prietary rights by the nontribal world is generally reserved for planted trees.19 Indeed, the planting of commercially val• ued perennials like rubber or coffee is recognized under both national and tribal laws as establishing rights not just to the trees but to the land under them (Wcinstockand Vergara 1987, 318-19). The proprietary implications of planting Para rub• ber were not lost on the tribesmen of Kalimantan, who often plant rubber largely for its tenurial benefits. This benefit of Para rubber acquired disproportionate importance because of the particular political-economic conditions prevailing in Indonesia at the beginning of this century. 61 Society and Non-timber Forest Products in Tropical Asia

THE POLITICAL In both colonial and post-colonial Indonesia, whenever a natu• ECONOMY ral resource has experienced a commercial boom and attracted OF RUBBER the attention of government and industry, the latter has taken PRODUCTION IN steps to restrict exploitation by traditional smallholders, al• SOUTHEAST ASIA ways out of ostensible concern for either the welfare of smallholders or the welfare of the resource in question.

GOVERNMENT INVOLVEMENT AND TRIBAL RESPONSE

There was an international boom in jelutong (from Dyera spp.] in the first decade of this century, as mentioned previ• ously. By 1908 the colonial government in parts of Kalimantan was requiring a license to tap the trees; in 1910 the govern• ment awarded all tapping rights to foreign concessionaires;20 and in 1913 the government imposed export levies on the native tappers (Potter 1988, 131-33). The government justi• fied these measures in terms of the need to avoid over- exploitation of the latex-yielding trees (Potter 1988, 131) or the need to protect smallholders against middlemen (Lindblad 1988, 19). But the resultant abuse of native rights in pursuit of European profit was so glaring that the Dutch legal scholar van Vollenhoven used it as a textbook case of the colonial government's abuse of its right to "wastelands" (Potter 1988, 134). Another observer was driven to comparing the right of the Dayak to tap jelutong to the right of the Javanese to cul• tivate land (CAPD 198a, 5519].

This active state effort to control production of the na• tive forest rubbers set the stage for the shift to Para rubber. I suggest that the shift to Para rubber was at least partly a response to the state effort at control. The shift to Para rub• ber was not a shift from a subsistence economy to a mixed or market-oriented economy, as some observers have suggested, because the tribesmen were already participating in the mar• ket through gathering the native rubbers. Rather, it was a shift, or the beginning of a shift, from a tribal political-eco• nomic formation to a peasant formation. Wolf (1966, 25) writes that "Ecotypes based on swidden can support a peas• antry only under exceptional circumstances or where The Impact of Cultivation on Peasant-State Relations 63

swiddens become 'anchored' to a nonswidden crop." For many of the swidden cultivators of Kalimantan, rubber be• came this anchor. But adoption of rubber provided these tribes• men with more than the means to produce a "fund of rent" for the state, Wolf's (1966) defining characteristic of peasant economic formations. Replacement of native rubbers with Para rubber is more accurately seen as part of a move from a limited involvement in the world economy, with limited vulnerability and limited power, to greater involvement with greater vulnerability and greater power. It was a move from limited engagement to greater engagement. It represented not submission to the state, but an aggressive reorientation to• ward the state.

CULTURAL EVIDENCE

The central role of Para rubber production in this political- economic transformation is reflected in ritual. Omen-taking was traditionally practiced both when gathering native rub• bers and producing Para rubber, but with a difference: omen- taking in native rubber production focused on the hazards of traveling to gather the rubber, whereas omen-taking in Para rubber production focused on the hazards of trading the prod• uct (Sandin 1980, 107, 112, 113, 115, 122]. The focus in the first case was on the physical dangers of the tribal world, and the focus in the second case was |and is} on the fiscal dangers of the outside world. This shift from "physical" to "fiscal" hazards aptly sums up some of the consequences of a transi• tion from a tribal to a peasant political-economic formation. The anxieties attendant upon this transition were also reflected in an extraordinary event that took place in pre-war Borneo. A panic swept Kalimantan and Sarawak in the 1930s and 1940s, based on a rumor that the spirit of the rubber was "eating" the spirit of the swidden rice (Dove in press; Free• man 1970, 268; Geddes 1954, 97). This panic obviously re• flects native anxiety about the impact of involvement in Para rubber on the traditional cultivation of swidden rice. But it also reflects a more generalized anxiety about the impact of 64 Society and Non-timber Forest Products in Tropical Asia

the new political-economy on tribal society. It can be inter• preted as a cultural caution against overinvolvement in rub• ber production—a caution that most tribesmen took to heart. Just as most Dayak tribesmen moderated their involvement in gathering the native forest rubbers—during the periodic booms, it was coastal Malays and Bandjars who did most of the gathering, not the interior tribesmen (Potter 1988, 132- 33; cf. Hudson 1967, 66)—so did they moderate their involve• ment in production of Para rubber. As Hudson writes (1967, 311]: Most villagers feel that the rubber market is a chancy thing. World demand varies and prices fluctuate. No one of them wants to be totally dependent on factors over which they have no control. Thus . . . rubber cultivation will continue in the foreseeable future as an activity ancillary to swidden farming.

The fact that they maintained rubber production as an "an• cillary" activity proved to be the particular strength of their system of smallholder production.

GOVERNMENT EVIDENCE

Political-economic tensions over the tribesmen's shift from the native forest rubbers to Para rubber were also manifested in government policy, albeit with different intentions. Where the tribesmen were concerned with minimizing their vul• nerability to the outside world, the state was concerned with maximizing it. The most notorious example of this occurred under the International Rubber Regulation Agreement, en• acted by the Netherlands, Great Britain, France, India, and Siam in 1934 and eventually extended to 1944 (Barlow 1978, 62-67; Boeke 1953, 124-25, 248; Thee 1977, 26). The agree• ment was in theory designed to stabilize rubber prices by limiting production. In practice, in Indonesia and elsewhere, the agreement was used to limit production by smallholders for the benefit of estates, through the imposition of export lev• ies ranging up to 83 percent on the smallholders' rubber (Dillon 1985, 116). This intention was frankly acknowledged by some The Impact of Cultivation on Peasant-State Relations 65

of the participants, as the following statement by the chair• man of the British North Borneo Company indicates:

One of the primary objects of the Rubber Control Scheme was to protect European capital in companies in Malaya, Borneo, and the Netherlands East Indies from com• petition arising from the production of rubber by the natives at a fraction of the cost involved on European-owned estates. (McFadyean 1936, cited in Barlow 1978, 72)

THE POLITICAL The development of Para smallholder rubber production in

ECONOMY Southeast Asia followed a very different path from that taken

OF RUBBER m jts homeland in South America. The fact that Para rubber PRODUCTION IN was regarded largely as a forest product in South America SOUTH AMERICA but an export "crop" in Southeast Asia, in particular, offers a new perspective on its development in both regions. Para rubber grows wild in the tropical forests of South America, at maximum densities of one to two trees per hect• are. The native tappers, or seringueiros, clear winding paths or "avenues" several kilometers in length through the forest to perhaps 100-200 trees, which they tap every day in season (Barlow 1978, 17). The tappers work under the authority of traders or patrons, who have rights to forest sections based on tax-paying and force (Murphy [i960] 1978, 18). Indian tap• pers working on state land and continuing some involvement in subsistence food-cropping are somewhat better off than peasant tappers on privately owned land who depend on their trader even for food; although even for the Indians, involve• ment in tapping has come at the expense of their own social system (Murphy (i960] 1978, 143, 153, 177)- Whereas Para rubber has empowered many of its smallholder adopters in Southeast Asia, its impact has been quite the reverse in South America (Padoch 1988, 131). This difference is reflected in the recent international attention focused on the plight of the South American tappers (and the killing of a well-known activist) and efforts to "empower" them. The lot of the South• east Asian smallholder, while lacking in many respects, is still so different from that of the seringueiros that a similar 66 Society and Non-timber Forest Products in Tropical Asia

movement for empowerment in Southeast Asia is both ab• sent at present and unlikely in the foreseeable future. This difference may be due chiefly to one fact: Para rubber had to be planted in Southeast Asia, but not South America. By planting or not, and thus by involving or not issues of nature versus culture and public domain versus pri• vate domain, the potential for involvement of the state and state-supported economic elites, versus the potential for de• velopment of the smallholder, is completely different. The structural role of Para rubber in South America is analogous

to that of the native forest rubbers in Southeast Asia; and what happened to the native rubbers in Southeast Asia—in terms of local people losing control to outside forces—is what happened to Para rubber in South America.

CONCLUSIONS Many of those interested in the development of non-timber forest products regard such development as basically a tech• nical and economic challenge. There is much discussion of the economic values of these products and much analysis of the costs and benefits of returns from various products com• pared with those from other uses of the forest (e.g., by log• ging or shifting cultivation). In fact the importance of politi• cal-economic factors may make such fine cost-accounting irrelevant—if not harmful since it draws attention away from the real issues. The analysis presented here shows that the most important issue may be not the size or efficiency of the return but rather who receives it (and who might like to re• ceive it instead). The development of non-timber forest products is not first a technological and economic challenge, but rather it is first and foremost a political challenge, involving mediation of relations between forest peoples and the state. This is at• tested to by the very different outcomes of Para rubber devel• opment in Southeast Asia and South America. The tree, the plant, is identical, but the two histories are completely dis• similar. This illustrates once again the ability of political- The Impact of Cultivation on Peasant-State Relations 67 economic factors to overpower biological factors in tropical forest development.

Acknowledgments—This analysis was written during the tenure of fellowships in the East-West Center's Program on Population and Program on Environment (in the former case with support from the United States Agency for International Development). The field research upon which the analysis is based was supported by the National Science Foundation (Grant #GS-426os). None of the aforementioned organiza• tions necessarily agrees with the analysis and opinions pre• sented in the paper, however, for which the author alone is responsible.

A subsequent version of this paper travel by river from the coast of was published in Society and Kalimantan (at Pontianak), and Natural Resources, Vol. 8, 199s, two days' travel by foot from the by Taylor & Francis, Inc., international border with Sarawak Washington, D.C. Another study (at Lubok Antu).. I gathered of the same topic was published in additional data during six years of Economic Botany, Vol. 48, No. 4, subsequent work on lava, with 1994, by the New York Botanical periodic field uips to Kalimantan Garden, Bronx, New York. and Indonesia's other principal islands. 1. Padoch (1980, 478), in contrast, maintains that among some Iban, 4. A rubber "smallholder" is at least, "The traditional 'jungle defined in the literature as produce'—da mar, camphor, gutta• someone with less than 25 percha, and rattan—has been hectares (Barlow and Muharminto replaced by the cutting and 1982, 86). In practice, most working into posts and shingles of smallholdings are much closer to Borneo ironwood or 'belian' the Kantu' average of approxi• [Eusideroxylon zwageri)." mately 4 hectares.

1. Compare the development of 5. Gums, resins, and latexes are rubber cultivation from outside all exudates, viscous liquid com• agriculture to the development of pounds that are produced natu• tree crops from within agriculture rally by forest plants and emerge (see Eder 1981 on Philippine from injured tissues; latexes, as orchards). distinct from the other two prod• ucts, are suspensions of salt, 3. I carried out research among hydrocarbons, and other organic the Kantu' for two years. The sub• compounds in waters (de Beer and group that I studied, the Melaban McDermott 1989, 37, 38). Kantu', live along a western, secondary tributary of the Kapuas 6. Richards (1981, ro6) also writes River, which was then two weeks' that gutta expeditions usually 68 Society and Non-timber Forest Products in Tropical Asia

preceded migration. Lian (1988, tion," due to overly intensive and rr8) writes more generally that destructive exploitation in South the gathering of forest products America (Purseglove 1968, 148). was the historical genesis of the 13. Another traditional technol• Iban custom of berjalai, or ogy that resembled and thus may "expedition." have facilitated the adoption of 7. Plants producing commercially Para rubber was the cultivation of viable latexes in Indonesia in• a wide variety of trees bearing clude, in addition to the intro• fruit, oil, and other valued (and in duced Para rubber, five other some cases marketed) commodi• introduced trees, two native trees, ties (cf. Cramb t988, 112). The two climbers, five lianas, and one perennial character of these crops climbing shrub (Anonymous 1935, may suggest, however, that this cited in CAPD r982, 352s). technology did not itself develop until the sedentarization of 8. It seems unlikely that the Borneo's tribal peoples, and their earlier-mentioned migration of the adoption of rubber, was relatively Kantu' to the Empanang valley is well advanced. attributable to this boom in the European trade in the 1840s 14. Some sources of gutta-percha, because just four decades later, at however, had to be felled to obtain the height of Kantu' settlement of their latex (Burkill [1935] 1966, the valley, they already numbered 2:1651), and even those trees that more than thirty longhouses could be exploited by tapping (Dove 1985, TI). This suggests quickly succumbed to pests or that the trade, which prompted disease. On the other hand, this the Kantu' migration, must have was also to some extent true of been the more ancient trade Para rubber until an improved within the region. tapping technology was developed, as a result of painstaking research 9. A modest resurgence in in Southeast Asia. Traditional interest in jelutong began in 1922, methods of tapping, employing with the discovery that it could be the hatchet in South America and used in (Burkill the cutlass in Southeast Asia, [1935] 1966, 1:891). were harmful to the tree (Barlow 10. The two most important r978, 21-21; Purseglove 1968, native latexes—gutta-percha from 161-64). Palaquium spp. and jelutong from 15. Lindblad (1988, 115) errone• Dyera costulata—are still traded, ously concludes that Para rubber albeit not in volumes approaching production responds more slowly historic levels (de Beer and to market conditions than the McDermott 1989, 40). production of gutta-percha, 11. The center of production was because of the time-lag from around the city and state of Para, rubber-planting to maturity. In from which the vernacular term the short term, however, the for Hevea brasiliensis comes. response mechanism for Para rubber is not planting but r2. The effort to plant rubber in resumption of tapping, which Asia was initiated "in order to consumes only two to four days— maintain the world's supply, far shorter than the time required which was in danger of extinc• The Impact of Cultivation on Peasant-State Relations 69

to find and exploit a stand of Shorea spp.], pantung \Dyera spp.], gut to-yielding trees in the forest. and maja [Terminate spp.) trees which had provided local people 16. The causal direction here is with a source of income. The actually two-way, that is, the company has ignored the protests adoption of rubber to some extent of the Riam Batang villagers, who promoted—just as it was used to make a reasonable living promoted by—the intensification from the products—nuts and of swidden agriculture. Posses• resins—from these trees." sion of producing rubber gardens provided an incentive for a more i9- In fact, the Kantu' say that sedentary pattern of settlement, their ancestors also planted some for example. of the native rubber trees, in particular Palaquium spp., the 17. Cf. Lian (1988, 118) on a major source of gutta-percha (cf. similar law among the Kenyah. Wijk 1941, cited in CAPD 1982, This law obviously only applied to r344, on planting of Dyera lowii tree species from which latex was and D. borneenis}. Presumably extracted by tapping, as opposed such trees—especially if planted to felling. in proximity to a house or to some 18. Note this report of the other easily recognized property— destruction of jelutong trees, were less vulnerable than among others, from the Indone• naturally grown trees in the sian daily newspaper Suara middle of the forest. Pembangunan (cited, in transla• 20. Even Charles Brooke (the first tion, in Down ro Earth 1990, ro}: English rajah of Sarawak), "A logging company identified otherwise known for his hostility only as FT SBK with a concession to European plantation interests, in Kotawaringin Timur district, gave Europeans a jelutong- Central Kalimantan, is suspected processing monopoly in Sarawak of cutting down thousands of (Reece 1988, 28-29). tengkawang [Isoptera spp. and

Anonymous. 1935. Rubber- Boeke, J. H. 1953. Economics and leverende planten in Ned.-Indie economic policy of dual societies: (Latex-producing plants in the As exemplified by Indonesia. New Dutch East Indies, with the York: Institute of Pacific Rela• exception of Hevea brasiliensis). tions. Algemeen Landbouw Weekblad Booth, Anne. 1988. Agricultural van Nederlandsch-lndie 11:1186- development in Indonesia. Asian 89. Studies Association of Australia, Barlow, Colin. 1978. The natural Southeast.Asia Publications Series rubber industry: Its development, No. r6. Sydney: Allen and Unwin. technology, and economy in Burkill, I. H. [1935I 1966. A Malaysia. Kuala Lumpur: Oxford dictionary of the economic University Press. products of the Malay Peninsula. Barlow, Colin, and Muharminto. 1 vols. Reprint, Kuala Lumpur: 1982. The rubber smallholder Ministry of Agriculture and Co• economy. Bulletin of Indonesian operatives. Economic Studies 18(2): 86-119. 70 Society and Non-timber Forest Products in Tropical Asia

CAPD {Centre for Agricultural Down to Earth. 1990. Loggers Publishing and Documentation). destroy local forest product 1981. Indonesian forestry supplies. No. 7(March): 10. abstracts: Dutch literature until Drake, Richard Allen. 1982. The about i960. Wageningen. material provisioning of Mualang CPIS (Center for Policy and society in Hinterland Kalimantan Implementation Studies). 1993. Barat, Indonesia. Ph.D. diss., Towards a planting materials Michigan State University. policy for Indonesian rubber Dunn, F. L. 1975. Rainforest smallholdings: Lessons from past collectors and traders: A study of projects. Agriculture Group resource utilization in modern Working Paper No. 14, CPIS, and ancient Malaya. Monographs Jakarta. of the Malaysian Branch of the Cramb, R. A. 1988. The commer• Royal Asiatic Society No. 5, Kuala cialization of Iban agriculture. In Lumpur. Development in Sarawak: Eder, James F. 1981. From grain Historical and contemporary crops to tree crops in the Cuyunon perspectives, ed. R. A. Cramb and swidden system. In Adaptive R. H. W. Reece, 105-34. Monash strategies and change in Philip• Paper on Southeast Asia No. 17, pine swidden-based societies, ed. Center of Southeast Asian Studies, Harold Olofson, 91-104. Laguna: Monash University. Forest Research Institute. de Beer, Jenne H., and Melanie J. Effendi, Suryatna. 1985. Improve• McDermott. 1989. The economic ment of smallholder rubber value of non-timber productivity in Indonesia. products in Southeast Asia. In Smallholder rubber production Netherlands Committee for and policies: Proceedings of an IUCN, Amsterdam. international workshop held at Dillon, H. S. 1985. Development the University of Adelaide, ro8- of rubber smallholders in north r5. Canberra: ACIAR. Sumatra. In Smallholder rubber Freeman, J. D. r970. Report on the production and policies: Proceed• Iban. London: Athlone Press. ings of an international workshop held at the University of Geddes, W. R. 1954. The land Adelaide, 116-26. Canberra: Dayaks of Sarawak. Colonial ACIAR. Research Study No. 14. London: HMSO. Dove, Michael R. 19B5. Swidden agriculture in Indonesia: The Government of Indonesia. 1992. subsistence strategies of the Statistik Indonesia: Statistical Kalimantan Kantu'. Berlin: yearbook of Indonesia. Central Mouton. Bureau of Statistics, Jakarta.

. In press. Rice-eating Hudson, Alfred B. 1967. Padju rubber and people-eating Epat: The ethnography and social governments: Peasant versus state structure of a Ma'anyan Dajak critiques of rubber development in group in southeastern Borneo. colonial Indonesia. Ethnohistory. Ph.D. diss., Cornell University, Ithaca, New York. The Impact of Cultivation on Peasant-State Relations y\

Jessup, Timothy C, and Andrew P. the tropical rain forest, ed. Julie Vayda. 1988. Dayaks and forests of Sloan Denslow and Christine interior Borneo. Expedition 30(1): Padoch, t27-4i. Berkeley: 5-i7 University of California Press.

King, Victor T. 1988. Social rank Peluso, Nancy. 1983. Markets and and social change among the merchants: The forest product Maloh of West Kalimantan. In The trade of East Kalimantan in real and imagined role of culture historical perspective. M.Sc. in development, ed. Michael R. thesis, Cornell University, Ithaca, Dove, 219-53. Honolulu: New York. University of Hawaii Press. Potter, Lesley. 1988. Indigenes and Lian, Francis J. 1988. The colonisers: Dutch forest policy in economics and ecology of the south and east Borneo (Kaliman• production of the tropical tan), 1900 to 1950. In Changing rainforest resources by tribal tropical forests: Historical groups of Sarawak, Borneo. In perspectives on today's challenges Changing tropical forests: in Asia, Australasia, and Oceania, Historical perspectives on today's ed. J. Dargavel, K. Dixon, and N. challenges in Asia, Australasia, Semple, 127-49. Centre for and Oceania, ed. J. Dargavel, K. Resource and Environmental Dixon, and N. Semple, 113-25. Studies, Canberra. Centre for Resource and Environ• Purseglove, J. W. t968. Tropical mental Studies, Canberra. crops: Dicotyledons. Harlow, Lindblad, J. Thomas. 1988. U.K.: Longman. Between Dayak and Dutch: The Reece, R. H. W. r988. Economic economic history of southeast development under the Brookes. Kalimantan, 1880-1942. In Development in Sarawak: Verhandelingen No. 134. Historical and contemporary Dordrecht: Foris Publications. perspectives, ed. R. A. Cramb and McFadyean, Sir A. 1936. Rubber R. H. W. Reece, 21-34. Monash News Letter, September 30. Paper on Southeast Asia No. r7. Center of Southeast Asian Studies, Murphy, Robert F. [T960] 1978. Monash University. Headhunter's heritage: Social and economic change among the Richards, Anthony. 1981. An Mundurucu Indians. Reprint, Iban-EngUsh dictionary. Oxford: New York: Octagon Books. Clarendon Press.

Padoch, Christine. 1980. The Sandin, Benedict. 1980. Iban adat environmental and demographic and augury. Penang: Universiti effects of alternative cash- Sains Malaysia. producing activities among Thee, Kian-Wie. r977. Plantation shifting cultivators in Sarawak. In agriculture and export growth: An Tropical ecology and develop• economic history of east Sumatra, ment, ed. J. I. Furtado, 475-81. 1863-1942. Jakarta: LEKNAS-LIPI. Kuala Lumpur: International Society of Tropical Ecology. Tremeer, R. E. T964. The early history of rubber planting in . 1988. People of the Sarawak, 1880-1910. Sarawak floodplain and forest. In People of Gazette 90:50-52. 72 Society and Non-timber Forest Products in Tropical Asia

Uljee, G. L. 1925. Handboek vooi D. borneensis) (Notes on the de Residentie Westerafdeeling van regeneration of the pantung Borneo. Weltevreden: Visser. [Dyera lowii and D. borneensw||. Unpublished. Weinstock, Joseph A., and Napoleon T. Vergara. 1987. Land Wolf, Eric R. r966. Peasants. or plants: Agricultural tenure in Englewood Cliffs, N.J.: Prentice- agroforestry systems. Economic Hall. Botany 41(2): JIMJ, Wolters, O. r967. Early Indone• Wijk, C. L. van. 1941. Enkele sian commerce. Ithaca, N.Y.: aantekeningen over'de verjonging Cornell University Press. van de Pantoeng (Dyera lowii en Extraction Interactions: Logging Tropical Timbers in West Kalimantan, Indonesia

Competition between large-scale industrial forest exploita• tion for timber and village forest management for agricul• ture and forest product extraction threatens both biodiversity and household welfare. Although the two systems for ex• tracting forest products have been seen as potentially sus• tainable under appropriate separate management, the inter• active impacts of their competition within the same forest territory have been analyzed only in limited ways. Not only is it crucial to understand the effects of individual forest management systems, but it is equally critical to understand the interaction of the various systems, given the multiple demands characteristically made on most of the world's for• ests today and in the future. This has been shown in the literature on Latin , particularly in regard to the adverse social and ecological impacts of cattle ranch• ing, mining, logging, and pioneer swiddening (Bunker 1985; Hecht, May, and Anderson 1987,- Schmink and Wood 1987; Norgaard 1984).

In Southeast Asia, the study of competing forms of for• est land use has focused on the conflicts between industrial forms of forest extraction or conversion, such as logging or plantation development, and traditional forest-based swidden .agriculture (e.g., Dove 1983, 1986; Anderson 1987; Hafner 1990). In trying to understand the relative impacts of differ• ent land uses on , sustainable management, community welfare, or state revenues, these and other ana• lysts also have examined the costs and benefits (social, po• litical, and economic) of these forest management systems in isolation from each other (see also Conklin 1957; Dove 74 Society and Non-timber Forest Products in Tropical Asia

1985; Weinstock 1983). Moreover, little (if any) field research has compared the ecological or sociological impacts of tradi• tional forest product extraction with industrial forms, and the interactions between extractive forms. The need to study villagers' management and extrac• tion of "minor" or non-timber forest products (NTFPs) has been recognized by the natural resources development com• munity (Gillis and Repetto 1988). The legal aspects of na• tional concerning customary law and commu• nity rights to the forest have been detailed for Indonesian forests in general (Zerner 1990). Other research on the re• gional trade in non-timber products and their roles within Dayak and other forest peoples' agroforestry systems has documented the critical role of trade in NTFPs to regional and local economies for more than a thousand years (Peluso 1983, 1986, 1989, 1992; Padoch 1982; Jessup 1989; Dunn 1975). Nevertheless, there is still a significant lack of analy• sis based on field studies of the direct and indirect effects of one system of forest extraction on other systems. This paper tries to redress a small part of the imbal• ance in the analysis of Southeast Asian forest policy by ex• amining a complex of interactions between different forms of forest use. In particular some of the indirect effects of log• ging operations on the production, trade, and management of Borneo ironwood—belian [Eusideroxylon zwageri)—are explored. A village in West Kalimantan (Indonesian Borneo), which was relatively isolated until this past decade, is used as an example. The rapid changes that developed when tim• ber operations began to approach the village's forest territory about seven years ago are examined. The village's relative isolation until recently highlights the initial impacts of na• tional management policies for "major" and "minor" forest products, including the infrastructures and technologies used for exploitation, on the traditional balance between the village's subsistence and commercial uses of the forest. The immediate outcome of these sudden changes has been in• creasing conflict within and between villages and between the village, state, and timber companies. Extraction Interactions 75

The paper also examines state attempts to foresee and prevent the two things that subsequently occurred—the overexploitation of a forest product that is difficult to pro• duce and the local people's loss of income and control over forest-cutting within their village territory. Conflict and re• source depletion resulted not only from the timber company's extraction of logs or penetration of logging roads, but also from the state's replacement of a well-entrenched system for governing access to forest products with an unrelated "mod• ern" form of management. State policies failed to protect the ironwood and local people's forest rights, because current state forest policies were developed from a contradictory frame• work vis-a-vis the value of the forest.

Indonesian forest policy addresses neither the interac• tions between the two conceptions of the forest nor the local people's ad hoc institutions of forest access bom of the clash between the two management systems. The traditional, state, and de facto forms of forest management and their relative impact on the management and benefit distribution from these forests are discussed here in further detail. This analysis is based on data from several sources. Field research was conducted in West Kalimantan in August and September 1990 in the initial phase of an FAO study on com• munity and household forest dependence.1 The research built upon previous work on East Kalimantan [Peluso 1983, 1986, 1992; Jessup and Peluso 1986) and a recent review for the FAO of the literature on household uses of non-timber forest products [Peluso 1989). The recent fieldwork focused partially on the changes in the nature of forest villagers' dependence on the forest. The village discussed in this paper was purposely se• lected for its location in the middle of a recently licensed (1983) timber concession and because of the parallel surge of ironwood exploitation in the selectively logged portion of the village's forest territory. Data were collected from group, key informant, and household interviews. Household interviews were conducted with two samples: eight of the eleven house• holds still living in the remaining longhouse, and a sample 76 Society and Non-timber Forest Products in Tropical Asia

of eight households selected randomly from the forty-eight poorest households (none of the second sample overlapped with the first). The "poorest" were defined by key informants to mean those households unable to produce enough rice to be self-sufficient from one year's harvest to the next. House• holds were ranked by a key informant, and the sample was selected from all those characterized as "never self-suffi• cient." The ranking process was secret because of the em• barrassment associated with the inability to produce enough rice. Interviews confirmed that these households were dis• advantaged.2 In addition to discussing various aspects of forest and land management with villagers, I compiled a history of for• est products collected and managed for subsistence and com• mercial purposes. Discussions on this latter topic were in• variably connected to the impacts that the encroachment of the timber company had on local people's methods and capa• bility of protecting "their" forest resources. Besides the vil• lage study, timber company managers and employees, sub- district officials, and provincial and district foresters were interviewed.

COMPETING TRADITIONAL LAW, FOREST PRODUCTS. AND FOREST

FORMS OF FOREST VILLAGES IN KALIMANTAN MANAGEMENT:

INSTITUTIONS Among Dayak swidden cultivators of Kalimantan, custom-

OF ACCESS J OR ^U^UM adat nas iong governed patterns of forest AND CONTROL , , , management by determining people's rights of access to for• est products, forest land cleared for agriculture, and trees planted or growing wild outside the forest. Serving as a struc• ture of social control, the system depends for its power on both material sanctions and psychological punishment im• posed on those who break the accepted rules of custom and action. Sanctions require the rule-breaker to pay certain fines to the wronged party and to sponsor a meal between the par• ties concerned and the arbitrators, or adat experts. Both the fines for particular "crimes" and the amount Extraction Interactions 77

and type of food for the peace feast vary according to the severity of the action's consequences. Fines generally include certain numbers of specific types of gongs, ceramic jugs, and other pottery or, more recently, cash equivalents of these goods. Besides their application to crimes against private prop• erty in land and trees, adat fines are also imposed for other social transgressions such as adultery, divorce, slander, and, most severe of all, the taking of another person's life, which requires compensation for every part of the lost individual's physical and spiritual self with appropriate and expensive goods. While the economic burden of an adat fine can be shared with sympathetic family members, the psychological burden of having been publicly shamed is experienced largely by the individual. Fear of this shame [malu] is arguably a more powerful deterrent to would-be rule-breakers than the economic costs.

The rights to convert or use particular forest territories and products are conceived in layered or nested sets of adat access rules. These vary slightly between villages and tribal groups.3 For some of the smaller Dayak groups,4 the broadest of these realms of adat would consist of all Dayaks in the same language group who live in contiguous villages and who would conceive "their territory" to include the forest falling within the bounds of particular geographical markers such as mountain ranges or rivers. The next largest territory would be that of an organizational entity that consists of three or four villages of Dayaks who speak the same language and share a temenggung adat, or customary law "lieutenant" who carries more authority than the kepala adat (head of cus• tomary law) operating at the village level. In some places this organization was imposed in the eighteenth and nine• teenth centuries by Malays from nearby sultanates (hence the Malay term temenggung) and has been integrated within the Dayak system. Nevertheless, such political groupings were based on traditional inter-village or inter-longhouse ties and the sharing of a common language.

Within these larger units, specific villages or longhouses recognize boundaries with other villages that marked the 78 Society and Non-timber Forest Products in Tropical Asia

village's territory. This territory—its lands and products—is considered village property until the conditions for private claim or control are invoked by individuals or groups. Thus these resources were essentially open access resources avail• able to local villagers until exclusive ownership of specific forest products or lands were recognized (cf. Jessup and Peluso 1986, 517). Privatization of forest product claims was recognized as a result of both prior claim and the investment of labor in the products' management (i.e., its production, protection, or maintenance). While forest clearing for swidden agricul• ture provides an obvious example of this principle, the man• agement of trees and other vegetation can also be illustrated. Clearing the brush around a particular tree to facilitate ex• ploitation or to eliminate species competition and thus to encourage fruit production would impart rights to the clearer. Similarly, finding a copal or resin tree (e.g., Agathis) in the forest and protecting it from fire during swidden clearing, or wounding the tree to encourage exudation of its or res• ins, involves investing labor in tree management for its prod• ucts. Planting trees in the forest or in old swiddens clearly imparted private rights to the planter. Products such as tengkawang (illipe nuts), although traditionally classified as "forest products" by state foresters, are actually collected from planted varieties of this tree [Shorea spp.). The ecological characteristics of a particular product also affected the nature of the access rules or expectations of products in a particular village's ulayat (territory). In all cases, for example, an outsider wishing to extract forest products from a village's territory is required to pay a "tax," in kind, to the village as common property owner or, if applicable, to a claimant who has been vested with private rights by prior claim and management (see also Peluso 1983). This tax was described as "two parts of ten," or 20 percent of the harvested product. A payment might consist of two rattan bundles per ten collected, two ironwood posts per ten cut; twenty bird nests per 100 collected; or two "balls" of coagulated resin per ten collected. Extraction Interactions 79

Most important here, however, is how the decision to allow access is adapted once a resource used in local produc• tion systems becomes scarce or if the resource is difficult to regenerate. In such instances, the subsistence or use value of the resource is given priority and valued higher than its ex• change value on local or regional markets. In this West Kalimantan village, even renewed products within the me• dium term were so restricted. Slender rattans, for example, grew wild in the forest around the village and were made into tools and baskets for producing and storing rice, respec• tively. They were also made into mats to dry rice in the sun and for use at home and in formal or ritual gatherings. Rattan mats and baskets were also an important source of cash income or trade items for nonlocal goods. Villagers traded mats across the Malaysian border and sold mats and baskets to buyers who came to their homes from other vil• lages and towns. Seed and harvest baskets were in particu• larly high and constant demand. Yet, in 1990, villagers would not sell their baskets, not even old baskets, because the sources of rattan accessible on a day trip (a three-hour walk from the village) were dwindling, and people were afraid they would not have enough baskets for their own needs. The vil• lagers did not cultivate rattan; rather, they limited its use and sale. And, although bamboo also grew wild in the forest, several informants claimed that "things were really bad if you had to make your baskets out of bamboo,- at most they would last a season or two." Thus, although substitution was possible, it was not preferred. In sum, the forest, aside from serving as a storehouse of nutrients to be accessed through slashing and burning to cre• ate a rich layer of ash for fertilizing agricultural crops, was locally viewed as a storehouse of materials to be eaten, to cure disease, to be made into useful tools and building mate• rials, and to be exploited for cash income when local needs or the market demanded. Villagers regulate access to forest products through a system of social control recognized by members of the same village, by individuals living in other villages and towns, and by traders. Regulating accessibility 8o Society and Non-timber Forest Products in Tropical Asia was particularly important for products that were difficult to regenerate and considered critical components of the subsis• tence economy. In such cases, subsistence uses were given priority over commercial exploitation and the potential needs of all village households weighed in deciding who should have exploitation rights.

NATIONAL LAW, DEVELOPMENT, AND LOGGING IN

KALIMANTAN

Indonesian forestry law changes the emphasis of control on forest access, giving the state control of all land, water, and the "natural riches" contained within the forest lands [kawasan hutan) of the country (Zerner 1990, 15). The Min• istry of Forestry has authority to manage and control access to the rights of forest exploitation. Although forests are sup• posed to be managed for multiple uses, in the past twenty years the rights to log have superseded virtually all other rights to exploit or use production forests (Barber 1989, in Zemer 1990). Indeed, certain forests are designated by the state as suited to production based solely on the presence and accessibility of timber. The value of forests to the Ministry of Forestry and to the Indonesian state in general lies primarily in timber ex• ploitation. About 40 percent or close to 64 million hectares of Indonesian forests are in production forests; this consti• tutes almost 60 percent of all production forest in Southeast Asia (Repetto 1988, 43). The most and the best tropical for• ests are found in Kalimantan, Sumatra, and West Irian. So lucrative and desirable was timber exploitation that timber concessions allocated by 1983 exceeded the amount of land classified as production forest (ibid.). Most production forests in Kalimantan have been leased to private or parastatal companies via territorial concessions [hak pengusahaan hutan: HPH) defined as:

A right to exploit the forest in a Designated Forest Area, through timber cutting, regenerating and maintaining the forest, and processing and marketing forest products, in ac- Extraction Interactions 8r

cordance with a Forest and Exploitation Workplan, as required by current regulations, and based on criteria of conservation and sustainable production.5

The Ministry of Forestry reserves the right to deter• mine which forest territories should be classified as produc• tion, protection, reserve, or conversion forests, and to regu• late "the activities of communities living in and on the bor• ders of forest areas." In terms of the relationship between national forest law and traditional rights or customary law governing access to the forest, traditional law is legally viewed as subsidiary to national law.

The enjoyment of adat rights, whether individual or com• munal, to exploit forest resources directly or indirectly . . . may not be allowed to disturb the attainment of [the tenets of Basic Forestry Law). (Barber 1989, in Zerner 1990, 114)

In Kalimantan, as in many of the forested areas of Indo• nesia, adat rights have always included rights to clear forest for swidden agriculture. Today, however, "government policy on shifting cultivation is directed wholly at stamping (shift• ing cultivation) out" (ibid., 116). Government policy makers and planners have been so preoccupied by their inability to stop swidden cultivation that they have largely ignored other aspects of forest uses by the swidden communities. Early for• est regulations peripherally addressed local people's rights to harvest forest products in HPH areas, but were capped by the decision to freeze the validity of traditional forest access mechanisms while timber harvesting operations were in pro• cess within the forest community's territory. In 1988, community rights to those forest products not harvested by the HPH were recognized as valid, and respon• sibility for damage to them was laid on the company (ibid., 115). Where compensation for damaging forest products is paid (and it is not always paid), it typically has been limited to rubber and fruit trees (ibid., 116; Dove 1986). Given that these trees have been managed (planted, cared for, tapped, or harvested) in ways comprehensible to and accepted by the company managers and the state, this pattern is not surpris• ing. Moreover, these products are viewed as agricultural prod- 82 Society and Non-timber Forest Products in Tropical Asia ucts to which people have more universally accepted rights than forest products. Local people's systems for protecting village supplies of difficult to renew forest products, such as ironwood, are not recognized as valid because their systems are not seen to have taken a direct role in ironwood's cre• ation or protection. In particular, where a community or a household takes a conscious decision to leave a certain amount or area of forest unmanaged in order that certain products will grow there, the management aspects of this decision are disregarded by policy makers and subsequently by field foresters and timber company personnel. Cases of dispute between timber companies and com• munities affected by or contained within their concession areas, at least in Kalimantan, are generally not arbitrated by state foresters. Nor is it usual for district and subdistrict ad• ministrators to monitor and protect the traditional rights to forest products of the communities within their jurisdictions. Subdistrict officials attribute this partially to the difficulty of travel to remote villages or to the broad extent of the subdistrict's territory. In addition, few administrators con• sider the protection of villagers' rights to forest products as part of their job descriptions. The subdistricts become in• volved only when villages advance complaints to the subdis• trict office or directly confront timber company foremen working in their areas. At times of extreme tension or dam• age to village property, villagers have organized and attacked timber company personnel and property. Then, subdistrict officials, police, and the local military contingent are called in to halt the violence and damage and to prevent personal injury. These kinds of confrontations occur much more fre• quently than reported by the local press. One timber com• pany had been attacked at least five times by local villagers, two times in the past two years by the same villagers. Three of these attacks resulted from land and compensation dis• putes; the two most recent attacks occurred as a result of controversies over controlling access to ironwood growing in the village territory. The reasons for this conflict can be Extraction Interactions 83

traced in part to the role of timber operations in the exploita• tion of ironwood and the failure of policy to address the in• teractions between the two forms of forest management rep• resented by the village and the timber company.

IRONWOOD: Borneo ironwood [Eusideioxylon zwageri) is one of the hard-

MAN AGE MENT AND est, densest known; it is perhaps the most dense tropi-

NONMANAGEMENT caj hardwood, capable of sinking in water (Burkill [ 193 5 ] 1966, 986). It reportedly takes several hundred years of growth to reach this high wood density. The seed takes a year to germi• nate, and although it is a crucial component of village forest subsistence supplies, villagers usually consider it too diffi• cult to plant.6 A few trees occur per hectare in these diptero- carp forests. As both a commercial and a subsistence forest product (having high exchange and use values), ironwood has a long history. It is used or sold for roof shingles, building frames, bridges, and boards. As early as 1820, the Chinese exported Borneo ironwood to China (ibid.). Some swiddeners of Borneo have been known to move their residential sites because subsistence supplies of ironwood have dwindled (Jessup 1989). The wood's most valuable characteristic from both a commercial and a subsistence perspective is that it is very resistant to termites and other tropical wood-eating insects and fungi. For this reason, the wood is in great demand for construction throughout Kalimantan. Indonesian law forbids its exportation. For the interior peoples of Kalimantan, the wood is also important. In many regions of interior Borneo, ironwood is exclusively used in traditional longhouses, including the thick posts forming the frame of the longhouse, the walls and floorboards, the roof shingles, and the log staircases lead• ing up to the longhouse, which traditionally was elevated on stilts. Today, where single family houses have become the norm in Dayak villages (because of government "encourage• ment"), ironwood is still employed in much of the construc• tion, even where supplies are dwindling. Ironwood is priori- 84 Society and Non-timber Forest Products in Tropical Asia tized for constructing house frames. Where transport is diffi• cult, precluding the import of zinc sheets or expensive clay tiles, ironwood shingles are still the preferred roofing mate• rial. Where ironwood has been depleted or become too ex• pensive, ironwood shingles are being replaced by thatch made from the leaves of the arenga palm [Arenga sp.). Such roofs must be replaced every year or two, whereas ironwood shingles last from three to five generations. In Sungai Dalam,7 ironwood was never planted but was used for all the purposes mentioned earlier. It was thus im• portant to retain a supply within the forest for the villagers' future needs. Young families, whether they plan to build a new apartment onto an existing longhouse or construct a separate single-family dwelling needed ironwood at least for the house frames; otherwise, their homes would be food for termites within a decade. When villagers needed ironwood, they could take it either from the uncut forest on the outer perimeter of their village territory, or from one of the two "forest islands" within the area cleared for swiddens. These forest islands serve as reserve areas for forest products on an open access, first-come, first-claim basis, because of a unique ecological characteristic. The areas are sprinkled with large boulders above and below ground that occur frequently enough to thwart efficiency in swidden farming. People use the 2-hectare forests and perimeter areas as reserve stocks of forest products, harvesting ironwood, other wood used in construction, medicinal plants, and rattan as needed. Some• one needing wood for construction from the village territory (not from someone's private land) stakes claim to a tree by painting his name on it. Because of the ironwood's weight and the lack of roads into the village, it was virtually never cut for sale until 1983.

Under national forestry policy, ironwood cutting is re• stricted to trees over 60 centimeter diameter at breast height. Ironwood has typically been described by officials as a "people's species" because of its long-established pattern of usage and importance in village subsistence systems. Also because it is a heavy "sinker" type of species in a system Extraction Interactions 85 where logs have been floated downriver to market, it has gen• erally been uneconomic for a timber company to harvest it. When logs are transported by truck this is less of an issue, although the wood's density and weight cause greater wear on logging and transport equipment. Its reputation as a tradi• tional species also makes timber company personnel reluc• tant to harvest it. In 1979-80 the government implemented a national forestry policy (approved nearly a decade earlier) requiring permits to harvest minor forest products. This policy was meant to formalize and thus protect the rights of forest-de• pendent communities to harvest and sell commercial forest products through the establishment of state-controlled co• operatives (see Peluso 1983). Three kinds of minor product concessions were provided for in these HPHH [hak pemungutan hasti hutan: permit to harvest forest products): (1) the rights to collect non-timber forest products such as rattan, resin, bird nests, and other non-timber, nonwood prod• ucts; (2) the rights to harvest wood in areas up to 100 hect• ares for village needs, including the rights to sell the wood harvested in these areas; and (3) the rights to collect dead- wood in the forest. Long-maturing woods such as various species of ironwood, ebony, and others were categorically grouped with other hard and soft woods that matured at faster rates. Proposals and plans for HPHH were to be submitted to the provincial forestry office by village KUDs [koperasi unite desa), the state-created cooperatives. Besides documenting the areas to be cut, the permits were to indicate the channels and individuals to whom the products would be sold. This procedure was also in the inter• est of the Forest Department because it ensured they would receive their taxes. Individuals or companies were also re• quired to purchase a license for resale of the forest products extracted under the specific HPHH. In addition to its stated intent to protect the forest community's rights, the plan would help formalize the trade in people's wood or non-tim• ber products, provide a check on smuggling or untaxed trade, and facilitate government taxation (i.e., a fee would be paid 86 Society and Non-timber Forest Products in Tropical Asia

for the trading license, and records could be checked for an• nual taxation on production and trade). Within the past several years, however, several forest provinces of Indonesia have revoked the HPHH system for the harvest and trade in woods, except for non-timber prod• ucts in some places. In West Kalimantan, the change came in 1989. The policy failed for several reasons, but primarily because the government did not explicitly build on existing village institutions for resource management, nor did they communicate the new system to many of the villages that would eventually be affected by the policy. These failures can be partially attributed to the industrial timber orienta• tion of Indonesian foresters and forestry policies. As a result, the local commercialization of a "traditional product" has negatively affected both the resource supply and the social or• ganization established to manage it. The exploitation of iron- wood in the forest of Sungai Dalam illustrates what can hap• pen when villages are not informed about the government- sanctioned system meant to "protect" their rights and thus adapt their own system of access control as a last-ditch effort to enjoy some of the benefits of "modern" forest exploitation.

IMPACT OF THE Through technology and infrastructure, the timber industry TIMBER INDUSTRY has brought new ways and means of exploiting the forest to ON IRONWOOD the people of Sungai Dalam. The ways in which they have EXTRACTION been integrated into traditional uses of the forest have clashed, however, with the state's vision of what is appropriate forest use. Yet, when roads began to open the region's forest, allow• ing passage of trucks, goods, forest products, and people, the roads were not accompanied by foresters or other officials informing the local people how they could organize to mar• ket or protect the forest products that form an integral part of their subsistence system. The only official who moved to the village was a soldier posted at the border area for security reasons and ostensibly "to help develop the village."8 Soon the accoutrements of modern forest exploitation accelerated and extended everyone's opportunities for forest Extraction Interactions 87 exploitation, facilitating the cutting of Borneo ironwood. Chainsaws and roads make the tasks of tree-cutting (for swiddens, local construction materials, or wood sales) and wood transport easier for local people, increasing access to outside markets. They also facilitate outsiders' access to re• mote areas rich in unexploited forest products. The HPHH system of forest product management was supposed to miti• gate the impact of outsiders and reserve the benefits of forest exploitation for local people. But five or six years after the main logging road was established within a three-hour walk of the village, and over the many years that ironwood has been taken by truck by outside buyers, no one in Sungai Dalam had heard of the HPHH permit system, nor was a state-sponsored cooperative ever formed for selling rice or forest products. Instead, local people applied the adat system of access to ironwood trees. Adat is a dynamic system with a histori• cal capacity for adaptation to changing circumstances. The large-scale and immediate commercialization of one of the village's forest products and the difficulties involved in planned production of ironwood have led to some unprec• edented mutations in the nature of rights to harvest it. As the logging roads penetrated deeper into village territory and as more outsiders became aware that ironwood was avail• able, the rights of access to whole trees and not just the wood itself also became more commercialized. Access was facili• tated and exploitation was accelerated through availability of chainsaws; the first arrived in the village before the ap• proach of logging roads in the early 1980s. Briefly, the progression of change was as follows. As traders sought to buy ironwood, most villagers hurried into their forested territory and painted their names on as many trees as they could regardless of tree size and regardless of the needs of other village households. Formerly, only one or two trees would be claimed in advance by villages within the commonly owned forest because one or two trees of sev• eral meters circumference perimeter were sufficient to build a single structure for a household. The new pattern effec- 88 Society and Non-timber.Forest Products in Tropical Asia

tively resulted in claims by individuals of trees in whole sec• tions of common territory. Traders made arrangements with individuals rather than the village as a whole, and taxes on contracted trees also went to individuals. Second, a variety of tenure arrangements emerged for sharing rights, for contracting trees and cutting or hauling labor, for buying ironwood "territories," and for transferring rights.9 Labor arrangements in the forest resemble those of logging, in which a team of cutters work on a tree—one han• dling the chainsaw and the other clearing away the deepen• ing sawdust. One of the team or another laborer would haul the boards or posts made in the forest to the roadside. Every task had its wages or shares. If a team wanted to work on trees or territory claimed by someone else, they were required to pay the traditional two of ten boards or posts uncut. The tree owner would be given the labor invested in making the boards but not the labor required to haul the wood products out of the forest. Third, a variety of trading arrangements developed for selling the wood products, none of which were related to the permit system. The villagers were at a distinct disadvantage. Although some could buy or rent chainsaws, and although many local people were willing to invest the labor in cutting and hauling them to the roadside, they lacked the capital to either buy or rent trucks, to pay the unofficial "user fees" required of all noncompany trucks using the logging roads for commercial purposes, or to buy fuel for their chainsaws and food and cigarettes for their workers. The lack of capital led to development of working arrangements with outsiders. Outsiders worked with shopkeepers from other villages and towns who handled the sales of forest products and trade goods, with building contractors interested only in the wood, and with various ad hoc traders who had no experience in the business of moving ironwood but who had access to capi• tal or connections with the individuals and groups control• ling use of the road.10 Outsiders came not only in the form of individual traders who made arrangements with local villag• ers to cut down ironwood trees, but also as groups of outside Extraction Interactions 89 collectors "contracted" by traders who bought the rights to ironwood trees or territories from individual claimants. As more and more outsiders came in, villagers began to com• plain about numerous forms of "theft"—of standing trees that local people had claimed, of wood already cut and waiting to be sawn into posts or hauled out, and of the rights of other villagers to a share of the ironwood in their forest territory. The involvement of both insiders and outsiders in the commercialization of this resource had many interesting outcomes. Under the circumstances, neither the locals nor the outsiders tried to prevent the overexploitation of iron- wood or the ecosystems where it might regenerate. Outsid• ers were simply unconcerned with the future of the local land and forest; they were either villagers from elsewhere or people whose economic well-being was derived from a much broader range of opportunities. They used their access to the ironwood as a fast way of accumulating capital with mini• mal, short-term investments. Even those in the forest prod• uct trade for the long term had no reason to cultivate the kinds of patron-client relationships that they might seek if the resource would renew within a few years." Villagers, on the other hand, saw outsiders flocking into their territory with the means to rapidly extract their forest products and responded by trying to benefit in whatever way possible before others did. Many villagers involved in cut• ting ironwood had no intention (or ability) to accumulate capital, to buy capital goods, or to reinvest the profits in other capital-producing enterprises. Some used their profits to buy chainsaws but bore excessively high operating costs. Chainsaws also facilitated the cutting of old growth for swiddens (but were not as useful for cutting the smaller brush and slash). Sungai Dalam villagers made swiddens in old for• est as well as swidden fallows aged six to twenty years. Chainsaws reduced the task of clearing mature forest trees from thirty to sixty days per hectare to five days per hectare. Chainsaws could also generate cash for the owners; some villagers were willing to pay to reduce their own cutting tasks and rented the chainsaw and its owner for a day of forest 90 Society and Non-timber Forest Products in Tropical Asia cutting. At Rp 15,000 per day, the chainsaw owner earned as much or more as in a day sawing ironwood (see also Colfer 1987]. Ironically, the villagers exhibited exactly the kind of modern entrepreneurship that the Indonesian government encourages as part of its development ideology, transferring the technology of the logging sector to their subsistence and commercial activities. They have monetized labor relations, individualized property rights, and organized in small groups for trade. They organized en masse to demand the timber company clear the forest for a feeder road from the logging road to give them easier access to other elements of develop• ment—medical care, regular and inexpensive supplies of trade goods and foodstuffs, and access to markets for agricultural products. Villagers have yet to receive either accolades or gov• ernment assistance. The availability of the HPHH for wood collection in West Kalimantan has been discontinued as of 1991, largely because of the ironwood "problem."12 Provin• cial forestry officials expressed considerable disappointment that their plan to protect village rights to traditional wood products had not worked. Their stated reason for revoking this type of license was that they did not want to issue over• lapping permits for the same territories—they did not want HPHH in the same area as HPH. However, the problems with the plan to vest power in the village cooperatives and preserve village rights to certain wood products went deeper than this superficial concern with overlapping harvest permits. The KUDs [koperasi unite desa), the state-operated cooperatives, have been notoriously cor• rupt and ineffective throughout Indonesia. Most were set up as mechanisms for marketing and for purchasing rice and agricultural inputs required for ricean d food production. They are registered with and administered by the Ministry of Co• operatives and Transmigration. The HPHH plan called for establishing KUDs to handle the production and marketing of forest products, and permits for harvesting to be obtained Extraction Interactions 91

from the Provincial Forest Service office. In much of Kalimantan, as in Sungai Dalam, no KUD had ever been es• tablished, either for food or forest product marketing, nor was it clear whether foresters or other officials should have organized the establishment of a KUD for forest products. When the time came to collect and sell ironwood, villagers were never encouraged by officials to set up a KUD. Ironwood production and trade are now illegal activi• ties, just as swidden cultivation in mature forest is forbid• den. Both continue, however. No one manages ironwood in the forest either to maintain biodiversity or to benefit indig• enous people. The subdistrict office does not attempt to man• age the woodcutting or trade; these activities take place out• side its jurisdiction, within the ioiest and along the logging road. The timber company, which the Forest Service has placed in full charge of managing ironwood cut from its con• cession area, is not willing to manage what the local people perceive as their forest resource. Because of the heavy use of logging roads by the company during the day, ironwood trucks are allowed access to the forest only from 6 P.M. to 6 A.M. The security guards and their military partners, who super• vise which trucks are allowed in, have a financial interest in continuing the trade. The size of the trees cut in the forest or who cuts them is not monitored by them. Finally, the village's ad hoc system of management lacks a means of ensuring a fair distribution of the benefits of iron- wood cutting. Within a few years, the ethics of access to this resource has all but disappeared.

SUMMARY AND Borneo ironwood takes more than a couple of generations to CONCLUSIONS grow to its most useful (some might say "optimum") size. For this and other reasons, ironwood can be considered to be difficult to produce. Yet, after thousands of years of use by the tribal and nontribal inhabitants of Borneo and at least two hundred years of commercial use, the wood's future seemed unthreatened. Now, other forms of forest extraction, 9i Society and Non-timber Forest Products in Tropical Asia most notably timber exploitation, and the availability of simple, yet efficient machine technologies are abetting the rapid depletion of ironwood. Ironwood is being overexploited in Kalimantan (and likely so throughout Borneo) because of environmental and sociopolitical factors. Changes in the modes and scale of for• est timber exploitation have not been accompanied by simi• larly paced changes in local institutions and legal procedures for traditional forest uses. Where policies or procedures to address this issue have been developed, government officials have neither communicated them to villagers nor made ef• fective use of existing forms of local resource management. Where no control or effective action has been taken, indig• enous people's responses to the changes have followed their own courses. Their underlying goals, however, have corre• sponded to the general goals of the Indonesian development ethic, which are (i) to be modern, (2) to engage in market- oriented activities, and (3) to develop self-sufficiency.'3 As these values are applied in the forest, both forest biodiversity and the future welfare of households will suffer. Although villagers realize the environmental and economic implica• tions of accelerated cutting of ironwood, they are practical about the implications of not participating (i.e., an absolute loss of benefits as opposed to the enjoyment of some benefits in the short term). These same scenarios might present themselves were the villagers to consider forming a commercially based coop• erative responsible to a state agency. As it was, the state os• tensibly tried to modernize a traditional institution without realistically addressing the limits or potential of these tradi• tional institutions. The result was an ultimate inability to control adverse impacts of forest exploitation on household welfare and on forest biological diversity. Yet, adapting the traditional management system to the constraints and opportunities of the modern economic sys• tem would not have required much planning effort. Simply recognizing the existence of a system for controlling access to the products within a particular territory, and backing up Extraction Interactions 93 this system in practice and, when necessary, with procedures of conflict management would have gone a long way toward establishing controls. Working with local villagers to ensure size limitations on ironwood trees cut and organizing the sale of ironwood to benefit the whole village rather than in• dividuals would also have been a more effective approach. Outside cutters could have been refused admission to the local forest under all but the most unusual circumstances. Such measures would require an active commitment to joint management of forest lands and the concurrent in• volvement of villagers, the timber company, forestry officials, and subdistrict administrators. Moreover, the notion that HPH and HPHH concessions should not overlap would lose credence as an issue. Each HPH would have to assume that there are large sections of its concession where joint man• agement plans need to be worked out with long-established villages. The ironwood "problem" symbolizes the basic and per• sisting conflict of natural resource management in New Or• der Indonesia and in many other countries. It is often as• sumed that an industrial, bureaucratic, centralized manage• ment style is "modern" and thus good, whereas a local, customary, simple, and practical management style is "primi• tive" and thus bad. A more appropriate approach to forest management will involve some mixing of rights to gain and control forest access. It is highly unlikely that Indonesia will give up exploiting timber in production forests and equally unlikely that all forest villagers are willing to give up their traditional forest uses. Because traditional management sys• tems in many parts of Kalimantan are still vibrant (or re• trievable!, it seems appropriate to suggest a more explicit approach to joint management of these lands. Joint manage• ment requires a more solid legal backing for traditional man• agement institutions as embodied in local customary law. Unfortunately, however, in many areas including Sungai Dalam, it may be too late to "save" the supply of this iron- wood. 94 Society and Non-timber Forest Products in Tropical Asia

A revised version of this paper 5. Peraturan Pemerintah No. 21/ was published as "The ironwood 1970 tentang Hak Pengusahaan problem: (Mis)management and Hutan dan Hak Pemungutan development of an extractive Hasil Hutan (Barber 1989, rainforest product" in Conserva• appended in Zerner 1990, roi, tion Biology 6{i): 210-19, 1992. retranslated by author). Reproduced with permission. 6. Except in one particular village r. The field research on which this in West Kalimantan, where paper is based was sponsored by Christine Padoch (pers. com., the Community Forestry Unit of August 1990) found that villagers the FAO Forestry Department as planted ironwood in their tree part of a larger study on household gardens. and community dependence on 7. A pseudonym for the study forests and trees. I am grateful to village. Ann Hawkins for her comments on an earlier draft of this paper. 8. He remained for three years, having left the year before I came. 2. Although one household turned During his tenure, he encouraged out to be self-sufficient in rice, the people to move out of their thirty- household was considered to be at door longhouse and establish more a disadvantage because the wife "appropriate" single-family had a severe illness, after which a dwellings. He also suggested that large tumor was removed from her families put an ironwood fence stomach. around their yard, because "they 3. Because of this variation, the had so much ironwood in their following remarks apply only to forest that it would be too bad not the two villages studied in West to use it or to sell it all to Kalimantan. For a more detailed outsiders." analysis of the East Kalimantan 9. These arrangements are too "patterns," see Jessup and Peluso numerous.and complex to discuss 1986, especially pp. 517-23; in this paper. Weinstock and Vergara 1987; and Dove r986. ro. These traders included members of the subdistrict police 4. Recent studies in West and military contingents. Kalimantan by Christine Padoch (pers. com.) and me have indicated 11. This is not the case with all the existence of numerous groups traders in non-timber forest of Dayaks who speak mutually products. Some of them have unintelligible languages, have not vested interests in the continued been "named" or described in the availability of forest product ethnographic literature, and call supplies. See Peluso (1983). themselves "Orang Dayak" (Dayak people). We are conducting 12. HPHH for non-timber forest more research on some of these products are still available. groups and their resource 13. Villages are classified by the practices, but preliminary study Department of Village Develop• indicates that these language ment as being at various stages of groups tend to be smaller than the "self-sufficiency"; the highest more well-known Dayak groups degree of self-sufficiency is called such as the Iban, Kenyah, Kayan, swasembada. Busang, and Ot Danum. Extraction Interactions 95

Anderson, James N. 1987. Lands Dunn, F. L. 1975. Rainforest at risk, people at risk: Perspectives collectors and traders: A study of on tropical forest transformation resource utilization in modern in the Philippines. In Lands at risk and ancient Malaya. Monographs in the Third World: Local level of the Malaysian Branch of the perspectives, ed. Peter D. Little Royal Asiatic Society No. 5, Kuala and Michael M. Horowitz, with A. Lumpur. Endre Nyerges, 249-68. Boulder: Westview Press. Gillis, Malcolm, and Robert Repetto. 1988. Public policies and Barber, Charles V. 1989. The state, the misuses of forest resources. the environment, and develop• Cambridge: Cambridge University ment: The genesis and transfor• Press. mation of social forestry policy in Hafner, James A. 1990. Farming New Order Indonesia. Ph.D. diss.. the forest: Managing people and University of California, Berkeley. trees in reserve forests in Bunker, Stephen. 1985. Under- Thailand. In Keepers of the forest: developing the Amazon: The Land management alternatives in failure of the modem state. Southeast Asia, ed. Mark Urbana: University of Illinois Poffenberger. West Hartford, Press. Conn.: Kumarian Press.

Burkill, I. H. 11935] 1966. A Hecht, Susanna, Peter May, and dictionary of the economic Anthony Anderson. 1987. The products of the Malay Peninsula. subsidy from nature. Human 2 vols. Reprint, Kuala Lumpur: Organization 47(1): 15-35. Ministry of Agriculture and Co• Jessup, Timothy C. 1989. Forest operatives. London: Crown Agents exploitation by shifting cultiva• for the Colonies. tors in Borneo. Draft of Ph.D. Colfer, Carol Pierce. 1987. Change diss., Rutgers University, New and indigenous agroforestry in Brunswick, New Jersey. East Kalimantan. In Whose trees} Jessup, T. C, and N. L. Peluso. ed. L. Fortmann and J. W. Bruce, 1986. Minor forest products as 306-9. Boulder: Westview Press. common property resources. In Conklin, Harold. 1957- Hanunoo Proceedings of the Conference on agriculture. Rome: FAO. Common Property Resources, 515-39. Washington, D.C: Dove, Michael R. 1983. Theories National Academy of Sciences. of swidden cultivation and the political economy of Indonesia. Norgaard, Richard B. 1984. Agroforestry Systems 1:85-91. Coevolutionary development potential. Land Economics 60(20): . 1985. Swidden agriculture 60-173- in Indonesia: The subsistence strategies of the Kalimantan Padoch, Christine. 1982. Migra• Kantu'. Berlin: Mouton. tion and its alternatives among the Iban of Sarawak. Instituut . 1986. Plantation develop• voor Tall-, Land-, en Voken- ment in West Kalimantan II: The kunde,Verhandelingen 98, perceptions of the indigenous Marinus Nijhoff, Leiden. population. Borneo Research Bulletin I8|I): 3-26. Peluso, Nancy Lee. 1983. Networking in the commons: A 96 Society and Non-timber Forest Products in Tropical Asia

tragedy for rattan? Indonesia, no. and the misuse of forest resources. 35(April}: 95-108. Washington, D.C: World Resources Institute. . 1986. Rattan industries in East Kalimantan, Indonesia. Paper Schmink, Maryann, and Charles commissioned by FAO, Rome. Wood. 1987. The political ecology of Amazonia. In Lands at risk in —. 1989. The role of non- the Third World: Local level timber forest products in shifting perspectives, ed. Peter D. Little cultivation communities and and Michael M. Horowitz, with A. households: Current knowledge Endre Nyerges. Boulder: Westview and prospects for development. Press. Jakarta: FAO. Weinstock, Joseph A. 1983. . 1991. The rattan trade in Rattan: Ecological balance in a East Kalimantan, Indonesia: Can Borneo rainforest swidden. extraction be reserved? In Non- Economic Botany 37(1): 58-68. timber products from tropical forests: Evaluation of a conser• Weinstock, Joseph A., and vation and development strategy, Napoleon T. Vergara. 1987. Land ed. D. C. Nepstad and or plants: Agricultural tenure in S. Schwartzman, 115-27. agroforestry systems. Economic Advances in Economic Botany No. Botany 41(2): 312-22. 9. Bronx: New York Botanical Garden. Zerner, Charles. 1990. Legal options for the Indonesian Repetto, R. C. t988. The forest for forestry sector. Jakarta: FAO. the trees* Government policies CHAPTER 7

Stephen F. Siebert

Rattan Management for Forest Conservation in Indonesian National Parks and Preserve Buffer Zones

Tropical forests in Southeast Asia are rapidly disappearing under the combined pressure of timber harvesting and forest conversion to plantations and subsistence farms (Denslow and Padoch 1988). Rattans, a large group of climbing palms that comprise one of the most important forest products in Southeast Asia, are also endangered by forest conversion and uncontrolled harvesting; already many rattans are in danger of extinction (de Beer and McDermott 1989; Dransfield 1987]. The socioeconomic importance of rattans is difficult to exaggerate. Rattans have been used for centuries in bind• ing, basket-making, and weaving and are now a major source of cash income (Corner 1966; de Beer and McDermott 1989). The rattan trade generates approximately US$3 billion an• nually (de Beer and McDermott 1989) and employs more than half a million people in collecting, processing, and cottage- scale manufacturing (IDRC 1980). The loss of rattan resources will have many repercus• sions. Rural poverty will be exacerbated because poor people depend on rattan income (Siebert and Belsky. 1985). Forest conversion pressures will increase as landless unemployed are left with no alternative but to cultivate forest land (Denslow and Padoch 1988). Finally, conservationists will be deprived of an economic incentive to maintain tropical for• ests. Conversion of tropical rain forests and decimation of wild rattan can be controlled only to the extent that alterna• tives to forest farming and unregulated rattan collecting are developed. 98 Society and Non-timber Forest Products in Tropical Asia

The Indonesian government has delineated buffer zones around many national parks, and government and non-gov• ernmental conservation organizations are identifying devel• opment activities that are compatible with conservation ob• jectives. Rattan cultivation and management is one poten• tial activity that can be used in forest conservation efforts because of the many traditional uses of rattan and its impor• tance as a source of cash income, and because it thrives in primary and secondary forests. Several rattan species have been successfully cultivated in plantations [Dransfield 1988; Manokaran and Wong 1985) and in swidden fallows (Weinstock 1983). However, the rattan trade relies primarily on the collection of wild canes (Dransfield 1987). Little re• search has been conducted on the ecological characteristics of economically important rattan species in the wild or on the cultivation of rattans in preserve buffer zones. This paper reviews the abundance, site preferences, and cultivation potential of Calamus exilis (rattan) in Kerinci- Seblat National Park in Sumatra, Indonesia. The economic value of C. exilis cane harvesting and recommendations for managed rattan harvesting and forest conservation are also presented.

RESEARCH Research was conducted in Kerinci-Seblat National Park SITE AND (KSNP), Sumatra, a 1.5-million-hectare preserve that ranges METHODOLOGY from 100- to 3,800-meter elevation and includes an extremely diverse assemblage of flora and fauna. Management of KSNP is complicated by the presence of a io-by-80-kilometer hu• man settlement, Kabupaten Kerinci, in the center of the park. Approximately 280,000 people inhabit Kerinci, the economy of which is agriculturally based. Irrigated rice and perennial cash crops, particularly [Cinnamomum burmanii) and coffee (Coffea robusta), are the major products. The pe• rennial crops are grown on hill slopes that have been con• verted from forests. Rattan has been used by local people in Kerinci for gen• erations, and the rattan handicraft and basket-making indus- Rattan Management for Forest Conservation 99 tries are centered in the village of Sungai Tutung (Siebert 1989). Calamus exilis, a small clustering rattan capable of vegetative propagation, is the principal species used. Research on demographic characteristics of C. exilis was conducted in primary forests between 1,200 and 1,400 meters approximately 25 kilometers within KSNP. This area is the major source of cane for the local rattan industry. The forest type is lower montane: upper dipterocarp to oak-laurel (Whitmore 1984). Epiphytes (e.g., Asplenium nidus) are ex• tremely abundant, and the forest is frequently shrouded in mist. Canopy height averages from 25 to 30 meters but is frequently broken by tree fall gaps of varying size. Soils are Tropohumults (Acrisols in the FAO system) with an organic A horizon averaging 30 centimeters thick. Demographic characteristics of C. exilis were deter• mined by sampling rattan plant and cane abundance, and various environmental factors in forty 0.05 hectare plots (a total of 2 hectares were sampled). The plots were selected at random intervals along four transects in the major collecting area. In each plot, the following data were collected:

1. Number of C. exilis seedlings, juvenile plants (i.e., those without harvestable canes), and mature plants (i.e., those with harvestable canes); 2. Number of canes per plant (both harvestable and nonharvestable length; the minimum length harvested was about 5 meters); 3. Length of all harvestable canes,- 4. Predominant light regime, qualitatively scored on a scale of 1 to 4: 1 = complete shade, 2 = intermittent sunflecks, 3 = intermittent sunlight, and 4 = full sun• light; 5. Competition from understory vegetation, qualitatively scored on a scale of 1 to 4: 1 = little growth on the forest floor, 2 = scattered growth of little competitive signifi• cance, 3 = moderate competition from dense understory growth, and 4 = severe competition from complete cover of understory growth; ioo Society and Non-timber Forest Products in Tropical Asia

6. Soil drainage characteristics, classified on a scale of i to 3: i = well drained, 2 = occasional restricted drain• age, and 3 = poorly drained, soils waterlogged.

In cases where sample plots had small gaps and/or only portions of the plot occupied by understory vegetation, the plots were split and information was gathered in the domi• nant portion. In addition, one soil sample, comprised of five subsamples, was collected between o and 15 centimeter depth (the observed rooting depth of C. exilis) in each plot. Nutri• ents were extracted with Morgan's solution (10 percent so• dium acetate in 3 percent acetic acid buffered to pH 4.8 us• ing a 1:5 |by volume] soihsolution ratio). Samples were ana• lyzed for pH (in water); available phosphorus (colorimetri- cally by stannous chloride reduction); potassium, calcium, and magnesium (by atomic absorption); exchange acidity (by titration of barium chloride/triethanolamine solution); and organic matter [by loss on ignition). Soil was analyzed by the Cornell Nutrient Analysis Laboratories in Ithaca, New York. The cultivation potential of C. exilis was evaluated by transplanting twenty seedlings (i.e., vegetative cuttings, each comprised of stock and above-ground shoot] beneath perennial crops (i.e., cinnamon, coffee, and rubber \Hevea brasiliensis]) in five farms. A total of 100 seedlings were trans• planted. The cultivation site was in an area recently cleared from primary forest where soil conditions appeared favorable for rattan growth and where many rattan artisans had estab• lished perennial farms. Seedling survival and growth rates were assessed two and twenty-four months after transplant• ing. Rattan abundance and seedling survival rates were sum• marized using descriptive statistics. Site preferences were evaluated using regression analyses. The value of rattan har• vested was calculated on a per hectare basis using 1990 prices paid to collectors upon delivery of clean (i.e., silica sheaths removed), air-dried cane to rattan artisans. Rattan Management for Forest Conservation 101

RESULTS AND The abundance of C. exilis plants and canes in the principal DISCUSSION rattan-collecting area is summarized in Table 7.1. There was an average of 283 C. exilis plants with 65 mature plants and 1,147 total canes with 191 mature canes per hectare. Cane lengths averaged 10 meters. Thus, an average hectare of for• ests produced approximately 1,190 meters of cane.

SITE PREFERENCES OF CALAMUS EXILIS

Populations of C. exilis showed a strong negative relation• ship with light intensity and competition from understory vegetation. For example, sites with high light intensities were

1 negatively related to the number of mature plants (R = o.6o; F < 0.0011, number of canes (R1 = o.64;'F < 0.001), and number of canes of harvestable length (R1 = 0.49; F < 0.001). Simi• larly, sites with dense understory vegetation were negatively correlated with the number of mature plants (R1 = 0.57; F < 0.001), total number of canes (R1 = 0.59; F < o.ooi), and num• ber of canes of harvestable length (R* =0.4, F < 0.001) (Siebert 1993)- Interestingly, rattan collectors use light regime and understory vegetation as guides to locating rattan. Specifi• cally, they avoid areas with large breaks in the canopy and associate dense growth of Strobilanthes sp., Salacca sp., Pan- danus sp., Zingiberaceae sp,, and other understory vegeta• tion with poor rattan growth and little harvestable cane.

Table 7.1 Frequency of Calamus exilis plants and canes in lower montane forests of Kerinci-Seblat National Park (n - 40)

Mean no./plot (standard deviation) Mean no./ha

C. exilis plants Total 14.2 (10.0) 283.6 Seedlings 7.2 ( 7.4) 143.0 Juvenile 3.8 { 3.4) 75.0 Mature 3.3 ( 3.3) 65.4 C. exilis canes Total 57.1 (50.6) 1,147.4 Harvestable 9.6 (12.9) 191.0 ioi Society and Non-timber Forest Products in Tropical Asia

It is unclear whether excessive light or direct competi• tion accounts for the paucity of rattan in some plots. C. exilis seedlings and juvenile plants were observed in gaps where understory competition was intense. In these sites, C. exilis was overtopped by other vegetation. As canopy gaps close and shade-intolerant competitors die, C. exilis seedlings may resume growth and cane development. Alternatively, exces• sive light and competition may result in high mortality of C. exilis. No significant relationships were observed between C. exilis populations and edaphic conditions, except for soil drainage. Mean soil conditions in the sample plots are sum• marized in Table 7.2. of C. exilis were confined to the highly acidic A horizon. This organic layer is moist but well drained. The only instance where C. exilis populations were observed to vary with soil conditions was in a single plot with waterlogged soils. Collectors reported that C. exilis is never found in poorly drained soils. Observations along the transect supported this view; in poorly drained sites, C. exilis was only observed on raised windthrow mounds.

CULTIVATION TRIALS OF CALAMUS EXILIS

The survival rate of transplanted C. exilis seedlings was ini• tially high; two months after transplanting all plants appeared vigorous and exhibited green shoots. However, twenty-four months after transplanting all seedlings were dead. A num-

Table 7.2 Soil characteristics in lower montane forests of Kerinci Seblat National Park (n - 40)

Variable Mean (standard deviation)

PH 4.7 (0.6) Available phosphorus (mg/kg) 1.4 (0.6) Available potassium (mg/kg) 148.0 (35.8) Available calcium (mg/kg) 1,706.1 (1,981.8) Available magnesium (mg/kg) 182.1 (166.5) Exchange acidity (cmol/kg) 98.8 (18.7) Organic matter, LOI (%) 63.0 (6.9)

cmol - centimoles LOI - loss on ignition Rattan Management for Forest Conservation 103 ber of factors may account for the initial success and subse• quent failure of the transplants. First, seedlings were trans• planted in the middle of the wet season [February). During the June through August dry season, soils are drier in farms than in primary forests. In addition, soil organic layers (A horizons) are less well developed, light intensities greater, and competition from weeds (particularly Imperata cylindrica) more pronounced in farms than in primary for• ests.

ECONOMICS OF MANAGED RATTAN HARVESTING

Rattan collectors reported that C. exilis can be harvested two to three years after cutting and that subsequent production is comparable to initial yields. Marked plants were resurveyed in 1992 to determine actual resprouting and cane-growth potentials. Thirty months after harvesting, C. exilis plants had produced an average of 3.2 (± 5.0) new harvestable canes per plant, and canes were already being harvested by local rattan collectors (Siebert in press). While C. exilis harvesting may be repeatedly harvested, it may not be competitive with alternative land uses, par• ticularly perennial cash crop cultivation. For example, 1,190 meters of C. exilis cane is worth approximately Rp 17,865 or US$10 (assuming 1990 market prices of Rp 4,500 per 100 3- m-length canes). Assuming plants can be harvested every three years, the value of sustained-yield C. exilis harvesting is about US$3/ha per year. In contrast, relay intercropping of coffee and cinnamon earns farmers Rp 2,000,000 or US$ 1,100 per year over ten years (assuming average cinnamon yields of 9 kilograms per tree, 1,000 trees per hectare, Rp 2,000/kg and harvest at age ten years; and average coffee yields of 500 kg/ha for four consecutive years and Rp 1,000/kg) (90 per• cent of the revenue is derived from cinnamon harvest). Both rattan harvesting and cash crop cultivation are widespread and illegal in KSNP. Rattan collectors reported that C. exilis is abundant in many areas that are considered too high, steep, or infertile 104 Society and Non-timber Forest Products in Tropical Asia for coffee or cinnamon cultivation (including those sites in• vestigated in this study). While less valuable than perennial crops, rattan harvesting may be economically attractive on some lands. Rattan also offers many opportunities to increase local employment and income through domestic value-added processing. For example, more than seventy Kerinci families rely on rattan handicraft or basket-making as their primary or sole source of cash income. These benefits were not in• cluded in estimating the economic value of rattan. Further• more, unlike the conversion of forests to farms, C. exilis har• vesting has little or no effect on other forest flora and fauna. Assuming that C. exilis plants are not damaged during cane harvesting, rattan gathering could be conducted sustainably in conjunction with conservation objectives.

IMPLICATIONS FOR FOREST CONSERVATION AND

MANAGEMENT

Forest conversion to perennial cash crops is widespread and uncontrolled throughout KSNP and is primarily the work of young, resource-poor households (i.e., families who lack ac• cess to irrigated rice lands, hillside farms, and alternative income-earning opportunities) (Siebert 1989). Many of these forest farmers are from the rattan village of Sungai Tutung. Informal interviews of these farmers revealed that most would prefer to remain rattan artisans even though it is less remunerative than cinnamon and coffee farming. Many have no alternative but to cultivate cash crops, however, because of a shortage of rattan supplies. Government prohibition against collecting forest prod• ucts within KSNP has partly contributed to the shortage of C. exilis canes. The cultivation of perennial crops is prohib• ited within KSNP as well; however, because forest guards rarely patrol within the park, once cinnamon and coffee reach local markets it is impossible to determine their point of ori• gin. In contrast, rattan is available only from within the park Rattan Management for Forest Conservation 105 and collectors can be apprehended when they attempt to transport or sell rattan canes in the local markets. If C. exilis canes were harvested on a sustained-yield basis, they could provide a source of income for collectors and raw materials for the local rattan industry. Thus, pres• sure to clear forests for perennial crops could be reduced and some local inhabitants provided with an incentive to con• serve forests. Indonesian park officials and non-governmental con• servation organizations should evaluate the feasibility of es• tablishing extractive reserves for managed harvesting of C. exilis canes from KSNP. One means of managing rattan har• vesting sustainably would be to demarcate management units (i.e., subwatershed areas) in the principal collecting areas and open these units to C. exilis harvesting once every three to four years. This approach will require park personnel to pa• trol well within KSNP, rather than remain in the buffer zones around the enclave. More rigorous enforcement efforts throughout KSNP are essential due to the current practice of clearing forests for farms as far as 25 kilometers within the park. Rattan collectors could be enlisted in park patrol ef• forts if they are provided an economic incentive (i.e., rattan collecting) to conserve forests. In addition, the feasibility of paying rattan collectors for patrolling the park should be evaluated. The opportunity exists to expand managed rattan har• vesting to other species in KSNP, such as high-value C. manan, which was formerly abundant (Siebert 1989), and other large diameter canes. The cultivation of economically important rattans in primary forests, where conditions more closely approximate the natural habitat of rattan, also war• rants study One particularly promising approach would be to establish enrichment plantings of desired species where they already occur naturally. Finally, research and develop• ment efforts should focus on increasing the profitability of local industries by improving domestic and international marketing outlets. io6 Society and Non-timber Forest Products in Tropical Asia

CONCLUSION Managed harvesting of C. exilis may provide a means of fos• tering local economic development that is both ecologically sustainable and compatible with forest conservation objec• tives. The long-term viability of KSNP and other national parks in Indonesia will be determined, in part, by the extent to which forest conservation benefits the rural poor who have few alternatives but to clear forests for immediate food and cash income needs. The cultivation and management of rat• tan and other non-timber forest products represent possible means of meeting conservation and development objectives. While non-timber forest products alone are unlikely to pro• vide the economic incentive needed to conserve tropical for• ests [Saw et al. 1991), they represent an important compo• nent in forest conservation strategies and warrant greater research and development consideration.

Acknowledgments—This work would not have been possible without the assistance and knowledge so generously provided by Tapri and Harun of Sungai Tutung. Research funding was provided by grants from the , UNESCO Man and the Biosphere, and the University of Montana. I greatly appreciate the comments and suggestions provided by Jill Belsky, Jeff Fox, and two anonymous reviewers.

REFERENCES Corner, E. J. 1966. The natural The biological aspects of rare history of palms. London: plant conservation, ed. H. Synge. Weidenfeld and Nicholson. London: Wiley and Sons.

de Beer, J. H., and M. J. McDer• . 1988. Prospects for rattan mott. 1989. The economic value cultivation. Advances in of non-timbei forest products in Economic Botany 6:190-200. Southeast Asia. Netherlands 1DRC. 1980. Rattan. A report of i Committee for IUCN, Amster• workshop held in Singapore, 4-6 dam. June 1979. IDRC-isse, Ottawa, Denslow, J., and C. Padoch. 1988. Canada. People of the tropical rain forest. Manokaran, N., and K. Wong. Berkeley: University of California 198s. The silviculture of rattans: Press. An overview with emphasis on Dransfield, J. 1987. The biology of experiences from Malaysia. The Asiatic rattans in relation to the Malaysian Forester 46:298-315. rattan trade and conservation. In Rattan Management for Forest Conservation 107

Saw, L, C, J. V. LaFrankie, K. M. Siebert, S., and J. Belsky. 1985. Kochummen, and S. K. Yap. 1991. Forest-product trade in a lowland Fruit trees in a Malaysian rain Filipino village. Economic Botany forest. Economic Botany 45:120- 30:522-33. 36 Weinstock, J. A. 1983. Rattan: Siebert, S. 1989. The dilemma of a Ecological balance in a Borneo dwindling resource: Rattan in rainforest swidden. Economic Kerinci, Sumatra. Principes 33:79- Botany 37(1): 58-68. 87. Whitmore, T. 1984. Tropical rain . 1993. The abundance and forests of the Far East. 2d ed. site preferences of rattan Oxford: Clarendon Press. {Calamus exilis and Calamus zollingeri) in two Indonesian national parks. Forest Ecology and Management 59:105-13.

. In press. Prospects for sustained-yield harvesting of rattan {Calamus spp.) in two Indonesian national parks. Society and Natural Resources. I. Fiber Grass from Forest Land: A Case from North India

Fiber grasses are important raw materials for India's rural rope-making industries and commercial paper mills. Bhabbar, or sabai grass (Eulaliopsis binata], grows abundantly on for• est lands in eastern India and in the Shivalik Range, an area stretching from Uttar Pradesh state to the Pakistan border at the base of the Himalayas. This paper explores how grass• lands are managed by the Haryana Forest Department (HFD) and how lease-harvesting rights are allocated to contractors and a paper mill. The paper also chronicles the experiences of the Hill Resource Management Societies (HRMS) in pro• tecting and using grass lease lands under a Forest Depart• ment program to transfer non-timber forest product harvest• ing rights to communities to improve forest management. The information presented is drawn from field studies and monitoring reports compiled under the joint forest man• agement support program from 1989 to 1992. The authors participated in numerous discussions with community man• agement groups (HRMS), with senior HFD officials in work• ing group meetings, with field staff of training programs, and with paper mill representatives. While the department's suc• cesses are admirable and their failures understandable, the significance of the HFD program is in the learning that has emerged. The HFD's experience in working with communi• ties to manage reserve forest lands may provide lessons that can help develop new approaches to managing and restoring India's valuable forests and grasslands. no Society and Non-timber Forest Products in Tropical Asia

FIBER-GRASS The heart of the fiber grass production region in north India LEASE is the Shivalik Hills stretching from the Nepal border to the MANAGEMENT Pakistan border along the base of the Himalayas. State For• IN THE HARYANA est Departments control most of this land and logged the SHIVALIKS area extensively from the latter half of the nineteenth cen• tury until the 1960s. The logging activities allowed grasses to grow abundantly. Communities throughout the region harvest bhabbar for rope production. In the Saharanpur dis• trict of northwestern Uttar Pradesh, for example, an estimated 200,000 people derive a major part of their livelihood from grass harvesting and processing. In neighboring Ambala district, the HFD leased approxi• mately 20,000 of 68,000 hectares of forest lands in the Shivalik Hills for bhabbai cutting. When the leasing system was es• tablished four to five decades ago, nominally priced leases were granted to a paper mill. When the mill's lease expired in 1980, HFD initiated an annual open auction for grass-cut• ting leases. Since then prices have risen dramatically, increas• ing from an average Rs 180 to Rs 289 per metric ton. This, however, is still below the open market value of Rs 1,000 to Rs 1,500 per ton. Planners and environmentalists have ex• pressed concern over the sale of forest resources to indus• tries at concessional rates. They argue that large leasehold• ers have neither the economic need to receive government subsidies nor the commitment to improve resource manage• ment. Others argue that forest communities can play an ef• fective role in improving forest lands if given priority access to leases in exchange for community commitments to im• prove protection and management. Over the past decade, the HFD has attempted to over• come legal, procedural, and organizational barriers as well as the vested interests of powerful economic groups in order to devise participatory management systems for reserve forest lands in the Shivalik Hills. The department has struggled to identify appropriate grass and forest management technolo• gies, and to open communication channels with hill com• munities in order to involve them in collaborative efforts to regenerate highly eroded watersheds. The Forest Department Fiber Crass from Forest Land: A Case from North India in

uses fiber-grass leases as incentives for encouraging commu• nity participation.

COMMUNITY The idea for the HFD program of giving local communities BHABBAR priority access to bhabbar leases originated from the suc- LEASE PROGRAM cessful projects in Sukhomajri and Nada watersheds when STRATEGY villagers stopped grazing their cattle and goats in exchange for irrigation water from small earthen dams built by the HFD. Resource managers argued that the increased bhabbar yields generated by improved protection should be given to the community as its rightful share for improved forest man• agement. This gives villages a greater incentive for improv• ing the management and productivity of the local forest. Hill Resource Management Societies were established for this purpose. In the Nada watershed, bhabbar grass productivity in• creased from 36 kg/ha (wet weight) in 1981 to more than 5,000 kg/ha in 1986 while under community protection and through enrichment planting. Similar increases were observed in the nearby watershed of Sukhomajri. Annual fodder grass (as opposed to fiber) production also rose from 35 kg/ha to 600 kg/ha. Soil erosion from the two catchment areas de• creased markedly during this period from 900 to 30 metric tons of soil loss per hectare (Bansal and Mishra 1982). The HFD traditionally sold leases to middlemen, who then sold cutting rights to villagers. The Sukhomajri HRMS asked the HFD to sell them the grass leases directly. Provid• ing leases directly to the local societies was a significant step in transferring some formal management responsibilities to communities for overseeing the use of neighboring govern• ment forest lands. In Sukhomajri, household income in• creased an estimated Rs 1,000 per year due to lower lease costs and increased grass production (Poffenberger 1990, 23). In 1986 the Sukhomajri-Dhamala HRMS paid the Forest De• partment Rs 22,000 for the lease, and in 1987 harvested from their lease area 150 metric tons valued at Rs 120,000—ap• proximately US$ 10,000. in Society and Non-timber Forest Products in Tropical Asia

When the Haryana state government first considered selling grass leases to HRMS, the HFD set the price at the average lease price in the open market over the preceding three years. The village of Sukhomajri, which had already formed an HRMS, received the first lease priced on this ba• sis. The neighboring communities of Logarh and Jattan Majri subsequently purchased leases on the same basis, as did Govindpur, Masoompur, and Raina villages (see Figure 8.1). The HFD also stabilized the prices of fodder and bhabbar grass leases. In 1988, the divisional forest officer of the area issued orders to increase lease rates 10 percent each subse• quent year. Prices on lease agreements with contractors and the mill, however, are calculated differently from those sold to

Figure 8.1 HRMS grass lease areas in 1991 in the Pinjore-Morni Forest Division, Haryana Fiber Crass from Forest Land: A Case from North India 113

the HRMS. The mill is committed to pay a gate price based on each metric ton of bhabbar actually harvested. The price is to remain stable during the first two years, followed by annual price increases of 7.5 percent during the next three years. Furthermore, the mill has no obligation or incentive to assist the HFD in improving forest management or in• creasing grass production. The HFD can, however, penalize the mill for leaving unharvested bhabbar if the yields in a particular year are lower than those in previous years. An HRMS, on the other hand, is required to pay the entire lease rent, regardless of actual production. A reliable system for monitoring annual yields does not exist nor does a mechanism for reducing the lease prices if yields fall due to natural or human factors. During 1989-90, the HFD at• tempted to make leasing terms for HRMS more rational by developing a joint management policy framework. Several meetings with members of HRMS have been held to draft new rules. These policies and the reactions of the mill, the contractors, and the HFD are examined below.

REACTIONS Many Shivalik Hill communities reacted favorably to the OF THE HRMS opportunity to lease grasslands in forest areas. As an increas-

TO NEW POLICIES mg numDer 0f HRMS purchased bhabbar leases, however, pricing policies and payment systems became problematic. For example, Surajpur, a multicaste community of nearly 1,000 households, recently registered an HRMS. The major motivation of the community for taking the grass lease was to curb exploitation of poorer households by a local contrac• tor who sold fodder cutting rights (datij to households at in• flated rates ranging from Rs 650 to Rs 1,300 per season. The society reduced the dati rate to Rs 400 and to Rs 200 in the first and second years, respectively, and gave dati free to wid• ows and poorer households. In 1990, the HFD, despite the paper mill's objection, sold the Surajpur society both fodder and bhabbar grass leases on condition that they not subcontract the fiber-grass lease to a contractor. Later, the HFD learned that the HRMS in- ii4 Society and Non-timber Forest Products in Tropical Asia deed had subcontracted this lease. During a discussion of the matter, Surajpur HRMS leaders explained that they were com• pelled to subcontract because of their inability to raise the entire Rs 68,000 lease fee for advance payment. Villagers re• ported they pleaded with the HFD's range officer to permit them to pay the lease money in installments (as contractors are allowed to) but were refused. Villagers pointed out simi• lar problems faced by neighboring HRMS who had also been forced to resell their leases to private contractors. Villagers were confident that if the HFD agreed to accept lease pay• ments in three installments, they could manage the harvest• ing and marketing themselves. The Banjara Sikh community of Kahinwala, where rope- making is the main source of income generation, provides another example of a community grass leasehold. The com• munity has 140 hectares of common property land (shamlat) from which some of its fiber-grass needs are met. Villagers often bought bhabbar grass from local contractors at inflated prices to meet their remaining raw material requirements. Alternatively, they bid for the bhabbar lease in open auc• tion. The community showed considerable interest when contacted by the HFD about forming a management society and taking the lease. The HFD indicated that in return for the lease the community would have to stop grazing and tree- cutting in the reserve forest. The community quickly drafted a handwritten resolution signed by representatives of all vil• lage households offering to keep livestock out of the forest in exchange for the bhabbar lease on concessional terms. All 46 households in the village reduced their open-grazing ani• mals from 135 goats and 6$ cattle in late r98o, to 55 goats and 26 cattle in early i99r. Villagers used the money from the bhabbar concessions to increase their commercial dairy• ing herd of stall-fed water buffalo from 39 to 91 head. In r989, the Kahinwala HRMS purchased a bhabbar lease for Rs 44,550—approximately US$2,500 (see Figure 8.2). In 1990, the price was raised another 10 percent to Rs 49,000. In addition, the HRMS had to deposit Rs 7,225 as income tax on the lease, later refunded after a tax exemption was granted. Fiber Grass from Forest Land: A Case from North India 115

Forty-one members of the community paid Rs 1,400 each to raise Rs 56,225 for bhabbar cutting rights (dati). Four fami• lies did not participate because they were not involved in rope-making. Each participating family harvested approxi• mately 3 metric tons in 1989, but only 2.7 metric tons in 1990 due to unseasonably abundant rainfall and the spread of Lantana camara, a pernicious woody shrub, in the leasehold. While the Kahinwala HRMS profited in both years from their lease, their leaders were not entirely satisfied with the Forestry Department's behavior. They felt that they had ful-

GRASS LANDS (SHAMLAT)

Figure 8.2 Kahinwala village and the HRMS grass lease area n6 Society and Non-timber Forest Products in Tropical Asia filled the terms of the joint management agreement with the HFD by halting grazing and cutting in the lease area, but objected to the HFD's 10 percent annual increase in lease rates. The HRMS felt the HFD was defaulting on the agree• ment since the villagers understood that the lease rate was to be set at the average price over the past three years leading up to the HRMS purchase with increases limited to 7.5 per• cent annually—as with the mill. Villagers also felt down• ward price adjustments should be made if productivity fell due to natural causes. In 1992, the HFD drafted a new policy that would fix the lease rate on production levels and allow installment payments by the HRMS.

REACTIONS OF THE PAPER MILL

Ballarpur Industry is one of the oldest and largest paper mills in north India. In 1992, the mill required approximately 400,000 metric tons of raw pulp material each year. Roughly one-half of its requirement is for short fiber pulp supplied from eucalyptus trees; the remaining raw materials must be long fiber, for which bhabbar, bamboo, rags, and other grasses are used. The majority of the mill's needs for bhabbar are met from the Shivalik Hill regions of Nepal, Uttar Pradesh state, and the Punjab because of the limited supplies avail• able from Ballarpur's home state of Haryana. Currently the mill purchases approximately 4,500 metric tons of bhabbar each year from Haryana, two-thirds of which comes from the Shivalik Hills. Between i960 and 1980, the paper mill had a long-term bhabbar grass lease for forest areas in the Shivalik Hills of Haryana. During that period, many Indian State Forest De• partments gave extraction leases virtually without cost. In the late 1970s the Ballarpur mill was paying Rs 6.70 per met• ric ton of bhabbar grass. In 1980, when the mill's long-term lease expired, senior forest officers established an open bid• ding system for bhabbar cutting rights. In the first year's bidding, the bhabbar rate rose to Rs 180 per metric ton. Throughout the 1980s, the mill continued to control most of Fiber Crass from Forest Land: A Case from North India 117

Haryana's bhabbar lease compartments either by bidding for individual compartments in open auctions or by purchasing harvesting rights from private contractors. In 1989, the mill signed a new five-year bhabbar lease with the HFD. From 1986 to 1989, seven HRMS purchased leases in Pinjore and Raipur Rani ranges. Initially, the mill did not react to the entry of community grassland management groups. As of 1990, more than 2,100 hectares of forest land, or 56 percent of the total bhabbar lease area in the Pinjore Range, was under HRMS management. When three new HRMS in another forest range requested leases for the 1990- 91 season, the mill's raw material procurement officer com• plained to the HFD chief (1) that personnel planning for bhabbar harvesting and collection by the mill would be up• set if additional areas were given to HRMS; (2) that Haryana state would be deprived of an important raw material if HRMS were to sell the bhabbar from their joint management areas to contractors from the neighboring state of Uttar Pradesh; and (3) that the mill deserves priority access to the bhabbar production of the state because its tax payments make an important contribution to the state's economy and exche• quer. Conflict of interests between the mill and the HRMS also became evident when HFD records revealed that the mill had purchased virtually all of the newly offered 1990-91 fod• der grass lease rights in the Raipur Rani. A mill official was asked what interest the mill had in fodder grasses. He re• sponded that although the mill could not use fodder grass as raw material for making paper, the mill's managers felt that the inexpensive fodder leases were a small price to pay for depriving villagers access to the area and thus facilitating protection of the bhabbar grass within the lease areas. ' The HFD's Joint Forest Management Working Group, which monitors development of the HRMS program, raised these issues in a meeting with a representative of the mill. All of the senior HFD officials present at the meeting unani• mously objected to the mill's purchase of fodder grass leases to facilitate protection of bhabbar. HFD foresters agreed that n8 Society and Non-timber Forest Products in Tropical Asia

the mill could not deprive poor villagers access to an essen• tial subsistence (fodder) to further its own interests. The HFD tentatively decided that beginning in 1991 the mill would not be permitted to purchase fodder grass leases. The HFD encouraged the mill to develop a bhabbar purchasing agreement with all existing and future HRMS, rather than to directly manage protection and harvesting. Such an arrangement could resolve HFD's problem in decid• ing between two claimants to bhabbar leases, while provid• ing the mill with a more reliable source of raw materials. The HFD would receive the benefit of social fencing employed by the villagers, the mill would be ensured access to the raw material it needs, and the villages would benefit from an in• crease in bhabbar production. This arrangement can work only if the mill is willing to pay the HRMS a reasonable price for the bhabbar and if the HRMS are willing to give the mill the first rights of purchase. The mill resists paying open market prices because it has received reduced rates from the HFD in the past. Resistance among commercial forest prod• uct users is likely to continue if Indian Forest Departments give forest communities priority access to low harvesting rates, while removing the possibility of low rates for con• tractors and industries.

REACTIONS OF PRIVATE CONTRACTORS

The system of auctioning annual bhabbar leases to private contractors essentially transfers forest protection responsi• bilities from HFD staff to an individual for the duration of the lease. In the past, the mill and private contractors have relied on policing methods to keep villagers out of forest ar• eas. Contractors, due to their personal interest in maximiz• ing their own financial returns, are more forceful than the mill in keeping villagers out of the forest. In addition, they charge villagers high rates for both fodder and bhabbar grasses. Not unexpectedly, local contractors have attempted to undermine the establishment of joint forest management agreements since the HFD's joint management program tries Fiber Crass from Forest Land: A Case from North India 119

to replace the individualized control the contractors wield by community management. During a joint meeting among representatives of four villages (Prempura, Jattan Majri, Khera, and Basaula) to resolve intervillage conflicts, a contractor prevented a meaningful discussion from taking place by rais• ing the fear of villagers being permanently deprived of their traditional rights in the forest area if a joint forest manage• ment agreement took place. In other cases, private contractors have benefited by obtaining grass leases that were intended for community management from Haryana Resource Management Societ• ies that were not yet able to manage their lease area by them• selves. From 1986 to 1988 in particular, a number of HRMS could not raise enough money to pay the lease fee. They re• lied on local contractors to deposit the lease money on their behalf, thereby allowing the contractors to get the lease at a concessional rate. In such cases the contractors retained most of the profit from bhabbar, giving only a small royalty to the concerned HRMS. Private contractors may be forced out of the market if the capacity of HRMS to harvest and market bhabbar increases in the future. It remains to be seen whether private contractors will attempt to further hinder consolida• tion of the HRMS program.

REACTIONS OF THE HARYANA FOREST DEPARTMENT

When the HFD began leasing bhabbar compartments to HRMS in the mid-1980s, senior forestry officials assumed that participating communities would produce rope [baan], thereby generating employment and income-earning oppor• tunities as had the pilot project in the harijan (scheduled caste) community of Nada. Subsequent monitoring, however, indicated that only the forest enclave communities of Bhanjjara Sikhs were significantly engaged in rope produc• tion. The Gujjar, fat, and upper caste villages had neither the experience nor the desire to make rope. Rather, they were interested in gaining control of the bhabbar lease and using this new asset to generate income. These communities usu- no Society and Non-timber Forest Products in Tropical Asia ally sublet their bhabbar harvesting rights to contractors af• ter permitting their own members to collect small amounts for personal use. Often the community did not even harvest the bhabbar themselves, but subcontracted the work to pri• vate contractors. The HFD is uncomfortable with this situation. Some senior officers contend that many HRMS are simply acting as middlemen between the HFD, the paper mill, and other bhabbar consumers. As a result, few benefits are realized by the villagers. They argue that if the HRMS are going to make a middleman's profit, the HFD could sell the bhabbar di• rectly. Many foresters feel that while it is permissible for villagers to use the forest produce for personal needs and small industries, it is improper for them to manage it commercially as a raw material. Other Forest Department officers believe that communities that intend to sell the raw material should not be given grass leases, and that only personal consump• tion should warrant giving HRMS priority access. Outside resource persons associated with the program countered that joint management should be based on assuring villagers a share of all forest produce in return for their participation in forest management, regardless of what they eventually do with it. Furthermore, they argued that since the HRMS were pay• ing commercial lease rates, often in excess of those paid by the paper mill and private contractors, the FIRMS should have the same management rights as do commercial concerns.

The Forest Department also questioned the value of encouraging HRMS acquisition of bhabbar lease rights if the contractors who sublease the area obtain much of the com• mercial profits. The HFD is concerned that some of the HRMS may only be performing their protection responsibilities for the months for which the lease is valid, just as the contrac• tors do. They note that if the societies are serious about longer term improvement of the joint management area, they should protect it throughout the year. The communities respond that they are often forced to sublease their rights to contractors to be able to buy the lease, and that only if the department allowed them to pay in installments, as the contractors do, Fiber Grass from Forest Land: A Case from North India in

could they escape from depending on contractors and cap• ture the profits desired. Some Forest Department staff, how• ever, fear that the HRMS might default on their installment payments since they have no signed contract or collateral. Since under government law the range officer is personally accountable for unpaid leases, their concern is understand• able. HFD field staff have expressed concern about being held accountable for damage done to forest areas that they are responsible for managing if the HRMS do not honor their protection commitments. Consequently, HFD field staff are often caught between two contradictory sets of orders—one conforming to their traditional policing role and the other demanding them to work supportively with forest commu• nities. From 1989 to 1992, attempts were made to respond to problems such as these by first identifying conflict-generat• ing issues through documentation studies, meetings, and informal discussions. The chief of the HFD issued guidelines to its staff, detailing ways in which they can respond to com• munity needs. For example, in response to the concerns of HRMS forest managers, it would be helpful to issue a com• prehensive memorandum outlining the terms and conditions upon which joint management agreements are based.

FUTURE As the HFD began developing formal lease agreements with DIRECTIONS an increasing number of villages, the transfer of resource rights generated a number of positive and negative reactions from program participants. In this section specific issues emerging from the transformation of bhabbar management systems are discussed and ways in which the department might respond are proposed.

HRMS FEDERATION

A major problem for the HFD and the HRMS is a lack of effective communication channels. In the past the HFD bhabbar lease compartments were leased directly to the pa- 121 Society and Non-timber Forest Products in Tropical Asia per mill and a small number of contractors. The limited num• ber of actors made negotiations relatively easy. Furthermore, lessees, such as the mill and private contractors, who pos• sess greater capital, mobility, and a superior knowledge of HFD procedures are able to interact with the department with relative ease. With eight HRMS now holding leases and many more communities applying, the department must interact with a larger number of participants who, unlike the mill and contractors, have limited mobility and weak organiza• tional capacity, and lack capital and knowledge of HFD pro• cedures. Thus, these participants are less able to interact with the Forest Department. The HFD must now go to participat• ing villages to discuss lease agreements and negotiate directly with villagers, putting greater time demands on its staff. When problems arise with HRMS leaseholders, the department must attempt to address them in the field. As the number of HRMS leaseholders grows, so will demands on staff time. An informal federation of HRMS has been proposed to allow discussion of common problems regularly in open fo• rums. The first attempt to establish dialogue between com• munity management groups was made at a meeting orga• nized by the Surajpur HRMS on 10 March 1991. Seventeen HRMS participated in the gathering. The meeting was held to address common problems confronted by HRMS in man• aging bhabbar grass leases and other forest resources. The following issues were raised and recommendations proposed as possible solutions to existing problems.

FEDERATION RECOMMENDATIONS

Lease Compartment Boundary Setting Some communities find managing or protecting existing lease compartments is too large for them. In some cases, two to four villages share traditional rights in an existing compart• ment and desire subdivision of the area between them to fa• cilitate more equitable sharing of lease costs and benefits. In one case the HFD has changed compartment boundaries with• out making a proportionate reduction in the lease amount Fiber Grass from Forest Land: A Case from North India 113 payable by the HRMS. Occasionally, access to parts of the compartment is through areas falling under the jurisdiction of the Forest Department of the Union Territory of Chandigarh. Sometimes these areas were then leased to other parties who often refused HRMS members the right-of-ways. The HRMS proposed that the HFD work closely with par• ticipating communities to resolve these problems. The HFD may have to subdivide some of the compartments to better correspond to areas that the communities are able to protect and harvest. That determination would best be made jointly by the participating communities and HFD field staff after collaborative field-mapping exercises. Primary consideration in determining the boundaries for bhabbar lease subcom- partments would be the communities' raw material needs, managing and harvesting capacity, and prior use rights, and acceptance by neighboring communities.

Lease Rate Setting HRMS are exempt from open-auction bidding, but their bhabbar lease rate is subject to a 10 percent annual increase. Rapid lease rate increases in 1989-90 and 1990-91 raised con• siderable concern among all HRMS who signed leases, al• though some HRMS had enjoyed a constant rate since the first lease was issued in 1986 to Sukhomajri and Dhamala. While some HRMS were experiencing considerable yield in• creases over previous mill and contractor lease periods be• cause of better protection and reduced theft, rainfall levels and the spread of lantana had a significant negative effect on grass production for other HRMS, sometimes lowering pro• ductivity compared to previous years. Consequently, fixed- rate lease-price increases did not reflect the value of the har• vest. In addition, annual price increases of 10 percent are deemed too arbitrary and too high by the HRMS. A new pric• ing system was proposed in response to these complaints. Under the proposed system, the HRMS would pay a per quin• tal (100 kg) rate for the actual amount harvested. In subse• quent years, the per quintal price may be increased by 2 per• cent annually. HRMS would be charged at only 20 percent of 124 Society and Non-timber Forest Products in Tropical Asia the normal rate for any production increases the HRMS are able to achieve through better protection and enrichment planting. Consequently, 80 percent of the.increased produc• tion would be their share for participating in improved man• agement.

Lease Payment Scheduling Under the new program, HRMS have not been allowed to make their lease payments in installments, unlike contrac• tors who are allowed three installments. Initial lump-sum payments of the entire amount put considerable pressure on the communities' ability to raise Rs 20,000 to Rs 65,000 (US$i,100 to US$3,400). In some cases communities were unable to generate such a large sum and lost the lease to contractors. The HRMS recommended that in subsequent years they be allowed to pay for their leases in three equal installments after an initial down payment of 10 percent. The HRMS would have to sign a legal contract with the HFD to qualify for an installment payment plan.

Income Tax Exemption For the past few years, all HRMS have been subjected to a 15 percent central government income tax on their lease price. This has raised the cost of the bhabbar lease even higher. Nonprofit societies may seek income-tax exemption under the Indian tax law. Furthermore, commercial raw materials used directly for processing are tax exempt. Therefore, HRMS using bhabbar grass for making rope could also be eligible for tax exemptions. The HRMS proposed establishing their accounting and profit distribution systems to support low- cost grass prices for members and community-level develop• ment activities, freeing them from further tax liabilities un• der the nonprofit societies clause. Rope-producing commu• nities can also claim exemption as raw material processors. The proposed HRMS Federation could facilitate the process• ing of tax-exemption claims. Pending receipt of tax-exemp• tion status by the HRMS, the HFD agreed to reduce per quin- Fiber Grass from Forest Land: A Case from North India 115 tal rates to Rs 24, or 15 percent below the rate given to the paper mill.

Enhancing Bhabbar Production While bhabbar productivity is determined in part by rainfall patterns, a number of strategies to enhance output are within the reach of the HFD and HRMS. Many HRMS have effec• tively controlled grazing in their lease compartments by sell• ing goats and low-grade cattle, while shifting to stall-feeding their remaining animals. HRMS should meet with neighbor• ing villages, whose grazing animals encroach on the lease area, to reach agreements on closing lease areas to grazing. Insufficient fodder availability in some communities has forced families to harvest new bhabbar shoots for fodder. The more valuable bhabbar fiber grass could be maintained for rope-making and paper pulp sales through supplemental fodder grass and leaf production on private, common, and forest land. Microplanning exercises could help HRMS mem• bers to estimate their fodder needs and plan supplemental planting schemes. Some HRMS have complained that the Kandi Water• shed Project funded by the World Bank is negatively affect• ing bhabbar production in their lease areas. They are par• ticularly resentful that the project is planting two grasses— dudonea and vetiver (Vetiveria zizanioides). Besides not be• ing a local species, vetiver is of no value to the HRMS and dudonea is naturally abundant in the forest. Kandi project authorities, HFD field staff, and local HRMS held joint meet• ings to discuss how community and Kandi project objectives might be made complementary. Joint microplanning exer• cises should be organized in each area where Kandi project and community management groups have interests. Priority should be placed on reaching agreements about the physical and vegetative soil conservation measures. HRMS members should have a prominent role in determining the species to be planted in their lease area. Local resource planning activi• ties should make use of a microplanning option menu (de- n6 Society and Non-timber Forest Products in Tropical Asia veloped by the Joint Management Support Team| to help vil• lagers and field staff identify desirable inputs. Lease prices should be correspondingly lowered wherever Kandi project activities reduce bhabbar yields. A number of HRMS are also concerned over the expan• sion of Lantana camara within their lease area. Lantana shades out bhabbar grass, reducing its production. Some com• munities have asked HFD to reduce their lease payments in exchange for removing lantana from their lease area. Forest• ers have noted that lantana will die off if broadleaf trees are planted over it; however, this would also reduce grass pro• duction. Throughout the program area, the regrowth of na- tivekhairtrees {Acacia catechu), while encouraging, has also begun to reduce bhabbar productivity. Succession patterns of forest regeneration and sustained yield resource manage• ment approaches in the Shivaliks require further study. Enrichment planting of bhabbar in open forest lands has been highly successful in a number of sites. Many HRMS have expressed an interest in bhabbar enrichment plantings on their lease area. Decentralized nurseries and transplant• ing systems for grass slips should be established with appro• priate incentives and extension support from the HFD. HRMS members are also interested in gaining access to the leaves of chhal trees [Anogeissus latifolia), an excellent leaf fodder. Microplanning would assist HRMS members and HFD staff in identifying optimal sites for planting and protection to enhance the productivity of fodder grasses, trees, and fiber grasses.

Improving Processing, Transport, and Marketing Systems Many HRJVIS feel that they could manage harvesting, labor, and transport arrangements if they were able to pay the leases in installments and if they could gain rapid approval from the HFD to begin cutting before grasses dried and lost weight (and, therefore, value). Communities could receive assistance in contracting trucks and arranging mill deliveries through the proposed HRJVIS Federation. The HFD also should assist HRMS in organizing direct harvesting and marketing arrange- Fiber Grass from Forest Land: A Case from North India 127

ments. Political disturbances in Punjab, a rope-making area, have disrupted rope distribution channels to larger markets, flooding local rope markets and forcing prices down. The HFD and the proposed federation should begin developing new mar• ket channels to allow rope-producing HRMS access to larger, more distant urban markets with higher prices.

SUMMARY This paper reviews the experiences of the Haryana Forest Department in transferring management and exploitation rights to an important non-timber forest product [bhabbar grass) from large industries and wealthy middlemen to forest communities. While some Indian State Forest Departments give usufructs to villagers for non-timber forest products from public lands, the HFD is perhaps the first such agency to help communities organize and to give them priority access to forest-harvesting leases. The program has effectively re• duced tensions, improved communication, and provided a framework for collaborative protection and production ac• tivities between foresters and forest communities. Commu• nities, attracted by the opportunity to gain new rights to for• est resources, have also shown a willingness to accept new management responsibilities. Grazing pressures have been significantly reduced or eliminated in many participating communities, as goats and cattle are sold and water buffalo purchased for commercial dairying are stall-fed. Furthermore, natural regeneration has accelerated and soil erosion has di• minished as the cutting of trees and shrubs has declined. At the same time, many problems remain. While some participating communities have gained secure access to fod• der and fiber grasses at considerably reduced costs, others have legitimate complaints about price setting, boundary problems, and payment requirements. The Haryana Resource Management Societies' fiber-grass lease agreements were undermined by irregular price-setting procedures, combined with changing yield levels and fluctuating or falling market prices. In many cases, HRMS, with insufficient capital to buy leases, were forced to borrow from contractors, creating n8 Society and Non-timber Forest Products in Tropical Asia

"shadow leaseholders." Finally, the HRMS had difficulties effectively marketing grass to the mill, which is also con• cerned with the loss of direct control over grass lease com• partments. These problems stem from the HFD's limited capacity to monitor bhabbar productivity and set prices accordingly, and to resistance from mid-level HFD staff against allowing HRMS to pay for leases in installments. HRMS had a limited ability to respond to marketing problems and opportunities, while the mill felt threatened and desired continued access to subsidized raw materials. The program needs to be strengthened in the coming years through better coordina• tion among the HRJVIS, improved communication of poli• cies and procedures within the HFD, and a clear marketing agreement with the mill. Forestry staff must develop stable relations with management groups since nearly one-half of the Pinjore Range bhabbar lease compartments are already under HRMS management. The rapid rate of transfer among range and beat officers should be reduced and systematic staff reorientation programs intensified. Community grass leasing arrangements should be in• tegrated with long-term resource management requirements of the larger forest ecosystem. As tree cover increases and forest succession proceeds, grass production will gradually fall and communities will require continued benefits through rights to other forest products. While it will take a number of years to evolve new policies and procedures to support the functioning of the HRMS, the program has identified many problematic issues and has begun to address them. The ex• perience of the HFD Joint Forest Management Program is significant as HFD has encountered the types of problems other Forest Departments and communities will likely con• front as they attempt to develop collaborative forest man• agement systems to enhance the productivity and profitabil• ity of non-timber forest products for rural families. This shift to collaborative management will continue to require major changes in Forest Department procedures and working plans. Fiber Grass from Forest Land: A Case from North India 119

Note: From Society and Natural Resources, Vol. 8, pp. 219- 30, 1995, Taylor & Francis, Inc., Washington, D.C. Repro• duced with permission. All rights reserved.

Bansal, R. C, and P. R. Mishra. Poffenberger, Mark. 1990. Joint 1982. Sedimentation of Sukuna management of forest lands: Lake, Chandigarh: Status report Experiences from South Asia. A 1982. Central Soil and Water and Ford Foundation program Training Institute, Research statement. New Delhi: The Ford Center, Chandigarh. Foundation. I

I

I Conclusions

Forest ecosystems are the source of many goods and services that societies both now and in the past have found very use• ful, if not essential. Produce from the forest not only meets many basic needs for people living in or near the forest, most especially in developing societies, but also may be impor• tant in economic terms, entering local, regional, and some• times international trade. The differential distribution of mineral and biological elements in the landscape, for a large part originally forested, has formed the foundation for the economic, social, and cultural development of the different groups worldwide and provided the basis of interaction be• tween social groups since earliest times (Ellen 1979; Wolf 1990). Southeast Asia in particular, as a region of great bio• logical diversity, has been the source of numerous products that have been important trade items for several centuries (Marsden [1811] 1966; Watt [1908) 1966; Burkill [1935] 1966; Viraphol 1977; van Goor et al. 1982). Indeed, the pursuit of such products by various societies has influenced historical patterns of exploration and settlement in many regions.

Until recently all forest products other than timber have been known collectively as "minor forest products," a phrase used as much as a consequence of their perceived stature in fiscal as well as physical terms. The increased demand for construction timber both during and after World War II, rein• forced by the forestry profession's historical focus on wood, plus undoubtedly the sheer size of the material, associated machinery, and infrastructure required for timber extraction (and the correspondingly large amounts of money exchang• ing hands in the process), has served to concentrate atten- 132 Society and Non-timber Forest Products in Tropical Asia tion in recent decades almost solely on the timber element of the forest ecosystem. By comparison the value, not to mention importance, of the great variety of non-timber products extracted from the forest, either in commercial trade or traditional use, has been difficult to assess. Statistics on non-timber forest prod• ucts (NTFPs), if available, are by all accounts generally in• complete, inaccurate, and inconsistent over time (Saulei and Aruga 1994; Yonzon 1993; Aryal 1993; Edwards and Bowen 1993; FAO 1991; Peters, Gentry, and Mendelsohn 1989; de Beer and McDermott 1989). Many NTFPs, however, neither enter into commercial trade nor are recognized by forest management authorities, in which case very little is known about their extraction and use. Consequently, the true ex• tent of the role these products play, in cultural or welfare terms, in the societies that use them often is not fully appre• ciated. The persistent and increasing demand, both regionally and internationally, for NTFPs over the past several years has been highlighted by several reports (Aryal 1993; FAO 1991; de Beer and McDermott 1989). Correspondingly, more devel• oping country governments are becoming interested in pro• moting these products as the value of NTFP exports has in• creased (FAO 1993). Worldwide, tropical forest products have generated increased interest in the past few years due to the great and growing concern over the declining area of tropical forests. Well-intentioned but short-sighted government poli• cies to promote economic development through the liquida• tion of timber resources and the conversion of forest areas to alternative land uses have resulted in the destruction of ex• tensive areas of tropical forest (Repetto 1988). Meanwhile, local and indigenous people in the surrounding areas often have found their livelihoods threatened and their lives im• poverished with the loss of forest resources. The potential for developing NTFP production as an alternative to logging and land-use conversion has been much heralded of late in both technical literature and the popular press (Peters 1992; de Beer and McDermott 1989; Nepstad Conclusions 133 and Schwartzman 1992; Malhotra, Deb, and Dutta 1991). "Extractive reserves," a term variously defined but generally referring to areas managed for the removal of forest produce, excluding timber, while preserving the traditional vegetation, are being promoted as buffer zones for national parks and other conservation areas, and as a potentially very profitable alternative to timber management (Peters, Gentry, and Mendelsohn 1989). Developing markets for forest produce and traditional products, the raw materials for which would be collected in the natural forest, is seen as a means to in• crease the sustainable income-earning capability of forested lands, especially for forest dwellers and forest fringe com• munities. Fundamental to this approach is the argument that the best, and maybe the only, way to preserve the tropical forest is to develop a more broad-based appreciation for the diverse resources contained therein.

The management of tropical forest for a wider variety of goods and services, especially NTFPs, seems to be an al• ternative that would meet the demands of many groups con• cerned about the future of the tropical forest. The develop• ment of NTFPs, it is argued, could provide not only employ• ment and cash income for local, including indigenous and tribal, people of forested regions, but also generate substan• tial revenues for government coffers. In drawing on local people's knowledge as well as resources, NTFP development is also seen as a means to conserve not only the biological but cultural heritage of tropical regions. In many cases, NTFP extraction may be compatible with limited timber harvesting and the development of small- scale wood products industries, and indeed may serve to miti• gate some of the cyclical effects of timber production in for• est communities. Preserving the structural integrity of the forest ecosystem would support the maintenance of biodiversity as well as the environmental or physical ben• efits of the forest. Moreover, protection of the amenity and wildlife aspects of the forest could provide the basis for de• veloping nature-oriented tourism. Finally, protecting the physical resources, which are so closely tied to the cultural 134 Society and Non-timber Forest Products in Tropical Asia

traditions of people living in these forest areas, is important for safeguarding the traditional knowledge of indigenous for• est peoples. This knowledge may prove to be the key for suc• cessful management and sustainable use of the immense biodiversity that the tropical forests contain. The development of the NTFP sector to these ends, though admittedly a tall order technically, would probably prove as much if not more of a challenge in social terms. Much may be learned, however, from past and present pat• terns of exploitation of NTFPs. The purpose of these papers then has been to present case studies that illustrate varying patterns of NTFP exploitation, to discuss the pressures ef• fecting changes in these patterns, and to analyze the impacts on the associated communities, both in terms of internal as well as external relationships. The papers gathered here ex• plore diverse aspects of the management and extraction of various NTFPs in several communities in South and South• east Asia. They address issues of collection, processing, and marketing of various non-timber forest products as well as the policies and institutions influencing the management of the forest resource and the distribution of non-timber forest produce.

VARIETY AS A The wide variety of NTFPs that are, have been, or could be HALLMARK—IN exploited is much too numerous even to list in this short PEOPLE, PLACES. piece (cf. Watt (1908) 1966; Burkill (1935) 1966; FAO 1991). PRODUCTS The several products discussed in the papers, including cam• OVER TIME phor, rattan, rubber, and medicinal herbs, all have a long his• tory of exploitation in the Asia region. Individual preference, as well as cultural tradition, influences the type and level of resource use in different areas. Similarly the degree and re• finement of processing varies throughout the region. As Brosius concludes, regional variation in items exploited may be attributed to the differential distribution of resources, popu• lation, and skills among the different forest communities. The variety in products is nearly matched by the vari• ety of cultures and social groups that use or trade NTFPs, Conclusions 135 for, as noted, forest produce has figured significantly in the economic development of nearly all countries. A wide vari• ety of people across the socioeconomic spectrum may be in• volved in exploiting NTFPs. Undoubtedly the largest group, in number though not necessarily in terms of financial re• ward, would be the collectors—those people who actually live in or on the edge of the forest (e.g., Morris 1982; Dunn 1975). Although some collectors appear to specialize in cer• tain forest products, the majority seem to be involved in col• lecting NTFPs only on a part-time basis (de Beer and McDermott 1989; Wolf 1990J. People who benefit from ex• ploiting NTFPs include not only the collectors, but trans• porters, traders, processors, vendors, and exporters. All authors stress the continuing importance to com• munity welfare of NTFPs that serve a variety of needs, from providing life's daily necessities—food, fuelwood, fiber, me• dicinal and culinary herbs—to goods that can be exchanged for cash. The gathering of forest products persists and co• exists with modern agriculture often because neither the type nor the quantity of food collected from the forest can be ob• tained in local markets, especially at prices rural people can afford. Undoubtedly familiarity and ease of access, as well as the low cost, strongly influence continued use of such tradi• tional items. The increasing commercialization of the rural economy, plus greater access to consumer goods, increases the demand for cash by rural people whose own landhold- ings are often only sufficient to provide meager subsistence rations of food. As Rajan indicates, in such cases the pres• sure on neighboring forest resources is particularly intense. Management strategies are also seen to vary as differ• ent groups exploit the resources at their disposal with differ• ent objectives. Overcuttingof resources either by local people for consumption or exchange or by outside interests contin• ues to threaten some species with extinction in certain areas (Browder 1992). As Siebert notes, overexploitation in the past has virtually eliminated the rhinoceros from Sumatra. The wide diversity of produce at the disposal of many groups, some of which may not be used by the group itself, plus in- 136 Society and Non-timber Forest Products in Tropical Asia

adequate understanding of the regenerative capability of vari• ous resources and the desire to profit to the fullest from the present and possibly fleeting market advantage, may explain in part the seemingly indifferent attitude of some forest peoples to overharvesting. As Rajan points out, the extrac• tion of forest resources by outsiders, including the govern• ment, can become a critical issue for local people to the ex• tent that the natural forest, a source of cash income and of products traditionally used to meet basic needs, is severely depleted, significantly altered ecologically, or converted to an alternative use. One of the significant attractions of NTFPs to rural peoples has been not only the diversity, but the flexibility and complementarity of NTFP exploitation with a variety of traditional agricultural systems as noted by Dove and Padoch. Indeed, adaptability has become characteristic of this sector as communities have adjusted to conditions of emerging and eclipsing markets in local, regional, and international trade.

CHANGE AS A Confirming the historical importance of NTFPs, authors here CONSTANT- also recognized an evolution in the social systems examined ADAPTABILITY in response to various internal pressures and outside stimuli OF SOCIETY associated with NTFP exploitation. As noted by Brosius, AND RESOURCES Padoch, and Peluso, changes in patterns of exploitation of certain NTFPs have implications for the organization of eco• nomic activities and the division of labor within a group that may well reverberate through the social fabric of the whole society. The traditional association of NTFPs with certain socioeconomic strata means that the development of such products can have significant implications for inter- and intra- society relationships based on gender, caste, or ethnicity. Expansion of the cash economy, development of the infrastructure, and extension of government institutions into the hinterlands have brought about increased exploitation of forest resources, in many ways undermining traditional au• thority and social control of forest use. For instance, with the expansion and development of transportation and com- Conclusions 137 munications infrastructure, collectors in the hinterlands may- find they no longer have to depend on middlemen. As Brosius discovered, although many of the patron-client relationships developed in the trade of NTFPs between people of the inte• rior and people on the coast still exist, these historical rela• tionships are now under severe threat. The implications of such change in relations within groups and between groups in different areas may be signifi• cant not only for the sustainable management of the primary resource, and thus the product in question, but also for other products from this ecosystem and perhaps the welfare of the community as a whole. To the extent that the promotion and development of a certain product or group of products may affect access to income-earning opportunities, it may alter historical patterns of wealth distribution and thus call into question traditional social and political alliances. As Dove illustrates, the development of a particular resource may become so influential as to dominate the relationship between the periphery and the center (i.e., national govern• ment). If a particular product becomes increasingly impor• tant in international markets and attracts outside interests, government authorities may move to regulate or even appro• priate the resource (cf. Dove 1993). The alienation, de jure if not de facto, of traditional users from their commonly rec• ognized resource base has seldom protected that resource from undue exploitation. As noted in Rajan, Fox, and Poffenberger and Sarin, the ultimate consequence of such institutional appropriation has been to come virtually full circle with the realization that local people must be involved in managing the resource if it is to survive and indeed thrive.

The introduction of new techniques and new technolo• gies into rural areas may have effects on forest product ex• traction that cannot be fully anticipated (Xu, Pei, and Chen 1993). The influence of the extension of roads and other com• munications facilities in stimulating the collection of NTFPs is noted by several authors. With improved harvesting equip• ment and enhanced infrastructure, plus the increase in popu• lation in formerly remote areas, the danger of overexploitation 138 Society and Non-timber Forest Products in Tropical Asia increases. This tendency may be further exacerbated if forest communities perceive that they are not participating in the benefits of national development. The overexploitation and subsequent extreme deterio• ration of the forest resource in the past, for whatever rea• sons, may force some forest-dependent communities into a redefinition of institutional arrangements for managing the resource. Thus we see that change in resource conditions, especially with regard to those NTFPs addressing basic needs or supporting agricultural production, has been important in forcing societies to face the issues of forest depletion and environmental degradation in many areas. Fox, Peluso, and Poffenberger and Sarin show that such pressure has been in• strumental in bringing about the reorganization of social in• stitutions for more effective management of local resources, resulting in both healthier forests and increased welfare of local communities.

Several authors have pointed out that among some groups when certain NTFPs become scarce in the natural environment and demand for the products increases, collec• tors tend to move seedlings of these species found in the for• est into private gardens, multicrop orchards, swidden plots, or fallow fields, and thus into semi-cultivation (see Xu and Ruscoe 1993; Homma 1992; Ellen 1979; Harris 1973). Thus, as Padoch illustrates, the line between "forestry" and "agri• culture" is increasingly recognized to be hazy if not some• what artificial, especially in the tropics. Researchers have shown that many so-called "virgin" forests are far from pris• tine and probably have been subjected to exploitation for cen• turies. Reportedly some of the first European botanical ex• plorers to visit the tropics could not distinguish between natural and indigenously managed forests (Anderson 1969). Learning more about the indigenous technical knowledge regarding such ecosystem manipulation should have high priority in research concerned with NTFP development. Changes in management and institutional arrange• ments are inevitable, if not essential, for developing the NTFP sector. As Padoch illustrates, some revision in the develop- Conclusions 141

products from tropical forests: FAO. 1991. Non-wood forest Evaluation of a conservation and products: The way ahead. FAO development strategy, ed. D. C. Forestry Paper No. 97. Rome. Nepstad and S. Schwartzman, 33- 41. Advances in Economic Botany . 1993. 1961-1991 . . . 20JO: No. 9. Bronx: New York Botanical Forestry statistics today for Garden. tomorrow. Rome. Harris, D. R. 1973- The prehistory Burkill, I. H. [193s] 1966. A of tropical agriculture: An ethno- dictionary of the economic ecological model. In The products of the Malay Peninsula. explanation of culture change, ed. 2 vols. Reprint, Kuala Lumpur: C. Renfrew, 391-418. London: Ministry of Agriculture and Co• Duckworth. operatives. Homma, A. K. O. 1992. The de Beer, J. H., and M. J. McDer• dynamics of extraction in mott. 1989. The economic value Amazonia: A historical perspec• of non-timber forest products in tive. In Non-timber products from Southeast Asia. Netherlands tropical forests: Evaluation of a Committee for IUCN, Amster• conservation and development dam. strategy, ed. D. C. Nepstad and S. Dove, M. R. 1993. A revisionist Schwartzman, 23-32. Advances in view of tropical deforestation and Economic Botany No. 9. Bronx: development. Environmental New York Botanical Garden. Conservation 20(1): 17-24, 56- Malhotra, K. C, D. Deb, and M. Dunn, F. L. 1975. Rainforest Dutta. 1991. Role of non-timber collectors and traders: A study of forest products in a village resource utilization in modern economy: A household survey in and ancient Malaya. Monographs the Jamboni Range, Midnapore of the Malaysian Branch of the District, West Bengal. Working Royal Asiatic Society No. 5, Kuala paper, Indian Institute of Bio- Lumpur. social Research and Development, Calcutta. Edwards, D. M., and M. R. Bowen, eds. 1993. Focus on jaributi. Marsden, W. [1811] 1966. History Proceedings of the Non-timber of Sumatra. Reprint of 3d ed. Forest Products Seminar, 12 May, Kuala Lumpur: Oxford University Kathmandu. Forest Research and Press. Survey Centre Occasional Paper 2/93 (September|, Ministry of Morris, Brian. 1982. Forest Forests and Soil Conservation, traders: A socio-economic study Kathmandu. of the Hill Pandaram. London School of Economics, Monographs Ellen, R. F. 1979. Sago subsistence on Social Anthropology No. 55. and the trade in : A New Jersey: Athlone Press. provisional model of ecological Nepstad, D. C, and S. Schwartz• succession and imbalance in man, eds. 1992. Non-timber Moluccan history. In Social and products from tropical forests: ecological systems, ed. R. F. Ellen Evaluation of a conservation and and P. C. Burham, 43-74. ASA development strategy. Advances Monograph No. 18. London: in Economic Botany No. 9. Bronx: Academic Press. New York Botanical Garden. 141 Society and Non-timber Forest Products in Tropical Asia

Peters, C. M. 1992. The ecology Watt, Sir George. [1908I 1966. The and economics of oligarchic commercial products of India. forests. In Non-timber products Reprint, New Delhi: Today and from tropical forests: Evaluation Tomorrow's Printers and of a conservation and develop• Publishers. ment strategy, ed. D. C. Nepstad and S. Schwartzman, 15-22. Wolf, E. R. 1990. Europe and the Advances in Economic Botany No. people without history. Berkeley: 9. Bronx: New York Botanical University of California Press. Garden. Xu, J., and M. T. Ruscoe. 1993. The use of indigenous knowledge Peters, C. M., H. Gentry, and R. in agroecosystem management for O. Mendelsohn. 1989. Valuation of biodiversity conservation: A case an Amazonian rainforest. Nature study in Baraugay Haliap, 339(June|: 655-56. Kaiangan, Ifugao, Central Repetto, R. C. 1988. The forest for Cordillera, Philippines. Institute the trees1 Government policies of Environmental Science and and the misuse of forest resources. Management, University of the Washington, D.C.: World Philippines, Los Baflos. Resources Institute. Xu, J. C, S. J. Pei, and S. Chen. Saulei, S. M., and J. A. Aruga. 1993. From subsistence to market- 1994. The status and prospects of oriented system and the impacts non-timber forest products on agroecosystem biodiversity. In development in Papua New Proceedings, Southeast Asian Guinea. Commonwealth Forestry Universities Agroecosystem Review 73(2): 97-t<>S- Network Symposium on Biodiversity, May, Bandung, van Goor, C. P., et al. 1981. Indonesia. Indonesian forestry abstracts: Dutch literature until about 1960. Yonzon, P. 1993. Jaributi exploita• Wageningen: Centre for Agricul• tion from Langtang National Park. tural Publishing and Documenta• In Focus on jaributi, ed. D. M. tion. Edwards and M. R. Bowen, 10-11. Proceedings of the Non-timber Viraphol, S. 1977. Tribute and Forest Products Seminar, 12 May, profit: Sino-Siamese trade, i6$2- Kathmandu. Forest Research and /S53. Harvard East Asian Survey Centre Occasional Paper Monograph No. 76. Cambridge: 2/93 (September), Ministry of Harvard University Press. Forests and Soil Conservation, Kathmandu. CONTRIBUTORS

J. Peter Brosius Assistant Professor Department of Anthropology University of Georgia Athens, Georgia 30602-1619

D. G. Donovan Visiting Fellow Program on Environment East-West Center 1777 East-West Road Honolulu, Hawaii 96848

Michael R. Dove Senior Fellow Program on Environment East-West Center 1777 East-West Road Honolulu, Hawaii 96848

Jefferson Fox Fellow Program on Environment East-West Center 1777 East-West Road Honolulu, Hawaii 96848

Christine Padoch Senior Scientist The New York Botanical Garden 200 Street and Southern Blvd. Bronx, New York 10458-5126 144 Society and Non-timber Forest.Products in.Tropical Asia

Nancy Lee Peluso Associate Professor of Resource Policy Yale University 205 Prospect Street New Haven, Connecticut 06511-2189

Mark Poffenberger Visiting Fellow Program on Environment East-West Center 1777 East-West Road Honolulu, Hawaii 96848

Rashmi Pachauri Rajan Private Consultant Bangalore, India

Madhu Sarin Society for the Promotion of Wasteland Development 1 Copernicus Marg New Delhi 10001, India

Stephen F. Siebert Assistant Professor School of Forestry University of Montana Missoula, Montana 59812-1063

RECENT AVAILABLE EAST-WEST CENTER OCCASIONAL PAPERS: ENVIRONMENT SERIES

The following publications can be ordered from the East-West Center Publication Sales Office (see address on inside front cover). Price per copy is $7, plus shipping. 18 Dam lies and other statistics: Taking the measure of irrigation in China, 1931-91,. by James E. Nickum. January 1995, xx, 129 pp. ISBN 0-86638-168-6.

17 Urban harvest: Recycling as a peasant industry in northern Vietnam, by Michael R. DiGregorio. September 1994, xx, 212 pp. ISBN 0-86638-164-3. 16 Legal frameworks for forest management in Asia: Case studies of community/state relations, ed. Jefferson Fox with assistance from Owen Lynch, Mark Zimsky, Erin Moore, and Helen Takeuchi. 1993/ 100 pp ISBN 0-86638-158-9. r5 Too many people, too little land: The human ecology of a wet rice- growing village in the Red River Delta of Vietnam, ed. Le Trong Cue and A. Terry Ram bo, with assistance from Kathleen Gillogly. 1993, xx, 207 pp. ISBN 0-86638-157-0.

14 South Asian seas: A review of the oceanography, resources, and environment,.by Philomene A. Verlaan. 1991, x, 54 pp. ISBN 0-86638-147-3 13 Keimami sa vakila na liga ni Kalou (Feeling the hand of God): Human and nonhuman impacts on Pacific Island environments, by Patrick D. Nunn. Revised edition, 1992, vui, 69 pp. ISBN 0-86638- 154-6. 12 Agroecosystems of the Midlands of northern Vietnam: A report on a preliminary human ecology field study of three districts in Vinh Phu Province, ed. Le Trong Cue, Kathleen Gillogly, and A. Terry Rambo. 1990, xiv, 232 pp. ISBN 0-86638-134-1. (Also in Viet• namese; translated by Nghiem PhuongTuyen and Vu Minh Hoa.)

11 Conservation or. conversion of mangroves in Fiji: An ecological economic analysis, by Padma Narsey Lai. 1990, xu, 108 pp. ISBN 0-86638-131-5. 10 International conference on the Sea of Japan: Transnational ocean resource management issues and options for cooperation, ed./ authored by Mark J. Valencia. 1989, xiv, 239 pp. ISBN 0-86638- 116-3. 8 Water in Nepal: An interdisciplinary look at resource uncertainties, evolving problems, and future prospects, by Dipak Gyawali. ^89, xu, 126 pp. 6 Forestry research capacity in the Asia-Pacific region: An evaluation model and preliminary assessment, by David N. Bengston, Hans M. Gregersen, Allen L. Lundgren, and Lawrence S. Hamilton. 1988, x, 96 pp.