Proceedings of the Society for Range Management International Affairs Committee Symposium February 17, 2002, Kansas City, Missouri, USA

RANGELAND PROFESSIONALS AND POLICY: Prospects for Effective Influence in the Developed and Developing Worlds

Symposium sponsored by: U.S. Department of Interior, Bureau of Land Management Dow AgroSciences LLC

Proceedings edited by: María E. Fernández-Giménez, School of Renewable Natural Resources, 325 BioSciences East, University of Arizona, Tucson, AZ 85721

Urs P. Kreuter, Department of Rangeland Ecology and Management, Texas A&M University, College Station, TX 77843-2126

TABLE OF CONTENTS

RANGELAND PROFESSIONALS AND POLICY: PROSPECTS FOR EFFECTIVE INFLUENCE IN THE DEVELOPED AND DEVELOPING WORLDS M.E. Fernández-Giménez and U.P. Kreuter…………………………………….………...………2

RANGELAND PROFESSIONALS AND POLICY DEVELOPMENT IN THE UNITED STATES L.D. Butler and D. Thompson………………………………………………………...…………10

RANGELANDS AND RANGELAND POLICY IN ARGENTINA I. Feldman…………………………………………………………….……………...…………..17

RANGELAND POLICY DIALOGUE, FORMULATION, AND IMPLEMENTATION: PERSPECTIVES FROM BOLIVIA, ETHIOPIA, AND KENYA D.L. Coppock, A. Aboud, H. Alzerréca, and S. Desta…………………………………………..22

RANGELAND POLICY IN SOUTH AFRICA: DRIVERS AND PASSENGERS D. Grossman………………………………………………………….………………………….34

EFFECTIVE RANGELAND POLICIES FOR IMPLEMENTING GLOBAL CONVENTIONS IN AFRICA G. Oba…………………………………………………………………...……………………….42

TAKING STOCK: POLICY, PRACTICE, AND PROFESSIONALISM IN RANGELAND DEVELOPMENT R. Mearns……………………………………………………………………….………………..52

1 Rangeland Professionals and Policy: Prospects for Effective Influence in the Developed and Developing Worlds

MARÍA E. FERNÁNDEZ-GIMÉNEZ AND URS P. KREUTER

Authors are Assistant Professor, School of Renewable Natural Resources, University of Arizona, 325 Biological Sciences East, Tucson, AZ 85721; Assistant Professor, Department of Rangeland Ecology and Management, Texas A&M University, College Station, TX 77843-2126

Introduction Rangeland management is a profession dedicated to the goals of promoting stewardship and scholarship of rangeland resources; efficient and sustainable management of rangelands for a variety of values and products; public appreciation of the social and economic benefits of rangelands; and the professional development of its members (SRM Mission Statement). Our membership is dominated by professionals with training and expertise in the applied life and physical sciences. Yet, whether we work “on the ground,” “behind a desk,” or “in the ivory tower,” rangeland professionals today negotiate an increasingly complex policy landscape. Often we encounter policy in its formal manifestations--laws and regulations--as bureaucratic obstacles to be overcome or avoided, and not as a process in which we should actively engage. In the developed world, changes in public values have both obstructed management through litigation, and fostered new and creative institutional arrangements such as the collaborative stewardship and LandCare movements in the USA and Australia respectively. Rangeland professionals in the USA were actively involved in the early prototypes of collaborative stewardship—Coordinated Resource Management Planning and the Experimental Stewardship Program. In the developing world, growing attention to biodiversity preservation, and the increasing vulnerability of pastoralists to climatic variation and economic and socio- political upheaval create opportunities for both disaster and constructive social change. While social values of rangeland resources have been evolving, the field of range ecology also has advanced new non-equilibrium theories of rangeland dynamics, a paradigm shift with important implications for rangeland policy and management. These changes in social values and ecological theory, coupled with the ongoing discussion of SRM’s role in Washington DC, led us to ask several questions: ?? How do science and politics affect rangeland policy development? ?? What is and what should be the role of rangeland professionals in developing and implementing rangeland policies? ?? How can rangeland professionals around the world meaningfully influence rangeland policies at the local, national and global level? ?? Should professional societies be actively involved in the policy arena?

2 ?? What, if any, is the appropriate place for expatriate influence in the domestic rangeland policy of developing nations? ?? What are the respective roles and effectiveness of decentralized vs. centralized institutions in formulating and implementing policies that promote sustainable rangeland management? ?? What can we learn from each other about the ways that rangeland professionals contribute, or don’t, to policy formation in our different nations? To address these questions, the International Affairs Committee of the Society for Range Management sponsored a symposium that was held during the 55th meeting of the Society held in February 2002 in Kansas City, Missouri 1. The purpose of this symposium was to initiate a dialogue on these issues, bringing them from the domestic arena to an international one. The papers presented during the symposium, which are included in this volume, offer diverse perspectives on these questions from academia, government, professional societies, and international donor organizations in the United States, South America, Africa and Asia. In this introduction to the symposium proceedings we provide an overview of the major findings and views reported in these papers, concentrating on the roles of rangeland professionals, range science, and politics in policy-making. In our concluding summary we highlight four major themes that emerge from the six papers presented: 1) short-term social and economic concerns override technical factors in most rangeland policy-making; 2) links between science, policy and management must be strengthened through adaptive management, participatory research and proactive communication by rangeland professionals; 3) rangeland policies must account for the historical environmental, social and economic conditions of specific locations; and 4) broad participation in policy-making should be promoted at all levels of governance. We conclude that rangeland professionals must engage actively in policy development if our goals of enhancing the sustainable management of rangeland resources are to be reached. We also identify areas in which we need to build our professional capacity to achieve this objective. Finally we suggest that this shift in our professional scope calls for rethinking range education.

The Role of Rangeland Professionals in Rangeland Policy Development In most of the world, rangeland professionals currently appear to play a minor role, if any, in the development of policies that affect rangelands. In fact, much of the developing world lacks policies directed specifically towards the conservation and management of rangeland resources or the well-being of pastoral or ranching populations that depend on them. As Feldman points out in his introduction to Argentine rangelands, Argentina lacks policies or regulations for its rangelands, despite the fact that the beef cattle industry, which depends heavily on natural pastures, is historically a major economic driver in Argentina. Feldman attributes the paucity of attention to rangelands in part to the more subtle physiognomy of rangeland vegetation types compared to dramatic forests, and the fact that changes in rangeland vegetation cover or composition are less noticeable to the untrained eye than the felling of a forest.

1 This symposium is the second in a two-part series. The first symposium in the series, presented in Kona, HI in 2001 was titled “Rangeland Professionals and Society: Future Directions.” The papers presented in that symposium were printed in Volumes 23.5-6 and 24.1 of Rangelands.

3 In their overview of rangeland policy in Bolivia, Kenya and Ethiopia, Coppock et al. also observe that policies specific to rangeland management are rare, and rangeland professionals have little influence on their design. The principal policies affecting rangelands and their inhabitants in these nations are those that dictate access to land and land use planning. In Kenya and Ethiopia, short-term political and economic considerations tend to override long-term sustainability in the political calculus of policy-makers, leading to policies that promote environmentally and economically risky investments, e.g. cereal crop production in fragile soils. In all three countries, pastoralists whose depend on the land are often politically marginalized, and key resources needed to sustain their production systems (the more productive patches in the matrix of arid land) are captured by political elites and converted to other uses such as cropland. In his review of past trends and future directions in World Bank rangeland investments, Mearns points out that "natural resource management" per se is seldom a priority identified by pastoralists, whose concerns tend towards more immediate needs such as water development and veterinary services. Land tenure and resource access again emerge as key policy domains influencing rangelands and their pastoralist populations in the developing world. Mearns also points out that historically, rangeland professionals influenced the direction of development on arid rangelands, with distinctly mixed results. The failure of many range improvement and pastoral development projects in the mid 20th century led to a pronounced decline in donor investments in these areas, despite the increasing vulnerability of pastoral populations and rangeland landscapes. More recently, the assumptions that underlay the conventional wisdom of pastoral development have been challenged by a new generation of range ecologists and development practitioners. Perhaps this will lead to renewed investment in pastoral development, with greater attention being paid to the integration of sustainable livelihoods and landscapes, increased sensitivity to the nuances of pastoral land tenure, and a commitment to a more participatory development that emphasizes local capacity-building over technological solutions. Butler and Thompson as well as Grossman (whose essays focused on rangeland professionals and policy in the USA and South Africa, respectively) offer specific examples of ways in which rangeland professionals have effectively engaged in policy-making. Butler and Thompson identify two principal pathways of rangeland policy development--policies originated through the political process by the public, and those with their origins in science--and discuss the role of range professionals in each. Grossman describes an unprecedented, national-scale participatory planning process that was initiated at the dawn of South African democratic government in 1994. Rangeland professionals were formally brought into this process in several ways, including the incorporation of professional societies such as the Grasslands Society of Southern Africa into discussions. The Land and Agriculture Policy Center was also established to assist in policy formation, and was staffed by both South African and expatriate experts.

The Role of Range Science in Rangeland Policy-Making In the developing countries discussed in this volume, science has had less influence in policy formation than politics and economics. Developed nations, such as the USA, and some of the more advanced developing nations, support mature scientific infrastructure for range science and management in universities and extension programs. Even Mongolia, among the least developed nations, invested heavily in science education and research during its socialist era, but

4 little of this research was geared towards management of range resources, which constitute over 80% of the country and sustain its dominant economic sector--livestock husbandry. In addition, USA, South Africa and Argentina (as well as Australia and Kenya) host professional rangeland societies, the youngest being the Argentinian Association for the Management of Natural Pastures, established in 1999. However, in developed and developing countries alike, there is often a dramatic disconnect between the knowledge generated in academic or research institutions and its communication to and application by producers, land managers or policy-makers. Several of the authors point out that scientists must learn to communicate their results clearly and simply to decision-makers, who are more interested in identifying the scientifically defensible course of action that will satisfy the largest number of constituents, than understanding the complexities and caveats embedded in the answers that scientists provide. Yet oversimplification holds its own perils. Grossman points out the need for extreme caution in making environmental "doomsday" predictions, however compelling these messages may be, because when the predicted scenarios fail to occur (as in the case of predicted desertification in South Africa), the credibility of range scientists suffers, and the chances for effectively influencing future policies are weakened. Similarly, rangeland scientists must resist the temptation to offer "one-size-fits-all" management or policy remedies, whether rotational grazing systems or land titling of grazing commons. Generalized technological interventions and land tenure policies were responsible for many of the range improvement and development debacles of the 20th century. This statement holds not only for past development efforts abroad, but also for policies implemented in the USA, including those promoted for American Indian Nations of the Southwest (Fernandez- Gimenez, personal observation). Oba points out that, if they pay attention to local conditions, range professionals and scientists could contribute significantly to the implementation of global policy initiatives such as the Kyoto Protocol by using past data gathered in developing countries to help establish empirical baselines and monitoring protocols for carbon sequestration investments and carbon trading. Another emerging question is, what is the relevance of recent advances in rangeland ecology--notably the rise of "non-equilibrium" theories of rangeland vegetation dynamics and plant-animal population interactions--for the development of range policy? Coppock et al. determined this debate to be superfluous in the context of the political and economic realities of range policy and management in the developing nations they reviewed. Similarly, in countries such as Argentina where no range policies exist, it is meaningless to discuss the influence that changes in range science have had on policy-making. Butler and Thomson point out that, in the USA, the integration of state and transition models into ecological site descriptions indicates the move in Natural Resource Conservation Service (NRCS) to institutionalize the new paradigm. Ironically, the site-specific and historic nature of the data needed to accomplish this, as well as lingering murkiness as to what exactly constitutes a "state," a "transition," or a "threshold," will make it a slow and challenging process to complete. According to Mearns, the place where the "new range ecology" appears to hold most sway is in the arena of multinational donor organizations such as the World Bank. In this context non-equilibrium theory supports development approaches that appreciate the importance of flexibility, mobility, diversity, and opportunism to pastoral livelihoods and rangeland sustainability. This is evident in the increased attention being paid to the need for greater flexibility, site-specific tenure policy, and a "two-track" approach to policy development within

5 the Bank that incorporates both local initiatives and national-level policy dialogue. Mearns also admits that in order to gain local support, natural resource management initiatives must in most cases be tied to locally applicable land policy. Yet, the relatively inflexible legal framework for land tenure in many countries makes implementation of policies that support flexible and mobile management challenging.

The Role of Politics in Rangeland Policy-making If science takes the back seat to politics in the formation of range policy, how do politics drive the process? The papers in this volume provide a fascinating cross-section of political processes around the world as they relate to rangelands and their management. Across the board, politics at the local, national, and even international scales can be seen to influence rangeland policies. It is relatively easy to understand how local politics come into play, particularly in poorer developing nations, through the capture of high productivity land and other assets by local political elites, and their emphasis on investments with short-term returns to perpetuate existing power structures. A more revealing theme that emerged from this set of papers is the importance of macropolitics, of taking a historical view of macropolitical transformations, and of attending to the interactions between politics at local, national and global scales, to understanding the role of politics in rangeland policy. The macropolitical structures and climate provide the overall framework for the ways in which local people, rangeland professionals, and politicians are able to contribute to policy. Where democracy and political stability prevail, and where there is a framework for decentralized decision-making, the opportunity for local and scientific input into policy is greater. Coppock et al. indicate that in some developing countries there is a trend towards interest group and advocacy politics, even at the local level. This development provides a possible avenue for involvement by rangeland professionals in the policy process as information resources for affected local groups. Even where the political system theoretically allows for democratic process, pastoralists and other rangeland users are often in the political minority and may be excluded from these processes. Furthermore, the priorities of the populace may have little to do with long-term sustainable economic development or resource management. And, rangeland professionals by and large have not actively sought to engage in the political process of policy-making at local, national or international scales. Grossman provides a well-documented historical account of the relationship between macropolitics in South Africa and the role that rangeland professionals have played in three phases of South African history, from the first decades following the nation's formation in 1910, through 45 years of Apartheid, to the "outbreak of democracy" in 1994. This historical overview brings home the point that personal interest on the part of a political leader can significantly influence certain policy sectors, including rangelands. It also highlights the fact that even solicited scientific advice may be emphatically ignored or even denounced if the science or its policy implications contradict the prevailing political ideology. Butler and Thompson point out that in addition to science-based and grassroots-originated policies, much policy making (and breaking) in the American political system happens through the Congressional budget negotiations. These negotiations in turn are influenced both by national political debates and local interest ("pork barrel") politics.

6 International politics intersect with local and national concerns when global conventions are implemented locally. Oba notes the potential for "carbon colonialism" when developed nations invest in carbon sequestration projects such as aforestation in developing countries, and reap double benefits of carbon mitigation credits and timber profit while leaving the host nation with modest land rents and unaccounted opportunity costs, land use conflicts, and displaced farmers. Oba also points to the potential for constructive involvement of rangeland professionals in renegotiating the Kyoto Protocol in its second phase, and helping local land users enter the global carbon market in a way that promotes sustainable resource management.

Summary and Conclusions The perspectives included in this volume do not pretend to represent an exhaustive survey of the role of rangeland professionals in policy-making around the world. Nevertheless, they form an instructive sample, from which several common themes emerge. ??Immediate social and economic concerns override technical factors in policy-making for most rangelands. To achieve conservation and management goals, rangeland professionals must therefore reframe policy agendas to integrate social and economic priorities with resource management concerns. Policy-making that affects rangelands, is driven largely by non-technical factors because politics and perceived short-term economic and social needs prevail over science and resource management. The lack of explicit policies that address rangeland conservation and management is an indication of the relatively low priority of resource management for many rangeland inhabitants and users, and consequently for their political representatives. Rather, the foremost concerns for many pastoralists and ranchers are the immediate well-being of families, communities and livestock. Community and individual well-being in a resource-dependent economy are, of course, inextricably linked to the health of the natural resources on which production is based. Thus, instead of despairing that rangelands are ignored, rangeland professionals must refocus their efforts to clearly articulate the interdependence of sustainable livelihoods, community well-being, and sound stewardship, to policy-makers and local populations. They should also support policies and aid programs designed to address economic, social and resource management concerns in an integrated fashion. In many arid and semiarid rangelands, the major areas of rangeland policy emphasis will be risk management, resource tenure, and poverty alleviation. Thus, improved access to credit and markets for pastoralists may be key in many areas.

??The links among science, policy and management must be strengthened, through adaptive management, participatory research, and clear and proactive communication by rangeland professionals. With some exceptions, primarily in the developed world, science seldom drives rangeland policy. However, rangeland professionals could do much more to bring their science to policy-makers, and to communicate it to rangeland users in a user-friendly form. Range science would likely be more relevant to resource users if research were conducted in a more participatory manner, integrating local and traditional knowledge in theory-building and research design. Adaptive management promises to integrate research and management by applying experimental design to alternative management practices. An adaptive management approach is also useful for implementing development projects.

7 Advancement towards these objectives will require commitment by individual investigators and a change in the institutional culture of academic and government research entities.

??Rangeland policies need to take into account the historical environmental, social and economic conditions of specific locations. National-level policies should provide sufficient flexibility to allow for locally-adapted implementation. It is difficult to overstate the importance of understanding local environmental and social history in the design of effective rangeland policies. In developing nations, and among indigenous peoples in all countries, it is crucial not to overlook traditional resource management institutions, and to understand the ways they have been transformed through interactions with national and global markets and political systems. Such institutions can be instructive for contemporary policy formulation even when they are no longer viable in their historic forms.

??Promote broad participation in policy-making at local, national and global scales. Countries with more stable and genuinely democratic systems of government appear to offer greater opportunities for popular engagement in policy-making, by rangeland users and professionals, as well as other stakeholders and interest groups. Transparency and accountability in policy-making at all levels helps to ensure that formal democratic processes are not derailed by corruption or dominated by narrow interest groups. In many cases, a concerted effort will need to be made in order to bring traditionally marginalized voices, including those of minority rangeland user groups, into the policy discourse. This is neither a simple task, nor one without risks, since it implies potentially profound shifts in the existing balance of power. The papers herein provide several positive examples of the "win-win" outcomes that inclusive policy-making processes can support, such as the Working for Water program in South Africa and the Grazing Lands Conservation Initiative in the USA. There are many other examples of locally-based multi-stakeholder collaborations that have helped to resolve long-standing or acute resource conflicts, or develop resource management plans with strong public support. While collaborative policy-making is not a panacea nor the most appropriate or efficient approach for all issues, public participation in policy-making, including involvement of rangeland professionals, is to be encouraged whenever possible.

We believe that rangeland professionals can no longer afford to view policy-making as outside our domain of expertise and appropriate involvement. As the papers in this volume make clear, if we, as rangeland professionals, do not take the driver's seat, we and the resources that we advocate for will be taken for a ride. It will be a journey that we have no hand in guiding and a destination potentially devastating to the principles of sound science and conservation that we uphold. In order to claim our place behind the wheel, however, we will need to rethink our role as professionals, and the scope of range science as a discipline. Many of us were drawn to range science because of the inherently integrative and interdisciplinary nature of the field, and the opportunity to study the interactions among soils, plants, climate and herbivores. Now we must go a step further to fully integrate human social and economic systems into our understanding of rangeland ecosystems and their management. It is no longer sufficient to be technically proficient in the biophysical foundations of range

8 science, with a working knowledge of relevant laws and policies. Today's rangeland professional must be able to perform integrated socio-economic and ecological analyses of rangeland ecosystems, and identify the "killer assumptions" underlying proposed management actions. These needs are nowhere more apparent than in the international arena. Indeed, international rangeland researchers and professionals have been at the forefront in integrating social and ecological systems in research and management for some time. As we argued above, in the developing world, rangeland conservation and management objectives will only be achieved if closely linked to social and economic development priorities. To be effective in the policy arena, rangeland professionals must also dramatically improve their communication skills. We must proactively share our knowledge and concerns with policy-makers, and to do so we must develop expertise in communicating ecological complexity in simplified terms without overreliance on "doomsday" scenarios. Our job will be made easier if we involve the public more directly in our science, through participatory research and adaptive management. Such collaborative learning efforts may have the added benefit of increasing the scientific literacy of the public, so that laypersons better appreciate the underlying complexities and uncertainties associated with rangeland science and management. Finally, these shifts in professional and disciplinary scope have important implications for range education. Knowledge, skills and abilities that students will require to meet the challenges outlined above call for increasing emphasis in range curricula on social sciences, integrated ecosystem analysis, and communication skills, including mediation and negotiation. It is clear that our role as professionals must change if we are to achieve our objective of promoting sustainable management of rangelands. Specifically, we must take a more active stance in influencing policies that directly and indirectly affect rangelands. This broadening of our professional role does not, however, imply that we should abandon the knowledge base, professional culture and networks that we have painstakingly built over the past half-century. Rather, we must expand our knowledge base to address the interactions of social, economic and ecological systems, and enlarge our professional networks to include a growing client base with increasingly diverse land management objectives. The papers in this volume provide a wide range of examples of the need for constructive involvement in rangeland policy-making by rangeland professionals, and ample fodder for debating effective avenues of influence. This compilation shows that rangeland professionals have much to contribute to the development of policies that support the conservation of rangeland resources and the well-being of communities that depend on them. Our effectiveness will depend on our ability to expand our professional capacity, and on our initiative and commitment as individuals, institutions and professional societies to engage at all levels of policy formulation and implementation.

9 Rangeland Professionals and Policy Development in the United States

LARRY D. BUTLER AND DENNIS THOMPSON

Authors are Director, Grazing Lands Technology Institute, Natural Resources Conservation Service, P.O. Box. 6567, Fort Worth, TX 76115; National Range and Grazinglands Ecologist, Natural Resources Conservation Service, South Ag. Building Rm 6150, 14th and Independence Ave SW, Washington, D.C. 20205-0016

Introduction The purpose of this symposium is to present diverse perspectives and initiate a dialogue on the role of rangeland professionals in developing and implementing rangeland policy in the United States and abroad. This implies there are diverse perspectives, maybe differences of opinion as to the role of rangeland professionals in policy development and implementation. Some say the Society for Range Management (SRM) should have no role in policymaking, that SRM is a forum to develop professionally by presenting and discussing the science and that political activity is beyond the purpose and objectives of the Society. We do not believe this to be possible if SRM is to represent the profession and science of rangelands. The members of SRM come from many different backgrounds and work in the field of range and rangeland science and management in a variety of ways. Some rangeland professionals achieve satisfaction in learning all they can about the ecology of rangelands and related resources and place little effort into policy development and implementation. The knowledge they gain will likely be passed on through organizations such as SRM and used by not only rangeland professionals but also others in the development and implementation of policy. This presentation will focus on process and not content of any particular policy. Although there are policies that affect both privately owned rangelands and federally managed public lands, there are others that predominantly affect one or the other. My co-author and I are employed by the USDA Natural Resources Conservation Service (NRCS). NRCS ‘s responsibilities lie with the transfer of technologies to managers and owners of private lands; hence our experiences have dealt primarily with policies affecting private lands. Rangeland resources are vast, demands on these natural resources are increasing and knowledge of their values, functions and capabilities is growing. Policy is going to be developed that impacts rangelands and their management. The rangeland professionals will be involved and should take the lead in the development of emerging policies. The benefits to the nations' rangeland resources will be proportional to the efforts of the range profession in education of the public, political representatives, agency leadership and other interest groups.

10 Rangeland Resources in the United States Definition of Rangeland Society for Range Management, 1989, A Glossary of Terms Used in Range Management: Land on which the native vegetation (climax or natural potential) is predominantly grasses, grass-like plants, forbs, or shrubs. Includes lands revegetated naturally or artificially when routine management of that vegetation is accomplished mainly through manipulation of grazing. Rangelands include natural grasslands, savannas, shrublands, most deserts, tundra, alpine communities, coastal marshes and meadows. National Resources Inventory, 1997: A land cover/use category on which the climax or potential plant cover is composed principally of native grasses, grasslike plants, forbs or shrubs suitable for grazing and browsing, and introduced forage species that are managed like rangeland. This would include areas where introduced hardy and persistent grasses, such as crested wheatgrass, are planted and such practices as deferred grazing, burning, chaining, and rotational grazing are used, with little or no chemicals or fertilizer being applied. Grasslands, savannas, many wetlands, some deserts, and tundra are considered to be rangeland. Certain communities of low forbs and shrubs, such as mesquite, chaparral, mountain shrub, and pinyon- juniper, are also included as rangeland. [NRI-97] Regardless of which definition we use, rangelands of the United States are truly a national treasure. There are many uses and values that make them extremely important, not only to their owners, but also to the entire nation. Privately owned rangelands constitute about one-third of the total land base of the lower 48 states. The majority of U.S. rangelands will not support cultivated crop production; however, most do support plant communities that can be grazed by livestock to transform this renewable resource into food and fiber products. Rangelands can be managed as a renewable natural resource while providing a sustainable form of agriculture production. Proper management is essential for the sustainable production of food and fiber, as well as supporting a wide diversity of other uses. Healthy rangelands provide an economic base for many regions of our country. They are watersheds, which contribute to good water quality and sustained stream flows. Tributaries, and the rivers that they feed, are the source of water for agricultural, domestic and municipal uses, power production, and fish and wildlife. Some private rangelands are also used for outdoor recreation, camping, hiking, hunting, and fishing. They provide essential habitat for many wildlife populations. National, state, and local policies are developed for rangelands for a variety of reasons including; assuring the sustainability of rangelands, to enhance these lands for future use and to inform the public of the benefits of making the appropriate commitment and investment. There are more potential conflicts than ever on rangelands within the United States. Demands by landowners and society for rangeland benefits, values, and products are increasing. Private rangeland owners utilizing their lands in a business enterprise require a solid economy in which to operate. Policies that impact their ability to implement needed management decisions may impact their economic sustainability. Society is increasingly recognizing the benefits from clean air, healthy wildlife populations and habitat, improved fisheries and aquatic systems, healthy riparian areas, soils and water, reduced potential for flooding; less sediment in streams and reservoirs, and economic and social stability. An available supply of food and fiber, which

11 also comes from rangelands, is now an afterthought because of their abundance in our country. The ownership and/or management of rangelands, and thus the control of the management decisions, change in a cyclic manner due to generational changes and economic conditions. The future condition of these lands and the benefits from them depend on the understanding and decisions of the landowners and managers. Most owners and managers are good stewards of these resources and implement sound conservation practices. Some have limited exposure to sound conservation principles, while others have received little of the necessary follow-through assistance to recognize and/or treat their conservation problems. These owners and managers need to be taught and new science and technology must continually be made available in a practical manner so that these landowners may make informed decisions concerning this vital resource.

What is a rangeland professional? What is a rangeland professional? One member of SRM recently said there are only a couple of hundred professionals in the Society. This is a very narrow definition of a rangeland professional. That definition is too narrow to be accepted in this discussion. For purposes of this symposium, the following definition will be used. A rangeland professional is any individual that derives a significant portion of his or her livelihood in the study of rangelands and/or the application of sound rangeland management principles and practices. This does not imply the person always acts and reacts in a professional manner; that is a subject for another symposium. Rangeland professionals include ranchers, land owners, land managers, agency personnel, consultants, academicians, scientists and others. Not all rangeland professionals are members of SRM. Are all members of SRM rangeland professionals? Probably not, as interest in rangelands does not make one a rangeland professional.

What is policy? One definition of policy is “A plan or course of action, as of a government, political party, or business, intended to influence and determine decisions, actions, and other matters.” A course of action, guiding principles, or procedure considered expedient, prudent, or advantageous. It can be defined as the settled method by which the government and affairs of a nation are, or may be, administered; a system of public or official administration, as designed to promote the external or internal prosperity of a state. Considering these definitions of policy, rangeland policies can be defined as plans and courses of action selected as the settled methods intended to influence and determine decisions by which the rangelands and associated resources are managed. Rangeland policies do not always deal specifically with rangeland management. Rangeland management was defined in 1989 by the Society for Range Management as “A distinct discipline founded on ecological principles and dealing with the use of rangelands and range resources for a variety of purposes. These purposes include use as watersheds, wildlife habitat, grazing by livestock, recreation, and aesthetics, as well as other associated uses.” It is obvious from this definition there are many issues involved in rangeland management; therefore, there are policies written specifically for rangelands and policies

12 developed for other concerns or issues that directly and indirectly affect rangelands. An example of the former is the 1983 United States Department of Agriculture (USDA) Departmental Regulation Number 9500-5 Subject: Policy on Range. The following is an excerpt:

It is USDA policy to implement progress to: a. Protect and enhance range ecosystems. b. Improve the productivity of soil, water, plant, livestock and wildlife resources of the Nation's rangelands. c. Contribute to the social and economic well being of people and communities that depend on range for their livelihoods. d. Support and strengthen national and international cooperation designed to improve range condition and productivity. e. Provide continuing research and technology developments to enhance range conservation and productivity and to make optimum use of all range resources. f. Provide continuing education and other technology transfer programs that improve livestock production on rangeland in balance with watershed management, wildlife and fish habitat, and other uses. g. Emphasize cooperation and coordination among Federal, State, and local agencies; private organizations and institutions; and individuals in planning and executing range programs to benefit the Department's range constituency and to prevent duplication and overlap of duties in reaching interagency goals affecting range. h. Provide technical, managerial, education, and other assistance programs for range owners, operators, and other users that will encourage the adoption and use of conservation practices that protect and enhance range resources. i. Provide periodic inventories of the Nation's range resources that are useful to rangeland users, that are compatible among agencies, that are based upon sound economical and ecological principles, contain data which identify multiple-use opportunities, and are useful for protecting and enhancing range resources at the local, regional, and national levels j. Manage the range resources of the National Forest System in an economically and ecologically sound manner, for multiple use, and for development and demonstration of effective range management.

Examples of policies that are related to rangeland management or would affect rangeland management would include the Endangered Species Act, the Clean Water Act, and Executive Orders such as Invasive Species.

Formulation and Implementation of Rangeland Policies Rangeland policies are formulated and implemented in the United States at various levels and in a number of ways. Policy is set by statute when Congress passes laws. Congress can guide policy with budget allocations. This can cause a desired action or inaction. The Congress can do this with line item funding, budget earmarks, and budget language. The policies set and/or affected by Congress are generally national in scope, but can be directed at a local level as

13 well with line item funding directed to specific congressional districts to make something happen. The Executive Branch of the U.S. government develops and establishes policy in a number of ways also. The President can issue Executive Orders. He also makes decisions to guide agencies to develop and set policy. The President’s Cabinet members also develop and set policy as well as Agency administrators. These can vary in scope of application also, and include proposed budgets and laws.

Types of Policies There are various types of policies. Two of these types are (1) those originated through political process by the public and (2) those that are science-based in origin. The Grazing Lands Conservation Initiative (GLCI) is an example of how the interested public has used the political process. The GLCI was originated by grassroots organizations to affect the conservation of private grazing lands in the U.S. These lands include all private rangelands. These organizations passed their own internal policy to support this effort. A National Steering Committee was established to include representatives of the organizations that established the GLCI. These are members of the dairy industry, the National Cattlemen's Beef Association, the American Farm Bureau Federation, the American Sheep Industry, the American Forage and Grassland Council, the Society for Range Management the Soil and Water Conservation Society, the National Association of Conservation Districts and the National Farmers Union. Other conservation, environmental, and producer organizations have been involved with the committee to one degree or another. There are also state-level GLCI committees comprised of representatives of state affiliates of the national organizations represented on the National Steering Committee and other interested groups and individuals. The national and state-level GLCI committees have been very effective in guiding and developing rangeland and pastureland policies at all levels. The 1996 Farm Bill included Sec. 386 the Conservation of Private Grazing Lands, authorizing development and funding of a program to address conservation needs on private grazing lands. Grazinglands which include rangelands are now included in much of the conservation debate occurring regarding the development of the next farm bill. Particulate Matter 2.5 and Regional Haze rules are policies being developed that will apply to many kinds and uses of lands. They are examples of policies that are both science- based and originated by a joint collaboration of agency personnel, scientists and interested publics. These policies can have far reaching impacts to rangeland resources and their management. The use of prescribed burning may be greatly impacted if range science is not part of the foundation utilized in development of policies impacting water and air quality. Imagine the nations' rangeland resources with prescribed fire no longer available to the range manager as a management tool.

14 Development of Rangeland Policy by Rangeland Professionals Rangeland professionals are expected to be part of the policy development process by the administration, congress, agency leadership, the public and other organizations. Rangelands are recognized as a unique resource that needs consideration in development of national policy. Policy development without the proper consideration of rangeland resources could merely result in missed opportunities or it could result in disastrous impacts on the resource. Rangeland professionals are often asked to write policy based on sound science. For example, rangeland professionals use science to find solutions to problems and provide policy makers with information to assist in the understanding of potential impacts of proposed policies. The prescribed burning policy of an individual agency is an example of this. The U.S. Farm policy while certainly science based is more general in nature and less specific to a particular science. Rangeland science is used to set background standards upon which to implement policies. The development of Ecological Site Descriptions with the inclusion of the latest state and transition theories is an example of this. These descriptions can be utilized as standards upon which to base rangeland health evaluations and other assessments. One role of science and politics in the development and implementation of policy is to help and potentially require agencies to use common policy. Politics is often the driving force to encourage science to come together with useable solutions. Rangeland professionals, such as ranchers, land managers, and academia, can meaningfully influence rangeland policies in a number of ways. They can provide direct input when called upon or volunteer the input at appropriate times to appropriate people and/or organizations. This can be at the local level through SRM sections and conservation districts or at the state or national level through the Society for Range Management or the National Association of Conservation Districts or a particular state association. As previously mentioned, the GLCI (which contains rangeland professionals) can provide inputs and call upon other rangeland professionals for input also. The same can be said for producer organizations, conservation groups, and others. Rangeland professionals are asked to prepare briefing papers for new administrations. This is done within agencies for distribution through the Executive Branch, Congress and elected officials and their staffs. This is also done by organizations, such as the SRM transition paper that was developed and distributed in 2000. When new programs are developed, rangeland professionals are called upon to participate in the writing of the rules and regulations for the programs. These must comply with the laws passed by the Congress and signed by the President and they must be technically sound. Rangelands professionals can be frustrated by decisions made by officials that may not be deemed by the rangeland professional as the most technically sound. The senior officials must take many other factors into account that may override the technical advice or technical choice. Often decisions are based on political factors and concessions must be made. The rangeland professional must not be overwhelmed by this and must not falter when next asked for rangeland technical input or stop offering volunteered advice. Conditions are different with each issue and over time.

15 Often when a program is being developed, an administrator or special assistant to the administrator will ask the rangeland professional for a “Rule of Thumb” or a “Magic Formula” that “Fits All” throughout the nation. This can be very frustrating as we rangeland professionals know we must be site specific to make rangeland health assessments, production potential forecasts, wildlife potentials, or any other ecological assessment. The organization and functions of the Society for Range Management can assist in bringing rangeland issues to the table and assist in policy development based on sound and appropriate science. Rangeland professionals must continue to develop the knowledge base and technologies and be prepared to reply and be proactive on appropriate issues.

16 Rangelands and Rangeland Policy In Argentina

ISRAEL FELDMAN

Author is President, Asociación Argentina para el Manejo de Pastizales Naturales, Mendoza 1819 Dto. A, 2000 Rosaria, Argentina.

Argentina’s Rangelands Argentina is one of the southern-most countries of South America. It extends from 21º 46’ S to 55º 21’ S and from 53º 38’ E to 73º 29’ E., covering a length of 3694 km from the north at the Bolivian border to the last island in the Southern Atlantic Ocean, and 1423 km from east to west. The mountain range called the “Cordillera de los Andes” is the western boundary that separates Argentina from Chile. On the east, Argentina is bounded by rivers that separate the country from Brazil and Uruguay. Less than 10% of the east boundary with Brazil is dry. The southern part of the east boundary is the Atlantic Ocean. The Atlantic Ocean is also on the very south, which is the boundary with the Antarctic. In such an extensive area, there are various climatic conditions, which range from the wet subtropical to the dry semidesert. Most of the climatic conditions are continental, with great annual variations. In the northeastern part of the country rainfall may reach as much as 1700 mm per year. In the western part of the Aconquija Mountains, precipitation may be 1000 mm per year. In the lake region in the west southern part of the country, precipitation may reach 2000 mm per year. The Andes Mountains are covered with snow during winter and some mountain peaks hold permanent snow. The Patagonia region is cold and windy and the central part of the country has a mild climate without snow during winter, even though there are freezing temperatures. The northwest is semi-arid, with high summer temperatures. The northeast is subtropical-humid. When the first Europeans arrived, there were no large herbivores. Some ungulates, such as deer, guanacos, vicuñas, lamas and alpacas could be found in the different ecosystems. There were also some smaller herbivorous mammals, birds, and insects. In 1536, Pedro de Mendoza introduced the first cattle and horses into our country, which reproduced and spread rapidly. The country’s economy started to depend mainly on the cattle industry. Beef production and its by-products became so economically important that the Spanish word “hacienda”, which means wealth, began to be used as a synonym of cattle in Argentina. For many years Argentina was known as the beef country and nowadays beef production is still very important. There are about 50,000,000 head of cattle in all the country, 70% of the total herd grazes on rangelands. About 85% of Argentina, almost 2,500,000 km2, is covered by woodlands, shrublands and rangelands. The total herd of cattle is managed by almost 250,000 farmers or ranchers. The size of the farms or ranches ranges from a few to thousands of hectares. There are also 2,200,0000 sheep

17 and almost 4,000,000 goats in Argentina. Most of the sheep are raised in the Patagonia region and in Corrientes province, mostly in big ranches of more than 5000 ha. There are also sheep in the marginal areas of the country, where they are raised by small peasants together with goats. A total of 140,000 enterprises are engaged with the sheep and goat raising activity in the country. There have always been prairies and rangelands in Argentina. At present, most of the rangelands are private property. Below are described some features of Argentina’s different rangelands. The most impressive prairies were in the area now named Pampa Region. Nowadays agriculture is more important than beef production in this area. However there still are more than 10,000,000 ha of saline alkaline low wetlands, inland marshes, in the region. There are also some other hilly rangelands that are used as pasture lands. Different species of Paspalum, Bromus, Stipa, Panicum, etc. are the most important species. The lower areas are used for raising cattle and the western region is used for fattening. The Mesopotamia region comprises the eastern provinces of Misiones, Corrientes, and Entre Rios. Misiones has more woodlands than rangelands. Corrientes could be mentioned as a complete range area, with wide extended low land areas on the northern part, and gently hilly on the southern part of the province. In the low wetlands there are species which belong to the genera: Paspalum, Leersia, Luziola, Panicum, Tipha, Cyperus, etc. In the southern part the dominant species belong to the genera: Paspalum, Bothriochloa, Andropogon, Schizachyrium, etc. The same species are found in Entre Rios, which has a gently hilly landscape. The misuse of rangelands produces erosion and in some cases shrubs take over the pasture fields. It is worth mentioning that soils are poor in phosphorus in the Mesopotamia region. Supplementation with this mineral is essential for beef production, especially in Corrientes province. The Chaco Region comprises the northern central part of Argentina. It is part of the overall Chaco ecosystem that extends from Paraguay and Bolivia. Some phytogeographers classify this region as woodland. Woods cover a very important portion of this area, but in general there are patches of woods surrounded by grasslands. From ancient times forestry was the main economic activity. The famous quebracho colorado, Schinopsis balansae and S. lorentzii, used for producing tannin, is one of the most important trees in this region. Generally, the ecosystem is unstable and periodic droughts affect growth, especially grasses and forbs. Forest overexploitation and lack of controlled grazing on the grassland areas produced, in some cases, complete desertification, and in better situations brush invasion. Similar situations can be found in some Texas and Arizona areas in the USA. The most important forage grasses in the Chaco Region belong to the genera: Chloris, Trichloris, Eriochloa, Panicum, Paspalum, Stipa, Spartina, etc. The Puna Region or the High Andean desert extends from north to south, along the western boundary from Jujuy up to Neuquén. In ancient times agricultural cultures developed there. Grasslands supported the native animals, like lama and alpaca. The introduction of cattle and horses and consequent overgrazing destroyed the rangelands and initiated desertification. The most important grass species present in this region belong to the genera Pennicetum, Festuca, Bouteloua, etc. The Sub-humid Pampa and the Semi-arid Pampa are very important beef producing areas. Most of the area is rangeland, although some areas may have been cultivated historically. But

18 since cropping is hazardous, it was largely abandoned and most of the land is dedicated to beef production. The western part of this region has sandy soils, and overgrazing or misuse of rangeland may favor wind erosion. Sorghastrum, Eriochloa, and Stipa, are the most common genera in this region. The Patagonia region, which can be divided into the Patagonian Steppe and the Patagonian Desert, was formerly occupied by guanacos. This native herbivore was almost extinguished and replaced by sheep. Inappropriate use of the grasslands brought about the present situation: desert or semidesert. Human life can develop only on the riversides, and on some low wet areas. The most common grasses belong to the genera Stipa, Poa and Festuca.

Range Management in Argentina There were and are different approaches to rangelands or native grasslands in Argentina. The simplest one is: “Just use it”. This concept comes from the old times, when it was believed that the endless grasslands would last forever. Another approach focused upon identifying species and/or vegetation types, which was accomplished for academic purposes, but this knowledge did not reach the farmers or ranchers. Research projects on range management started about two or three decades ago, but they were very few in the beginning. There are some studies about grasslands and grassland management, and there is some extension work done on this matter that is not widespread. In general, native grasslands are undervalued in Argentina. They are considered to be good enough for raising or repasturing calves, while cultivated pastures are generally used for fattening cows or steers. Different federal, provincial, university, and private organizations are doing research and extension work on grassland management. In the frame of the Ciencia y Tecnología para el Desarrollo (CYTED) Program (Science and Technology for Development), studies are being carried out that concern biodiversity on grasslands and savannas. INTA (Instituto Nacional de Tecnología Agropecuaria), has a specific research and extension program about grassland management. Different universities are also conducting research work on grasslands, in some cases CONICET (Consejo Nacional de Ciencia y Técnica) may support projects of this type. It is worth mentioning two projects, both concerned with land use and biodiversity in relict rangeland areas. One project studies psammophytic steppes of the west inland pampa region, dominated by Sorghastrum pellitum and Elyonurus muticus and the other project is focused on one of the grasslands of the flooding pampa dominated by Paspalum quadrifarium. In spite of the importance of grasslands in the Argentinian economy, there are no regulations or laws about their management. There are laws about forest management, and in some provinces there are specific laws about preventing soil erosion. Soil erosion problems arose in the late 20th century, due to inappropriate use of cropping systems. Dust bowls in Argentina were much like those of the United States, mainly in the sandy soils in southern Buenos Aires and La Pampa provinces. Hidric erosion problems are frequent in areas with gently slopes, like Entre Ríos province. Some rules and laws about crop systems that prevent soil erosion were established.

19 Grasses and grasslands were the plants and vegetation “Cinderellas” in Argentina. There are several reasons for this situation. Forest is impressive--no man can walk but in its shade— while grasslands can be trodden. When a big tree is cut down, one has a feeling that many years of life have been destroyed. Grasses are feeble, and not much noise is made when they are eaten. Forest as a vegetation type is more stable than grasslands. Short-term climatic variations, such as droughts or floods will not affect the vegetation matrix. Grasslands can be affected by stressful climatic conditions, but their recovery can be fast as well. When a forest disappears it changes the landscape. A grassland may disappear, other species take over, and the layman won’t notice it. The harvest for human benefit in grassland is not the primary production. Meat, milk, skins, animal fibers, etc. are gathered in grasslands. A forest, in contrast, will produce fruits, or timber, that are primary products that can be collected and used directly by people. All these situations lead the general public to feel strongly about forests but care little about grasslands. The control of the wood production is relatively easy. Laws and regulations for forest management can be implemented without an ecosystemic approach. This does not mean that they should not take an ecosystemic approach, but rather that they are often implemented without one. An ecosystemic approach has to be considered for range management, but each situation needs a different approach, and general rules or laws have to be adapted for site-specific conditions. Cattle raising farmers and ranchers pay much attention to the breed of their herds and how to manage them in order to get the highest production and profits. But they do not pay enough attention to range management. The situation is quite different among ecosystems. For example, the Pampa Húmeda region differs completely from the more xerophytic regions. The cultural practices also differ. There is more fencing in the Pampa region and plots are smaller than in other regions. Water is more sparsely distributed and animals have to walk long distances to reach water in xerophytic areas. Although there is scattered information about grasses and grasslands and their management, this information does not always reach the ranchers or cattle farmers. Since the early 1980s the local office of INTA in San Cristóbal, Santa Fe, has been organizing a Biannual Meeting of Range Management. During the 4th meeting held in March 1999, the Asociación Argentina para el Manejo de Pastizales Naturales (AAPMPN: Argentinian Range Management Society) was founded. Its main goals are: ?? Promote research and extension work on rangelands and range management ?? Provide society with information about the economic and social importance of rangelands ?? Stimulate the organization or creation of specific range management courses at different levels: high schools, under and postgraduate students, and for general public. ?? Develop laws and regulations in order to stimulate multiple and sustainable range management policies in collaboration with government officials, legislators and NGO’s. To fulfill these goals, The 1st Argentinian Range Management Congress was held in August 200, at San Cristóbal, Santa Fe. There were 120 scientific and technical papers submitted and more than 400 scientists, technicians, students, and range owners attended the meeting. We felt highly honored by the presence of Jim Linebaugh from the SRM (SA).

20 The AAMPN also organized different courses for cowboys, ranch owners and range management professionals. Specific contacts are being established with different federal and provincial officials and politicians from several parties represented in Congress in order to call their attention to the importance of laws related to sustainable range management. Range management practices should be locally adapted or developed, not only for each specific area but for each specific situation. Professionals, ranchers, and cattle farmers, individually or through their organizations have to play an important role in this task. Credit institutions should be involved in order to develop specific credit lines for range management activities. The society as a whole must become aware of the importance of rangelands for human welfare, quality of life, and high productivity. On this note, I close with the words of the American poet, Walt Whitman from Leaves of Grass.

“A child said What is grass? Fetching it to me with full hands, How could I answer the child? I do not know what it is any more than he. I guess it must be the flag of my disposition, out of hopeful green stuff woven. Or I guess it is the handkerchief of the Lord, A scented gift and remembrancer designedly dropt, Bearing the owner’s name someway in the corners, that we may see And remark, and say Whose? Or I guess the grass is itself a child, the produced babe of the vegetation.”

21 Rangeland Policy Dialogue, Formulation, and Implementation: Perspectives from Bolivia, Ethiopia, and Kenya

D. LAYNE COPPOCK, ABDILLAHI ABOUD, HUMBERTO ALZERRÉCA, AND SOLOMON DESTA

Authors are Associate Professor, Dept. of Rangeland Resources, Utah State University, Logan, UT 84322; Dean, Faculty of Environmental Studies and Natural Resources, Egerton University, P.O. Box 536, Njoro, Kenya; Private Consultant, Calle 4, Casa 300, Villa San Antonio Bajo, Casilla de Correo 13697, La Paz, Bolivia; Post-Doctoral Associate, Dept. of Rangeland Resources, Utah State University, Nairobi, Kenya, and Addis Ababa, Ethiopia

Introduction Rangeland professionals worldwide have begun to recognize the pervasive influences that the policy environment has on the management of rangeland ecosystems and the economic development of pastoral and agro-pastoral societies. Policies affect investment in infrastructure and public services, resource use and tenure, and rates of technology generation and transfer. People inhabiting arid and semi-arid lands are among the poorest on earth (Coppock pers. obs.). Such societies are commonly marginalized from mainstream political processes in their home countries and rangelands consequently become seedbeds for poverty traps, food insecurity, ethnic conflict, natural resource degradation, and violence. Despite decades of technical rangeland research and extension, sustained impact of technology or improved management systems on the livelihoods of pastoral people remains elusive (Coppock 1994, Coppock et al. 2001). This suggests that non-technical factors are what really control the fates of many rangelands and their inhabitants. The vast majority of rangeland scientists are still educated in the biophysical sciences (Coppock pers. obs.). As a consequence, rangeland scientists often have an over-simplified view of rangeland management and pastoral development and have limited perceptions of the socio- economic sphere that profoundly affects rangeland systems. Rangeland professionals commonly have an inadequate understanding of how rangeland policies are formulated and implemented. Clarifying how rangeland policy making occurs in three developing countries is the main objective of this paper. Once rangeland professionals better understand policy-making processes they could become more influential in rangeland management and pastoral development. Our main message in this paper is that rangeland advocates are typically bystanders in the policy arena for Bolivia, Ethiopia, and Kenya. Despite this daunting premise, we feel there is potential for rangeland professionals to have significant influence in all three countries. Before we delve into the main themes of this paper we will first describe some general aspects of rangelands in each country.

22

Rangelands of Bolivia, Ethiopia, and Kenya Bolivia The temperate grasslands and cold-desert scrublands of Bolivia predominate at high elevations in the Andes mountains, while warmer, more mesic ecosystems prevail at lower elevations. The temperate Andean rangelands serve as the historical home of traditional pastoral and agro-pastoral societies. Tropical humid rangelands occur in the lowlands and these have been settled only during the past few decades. They serve a relatively small, but prosperous, population of modern ranchers. Overall, the Andean rangelands make up 25% of Bolivia’s land area and are home to 16% of the total human population of 8.3 million (Alzerréca pers. obs.). Andean rangelands were a cradle of Latin American civilization, being the birthplace of Quechua, Incan, and Aymara cultures that managed ecologically diverse production systems stratified by elevation (Jetté et al. 2001). The regional dominance of Andean peoples, however, was crushed over 400 years ago by a combination of Spanish colonialism and successive republican governments that brutally suppressed the indigenous people. For example, Aymara inhabitants of the Bolivian altiplano were typically denied access to higher-quality land, modern production technology, formal education, and agricultural markets until the 1950s (Jetté et al. 2001). The policy environment has continued to slowly improve, however, for indigenous people throughout Bolivia in recent decades. This includes the passage of laws that provide the means to invest public funds in development of rural Andean communities and the democratic election of indigenous people to high public office (Jetté et al. 2001). The stability of the monetary system has also markedly improved during the past 15 years. Market integration and monetization is steadily increasing in many rural locations, but there is still a large scope for improved socio-economic integration of Andean people with the rest of the nation (Alzerréca pers. obs.). Producer prices for range crops and livestock can be low and volatile, and this is a disincentive for rural people to work the land (Coppock et al. 2001, Valdivia et al. 2001). Another noteworthy phenomenon is a pervasive emigration of youths from rural rangeland communities to major metropolitan areas such as La Paz, El Alto, and Cochabamba. This has contributed to a net population decline in some altiplano communities (Alzerréca pers. obs.). These youths are seeking educational opportunities, higher incomes, and an improved standard of living (Jetté et al., 2001). Range livestock in Bolivia include the Spanish Criollo sheep and Criollo cattle as well as improved cross-bred sheep (Merino or Corriedale x Criollo) and cross-bred dairy cattle (Holstein x Criollo) in locations having fresh water and a higher potential for cultivated forage production (Jetté et al. 2001). Indigenous camelids have generally declined in importance over the past century. Alpaca (valued for wool) and llama (valued for meat) persist in more marginal production environments at higher, colder elevations. Potatoes, small grains, and irrigated forage comprise most of the cultivated crops in the agricultural zones of the Andes (Jetté et al. 2001). Drought, frost, salinity, and resource degradation from inadequate soil management are common challenges for crop and forage production (de Queiroz et al. 2001). In the proximity of large urban areas, market opportunities for rural producers to sell livestock, dairy products, and food crops can be quite satisfactory, but this is not the case for those living in more remote areas. Poorer households may rely more on sales of fuel wood and manure to traders who transport these commodities to peri-urban locales (Valdivia et al. 2001). Food insecurity can be an issue

23 for poorer households in rural areas. In contrast, wealthier households may be well diversified where market access is good, and strong economic links between urban and rural family members enhances opportunistic economic behavior (Valdivia et al., 2001). Access to basic public services in Andean rangelands is generally poor by Latin American standards, but it is better than what occurs in sub-Saharan Africa (Coppock pers. obs.).

Ethiopia Subtropical and tropical savannas and bushlands predominate in the lowlands of Ethiopia while more mesic agro-ecosystems prevail in the highlands. Overall, rangelands make up 61% of Ethiopia’s land area and are home to about 12% of the human population of 70 million (Coppock 1994). Never colonized by European powers, Ethiopia is, however, a vestige of African empire. Governance has veered from centuries of medieval feudalism to imperialism (ca. 1890 to 1973), Soviet-style socialism (1974 to 1991), and a contemporary federal system having a mix of socialist and democratic elements. Decision-making is being decentralized from the highlands federal capital of Addis Ababa to an array of semi-autonomous states (Moris 1999). The federal government struggles to control a diverse assortment of ethnic groups and several insurgency movements. Maintenance of security is thus a high priority of the federal government, and this is focused on the relatively remote rangelands that encircle the central highlands (Desta and Coppock in press). It is notable that all of the international borders of Ethiopia occur in rangeland environments, and with the exception of Kenya and Djibouti, neighbors such as Eritrea, Somalia, and Sudan tend to be less predictable and are often hostile. As one of the world’s poorest nations, Ethiopia continues to be beset with major problems of military conflict, food insecurity, poverty, and a low capacity to deliver public services (i.e., formal education, agricultural extension, health care, etc.). Ethiopia has rugged terrain and is one of the least developed nations in terms of infrastructure, although some pockets in the southern rangelands are well served with roads and livestock market facilities (Coppock 1994). The relative lack of suitable roads and trucking throughout the nation inhibits market development and inter-regional trade, and therefore contributes to food insecurity. The semi-arid rangelands of Ethiopia continue to support relatively large numbers of people under fairly traditional means. High rates of illiteracy and few urban job opportunities mean that few pastoralists can emigrate to non-pastoral opportunities (Coppock 1994). Pockets of maize cultivation continue to spread as slow, but steady, growth in human populations gradually outstrips the capacity of traditional, livestock-based systems to provide sufficient human food in the form of milk and meat. Despite the fact that Ethiopian pastoralists have a need to procure an increasing amount of their calories from the marketplace, domestic opportunities to reliably sell pastoral livestock and pastoral livestock products are often limited (Coppock 1994). This is due to inefficient rangeland marketing systems, market saturation from highlands livestock producers, and a low purchasing power of most urban dwellers. Range livestock exports have traditionally focused on markets in the Arabian Peninsula, but market access has been erratic due in part to trade barriers (Coppock pers. obs.). Kenya may be the largest livestock export market for Ethiopia at the present time (see below). Former drought grazing-reserves in the rangelands are being undermined by population growth and grazing-induced bush encroachment (Coppock 1994). Range livestock typically include zebu cattle, fat-tailed sheep, goats, and dromedary camels. Livestock holdings vary with

24 ecological zone and ethnic group. Food crops in the lowlands are uniformly dominated by maize. Economic diversification of pastoral households away from livestock is quite rare (Desta 1999).

Kenya Like Ethiopia, subtropical and tropical savannas and bushlands predominate in the lowlands while more mesic agro-ecosystems prevail in the highlands. Overall, rangelands make up about 75% of Kenya’s land area and are home to less than 15% of the total population of 30 million people (Pratt and Gwynne 1977, Coppock pers. obs). Kenya was a colony of Great Britain until 1963 and this has had a profound influence on rangeland development and some aspects of natural resource policy (Aboud pers. obs.). Post-independence rule has been characterized by stability. Only 2 presidents (i.e., Jomo Kenyatta and Daniel Arap Moi) from the same political party have maintained control over the capitalist system for nearly 40 years. Although the central government has always been based in Nairobi, many aspects of local development administration were consigned to districts during colonial times, a pattern maintained after independence (Moris 1999). National policies are formulated and enacted by a parliament. Parliament members are democratically elected, but candidates often represent one ruling party (Coppock pers. obs.). Like Ethiopia, except for Kajiado Maasailand to the southeast, most rangelands in Kenya occur far from major urban areas, and they form an apron around a central core of highland production systems (Pratt and Gwynne 1977). Rangelands comprise most of the borders with Tanzania, Uganda, Sudan, Ethiopia, and Somalia. Unlike Ethiopia, the relations that Kenya has with its neighbors are generally peaceful and it lacks significant insurgencies (Coppock pers. obs.). Unlike Ethiopia, however, the more remote Kenyan rangelands suffer from more internal security problems such as banditry, livestock rustling, and increasingly violent ethnic conflict (J. Swift, IDS unpublished). To some degree, these internal security problems may result from internal political dynamics and organized crime (Coppock pers. obs.). Despite recent economic decline, many Kenyans appear to enjoy a considerably higher standard of living than that in neighboring countries. One consequence, is that, on average, Nairobi residents are able to pay higher prices for livestock products that residents of Addis Ababa (D. Bailey, Utah State Univ. pers. comm.). Nairobi is therefore the major terminal market for high-quality cattle reared in southern Ethiopia mainly because of this economic pull. The quality and quantity of Kenyan cattle produced in Kajiado Maasailand, or to the north in Samburu, Isiolo, Marasbit, and Moyale Districts, are apparently inferior to southern Ethiopian breeds, another reason why Ethiopian cattle are in demand in Kenya. The annual flow of Ethiopian cattle to Nairobi is currently about 50,000 head despite the risk of theft and the poor marketing infrastructure in northern Kenya. Such factors are important transaction costs for livestock traders (Barrett et al. 1998, P. Little, Univ. Kentucky pers. comm.). Many of the semi-arid rangelands of Kenya continue to support relatively large numbers of people under fairly traditional means. Kenyan pastoralists, however, have greater access to formal education and employment than their Ethiopian counterparts, so household linkages between rural and urban areas in Kenya are more common (P. Little, Univ. Kentucky pers. comm.). Like Ethiopia, dry-land cultivation is spreading in Kenya, and for similar reasons (above). Food aid is also ubiquitous in the Kenyan rangelands. Informants report that food aid

25 floods local markets, lowers crop prices, and therefore creates disincentives for farmers to produce (Coppock pers. obs.). Some observers contend that the Kenyan rangelands have been overgrazed for decades and this has resulted in widespread conversion of rangelands to less desirable forage associations and brush encroachment (D. Herlocker, private consultant pers. comm.). Others, however, dispute whether pervasive grazing impacts have occurred in arid Kenya (Ellis and Swift 1988). There seems to be more agreement that concentrated harvest of shrubs and trees for fuel and timber in peri-urban locations is the most destructive practice in the Kenyan dry lands (G. Keya, KARI pers. comm.; J. McPeak, Cornell Univ. pers. comm.). This may become more pronounced as settlements and towns grow (Coppock pers. obs.). As in Ethiopia, range livestock typically include zebu cattle, fat-tailed sheep, goats, and dromedary camels. Livestock holdings vary with ecological zone and ethnic group. Food crops in the lowlands are uniformly dominated by maize. Economic diversification of pastoral households varies with location and other factors and is generally higher than that observed for Ethiopia (Little et al. 2001).

Policy Questions The main findings of this paper are broken down into 4 categories within this section to illustrate broad aspects of policy environments as well as a few examples of policy formulation and the varied roles that rangeland advocates and rangeland science play in policy processes. The information we present merely serves as a general introduction to issues posed in this symposium. We have combined our collective, practical experiences for this paper rather than conducting a detailed analysis of the open literature. We recognize the importance of policy matters for rangeland management and development, but we admit to having limited first-hand exposure to the policy making process. We make no claim to be policy scholars.

Examples of Land Policies We divide policy matters into two broad groups: (1) those that affect the allocation or ownership of land; and (2) those that affect the management or use of land and associated natural resources. Here we will focus on the former. In general, national land policies in Bolivia, Ethiopia, and Kenya have been rarely modified to adjust for local variation in ecology. Rather, the potential productivity—and hence value—of land appears to be more important in determining patterns of land access and control. For example, arable land, regardless of whether it occurs in humid or arid zones, is usually subject to a higher degree of controlled access compared to that for grazing land.

Land Ownership and Control National policy in Ethiopia still prohibits private ownership of land, and this is an artefact from the socialist era (Desta pers. obs.). There is no indication, however, that this legal framework will be repealed. In contrast to Ethiopia, private ownership of land occurs in both Bolivia and Kenya (Aboud and Alzerréca pers. obs.). In these cases land is excised from communal (Bolivia) or public trust (Kenya) lands through various legal and illegal means. In all

26 three countries people also gain long-term control over land using leasing arrangements (Valdivia et al. 2001; Coppock pers. obs.). Access to land in Bolivia. Lands on the Bolivian altiplano are ostensibly administered through local government districts called cantóns. Cantóns encompass pastoral or agro-pastoral communities that—in many cases—have persisted for hundreds of years (Jetté et al. 2001). In some cantóns higher-value croplands have been effectively privatized since the 1950s when a revolutionary government instituted land reform that gave control of such parcels to the current users. These parcels were not deeded, but control has been kept within family units and plots are passed down from one generation to the next. Other transfers of arable lands require approval by local community councils (Jetté et al. 2001). In contrast to croplands, lower-value grazing lands typically retain the traditional character of open-access and shared communal use. There are some wrinkles in the system, however, that help promote community welfare and flexible resource use. For example, access to cropland and livestock can be shared following customary rules—this allows households with more labor and those with more land or animals to combine resources and increase efficiency. Control of land can shift depending on weather patterns and resource pressure. During times of drought communal use of all lands can increase, and vice versa (Valdivia et al. 2001). Access to land in Ethiopia. Ethiopian rangelands are ostensibly administered at the woreda (district) level, and collaboration may occur with traditional pastoral organizations in some instances (Desta pers. obs.). Despite the fact that land cannot be owned per se, higher potential land suitable for cultivation or fodder reserves can be annexed in the traditional system when land users demarcate claims by investing in land preparation and bush fencing. For high- potential, arable land near settlements and towns, urban dwellers and farmers can gain controlled access via leases arranged through local political channels (Desta pers. obs.). Lower-value grazing lands typically retain the traditional character of open access and shared use, but local patterns are mediated by the proximity of sites to water and the presence of hostile neighbors (Coppock 1994). Access to land in Kenya. Kenyan rangelands are ostensibly administered by County Councils, and collaboration may occur with traditional pastoral organizations in some instances (Aboud pers. obs.). Unlike Ethiopia, in the semi-arid zone of Kenya (i.e., Kajiado Maasailand) there were large-scale experiments with group ranches and similar organizations since the 1960s. Group ranches were supposed to improve stewardship of natural resources by assigning group rights and responsibilities for group titling of land (Pratt and Gwynne 1977, Aboud pers. obs.). While generally considered a failure in terms of the socio-economic transformation of pastoralism and improvements in land management, group ranches have been considered successful in terms of helping pastoralists retain control over higher-value rangelands in the face of encroachment from farmers and other non-pastoralists (Galaty 1992, Kimani and Pickard 1998). In recent years some group ranches have been reportedly subjected to rampant land grabbing and subsequent resource fragmentation. Parcels are sold to outside interests by influential group-ranch members. This process may threaten the ecological integrity of Kaiado Maasailand because of the privatization of key resources and subsequent impediments to wildlife migration. Similar patterns are occurring on Marsabit Mountain in northern Kenya where expansion of farming is usurping the former dry-season grazing lands of pastoralists. Land sales to cultivators through the Marsabit County Council are brisk (Coppock pers. obs.). The rapacious political culture currently found in Kenya is reportedly fueled, in part, by uncertainty regarding

27 political succession following the Moi regime (Coppock pers. obs.). The remaining Kenyan rangelands that continue to be dominated by pastoralists typically retain the traditional character of open access and shared use, but as in Ethiopia, local patterns are mediated by proximity of any given site to water and the presence of hostile neighbors.

Policy Making and Roles of Rangeland Professionals and Rangeland Inhabitants Bolivia In Bolivia, policy issues regarding use of land, water, and forest resources are reviewed at regional and national levels through public fora that elicit voluntary input from citizens, special interest groups, professionals, and representatives of non-governmental organizations (Alzerréca pers. obs.). Policies are then articulated and approved through a democratically elected congress. Although this process has merits, one flaw is that the Andean rangelands seem to be notably absent from the resource use dialogue. This is primarily because the Andean rangeland population is economically and politically weak (Alzerréca pers. obs.). Rangeland scientists and traditonal producers also have little influence in policy debate. It is therefore not surprising that when a range-related policy happens to be enacted, it is typically too general or simplistic for effective local implementation (Alzerréca pers. obs.). Many land-use decisions are made on the basis of political expediency. If any pastoral group has influence it is the wealthier, mixed-race ranchers of the tropical lowlands. These producers form effective special interest groups to influence policy in order to protect their privileged lifestyles (Alzerréca pers. obs.).

Ethiopia In Ethiopia, policy regarding use of natural resources is increasingly formulated by the semi-autonomous states (Moris 1999, Desta pers. obs.). The politicians who deal with policy matters may be democratically elected, but they come from a restricted slate of candidates. Despite a dearth of public funds, recently formed state governments still appear committed to extending their influence even in the most remote locations (Moris 1999). Land-use decisions appear to be governed by political expediency, however, as comprehensive land-use planning guidelines are often non-existent (Coppock 1994; Desta pers. obs.). Compared to Bolivia or Kenya, the relative lack of a well-educated public, a scarcity of effective special-interest groups, and an absence of formal democratic traditions means that policies in Ethiopia are crafted in more of a top-down fashion (Coppock pers. obs.). Despite the lack of a true grass-roots democracy, local politicians often support resource-use decisions that favor popular, short-term goals of the masses that exploit the environment with little concern for proper resource management or conservation (Desta pers. obs.). Examples include instances where local administrators promote risky cereal cultivation on fragile rangeland soils or allow pastoral livestock to graze unhindered in gazetted conservation areas (Coppock pers. obs.; Desta pers. obs.). The input of range professionals in such policy-related decisions is minimal, even in cases where local professional expertise is easy to solicit (Desta pers. obs.). Simply put, rangeland management is not seen in the Ethiopian political culture as a means to rapidly create more food or wealth to solve acute social problems, and as such the range profession has little political influence compared to that held by agriculturalists or development advocates.

28 Kenya In Kenya, formulation of resource-use policy at the national level is in the hands of legislators who have often been democratically elected from a restricted slate of candidates. Special interest groups, lobbyists, and non-governmental organizations are increasingly involved in national policy debates and have some influence on policy making (Aboud pers. obs.). Interpretation, adaptation, and enforcement of natural resource policies, however, appear to be decentralized to district levels, a trend evident since independence (Moris 1999). Kenya, in contrast to Ethiopia, has enjoyed a relatively long period of political stability, and this will hopefully promote more wisdom and continuity in resource planning and management. Kenya has other potential advantages compared to Ethiopia in terms of: (1) a better-educated citizenry with generally higher incomes and better living standards; (2) a larger cadre of well-trained range professionals; and (3) more experience with formal, grass-roots democratic traditions (Moris 1999; Coppock pers. obs.). Unfortunately, however, acute fiscal constraints, a lack of government commitment, an increasingly rapacious elite culture, and a drying-up of support from external donors have all contributed to a marked contraction of public service delivery in Kenya (Moris 1999; Coppock pers. obs.). This is most readily observed in remote rangeland areas. Land-use decisions in rangeland areas appear to be mostly governed by political expediency and the greed of influential people (Aboud pers. obs.). Land-use decisions favor short-term gains for elites and the populace at large. For example, risky cultivation is common on Kenyan rangelands and deforestation of ostensibly protected trust lands for fuel wood production, charcoal production, and timber is locally acute in many instances (Coppock pers. obs.). Rangeland professionals may channel technical information and recommendations to policy makers via roles as resource persons for special interest groups, lobbyists, and non- governmental organizations. The relative influence of rangeland professionals or rangeland inhabitants on policy processes, however, remains small in Kenya. Short-term economic concerns clearly override all other considerations for land policy decisions (Aboud pers. obs.).

Rangeland Policy Trends—Promotion of Opportunism or More Control? It is evident from the previous section that whether one is talking about Bolivia, Ethiopia, or Kenya, rangeland professionals and rangeland inhabitants have a very long way to go in their quest to be more influential in the policy arena. In light of this finding, a key question posed by the organizers of this symposium—namely, whether policy makers have embraced new paradigms in range ecology—seems clearly moot in our view. We feel that range advocates first need to fight hard simply to get a seat at the policy-making table. After that happens we can concern ourselves about the theoretical details. Work by Ellis and Swift (1988) and others advocated that the dynamics of arid rangelands are controlled to a very large degree by external phenomena like precipitation variability. They labelled such systems as “non-equilibrial.” This is in contrast to “equilibrial” systems where internally generated phenomena like grazing pressure have a greater influence on attributes such as vegetation change, herbivore population dynamics, etc. Work by Ellis and Swift (1988) and others stimulated great interest among range professionals in revising range management paradigms. This is because non-equilibrium theory has important implications for undermining the utility of concepts like carrying capacity and range trend. More recently, however, other researchers have challenged non-equilibrium theory (Fernandez-Gimenez and

29 Allen-Diaz 1999; Illius and O’Connor 1999; Desta and Coppock in press). Despite on-going debate, we feel there is little doubt that work by Ellis and Swift (1988) and Westoby et. al. (1989) stimulated a renewed appreciation among range professionals that rangeland dynamics are not deterministic but rather characterized by risk and uncertainty. We suspect there is a professional consensus—regardless of labels like “equilibrial” or “non-equilibrial”—that pastoral production systems require various degrees of mobility, flexibility, diversity, and opportunism to be successful. This is the perspective from rangelands that makes them unique relative to other agro-ecosystems. Risk and uncertainty is the converse of the deterministic thinking that pervades intensive production agriculture, and it probably also influences the world of policy making. The major contribution of range professionals, when they get to the policy table, is to convey the need for investments and interventions for rangelands and range-based societies, which promote wealth accumulation and sustainable production through improvements in risk management. Ironically, for the land-use trends we observe on the rangelands of Bolivia, Ethiopia, and Kenya, they appear to be the antithesis of what pastoral societies need to better manage risk. The best example is the gradual loss of key rangeland resources, whether it occurs from annexation by external entities or from internally driven, de facto privatization. There seems to be a lack of policies (and/or lack of enforcement) that prohibits such trends from occurring. Similarly, in Ethiopia we continually hear that federal policy makers aspire to settle all pastoralists in the country to promote their development (Coppock pers. obs.). Settling pastoralists has been attempted many times in sub-Saharan Africa, and the results have often been disastrous. This is primarily because mobility of livestock herds is compromised and settled producers become more vulnerable to drought and localized resource degradation (Coppock pers. obs.). Conversely, actions that help open markets and stimulate trade can benefit rangeland people in some instances. One case is a recent situation where efforts have been undertaken to stimulate cross-border commerce between Kenya and Ethiopia. We see this as probably beneficial for pastoralists in both countries (Desta and Coppock pers. obs.).

Increasing the Influence of Range Professionals in Policy Making Despite the apparent lack of influence that range professionals have in the policy- making process in the three countries we reviewed, we believe that this situation can be improved. The status quo can be blamed on rangeland professionals to a high degree. A major factor, in our view, is simply that most range professionals in the past have not been concerned enough about whether their technical research would be translated into transferable technology or innovative policy down the road. We believe this has resulted because many rangeland scientists—first and foremost—have their greatest affection for the biophysical aspects of rangelands and less interest in the welfare of people who must make a living from rangeland resources. We suspect the tradition has also been for range scientists to see the application of research results as out of their purview—namely, that technology transfer or policy implications would simply be “handled by someone else” down the line. Those of us interested in the real-world impact of our work no longer believe that research and development in the range profession really works that way. Facilitation is needed. We believe that rangeland scientists must be far more active outside of the traditional research sphere if they hope to have a positive influence on policy. Those of us working in East Africa—in particular—have found that if we make more concerted and sustained efforts to identify policy makers and transmit useful information, they are accessible

30 and willing to listen. In our experience the typical policy maker in Bolivia or East Africa knows little about the unique aspects of rangelands or the special needs of rangeland inhabitants. The process we envision means that not only do range professionals need to be more aggressive in linking themselves to policy makers, but they must also be more clever about the form and content of messages being conveyed. Rangeland development concepts must compete for attention with those from other sectors and special interests. Range has a disadvantage because it is challenging to make range issues simpler for laypersons to understand. It is also more of a struggle to convince policy makers that investment in the smaller human populations that inevitably reside in rangelands can yield economic returns comparable to those made in other sectors with much higher levels of base productivity or having considerably larger populations. Our belief is that planners and policy makers put a heavy weight in the decision- making process on comparing economic returns from different investments. Economics, therefore, is probably not the trump card for rangeland advocacy. A different focus that includes more emphasis on humanitarian concerns, poverty reduction, and investment to promote security could be useful. Finally, we see that rangeland scientists in academic or governmental circles probably still have few tangible incentives for engaging in policy outreach other than self- satisfaction. In academics numbers of research publications and teaching success is what is typically rewarded in terms of tenure and promotion. Technology transfer has been traditionally identified with extension agents. Policy outreach falls into a nebulous zone between research, extension, and teaching. Therein lies part of the problem as to why it has been relatively ignored.

Conclusion In all three countries covered, rangeland inhabitants comprise a small minority of the national populations. The inhabitants of rangelands are largely devoted to extensive livestock production using traditional means. Compared to other sectors of the national economies, the contributions of rangeland production systems are relatively weak. Political and economic power tends to be held by groups that come from places other than rangelands, which contributes to the marginalization of rangeland societies and relegates rangeland professionals and advocates to the “back benches” of policy influence. Rangeland inhabitants are in danger of being marginalized even more if their interests cannot be better protected through policy channels. This seems especially true where key resources are being systematically lost due to a variety of factors internal and external to pastoral systems. Rangeland professionals must shoulder most of the burden of becoming more effective communicators with policy makers. This can create a win-win situation because demonstrable impact via policy channels helps rangeland professionals justify their existence in an era of greater emphasis on accountability. Tenure, promotion, and other reward systems should be altered to value time spent by rangeland scientists on policy outreach. Rangeland professionals need to be more creative in their attempts to communicate with policy makers. While economic returns from rangeland investment may not be competitive with those from other sectors, other societal values can be brought to the table. We need to be cognisant of the societal costs of not investing in rangelands given the increasing role of rangelands as safe havens for poverty and various social ills. It has been recently noted, on a global scale, that the dominant multi-lateral donor activities on rangelands—especially those in sub-Saharan Africa and Central Asia, but also pockets of the Andes—have shifted from

31 development investment to a mode of chronic relief and rehabilitation over the past 40 years. This bottomless pit of crisis mitigation has emerged precisely because rangelands and rangeland inhabitants have been undervalued and ignored by decision-makers in the past. The long-term costs of relegating rangeland societies to the “waiting room of development” are becoming more apparent today. Rangeland professionals need to be more ambitious about the positive impacts they hope to have on society at large. Rangeland scientists, in particular, need to alter old assumptions about how they should allocate their professional efforts. We believe that rangeland professionals can increase their effectiveness to society-at-large in terms of: (1) making policy makers and the public more aware of the importance of proper resource use and management on rangelands via information distributed in the popular media; (2) making a stronger effort to disseminate simplified and eye-catching summaries of technical research (and research implications) to policy makers; (3) attempting to educate policy makers on the unique virtues and attributes of rangelands; (4) increasing personal participation in local, regional, and national decision-making bodies—in both the public and private sectors—that deliberate on natural resource issues; and (5) providing professional services to lobbyists and special interest groups. While some aspects of this approach may seem beyond the realm of extending results of objective research, we emphasize that simply providing information to more types of people in more compelling ways does not equate to a biased advocacy. It is also useful to remind ourselves that the rangeland profession is in stiff competition with others to gain the attention and favors of policy makers. If the competition is engaged in such activity, so should rangeland professionals.

Acknowledgements Some of the research cited in this paper was conducted under the auspices of the Small Ruminant Collaborative Research Support Program (SR-CRSP) or the Global Livestock Collaborative Research Support Program (GL-CRSP). Both programs received funding from grant DAN-1328-G-00-0046-00 or grant PCE-G-98-00036-00, respectively. Both grants have been administered under the auspices of the Office of Agriculture and Food Security, Global Bureau, United States Agency for International Development (USAID). Opinions expressed in this paper are those of the co-authors and do not necessarily reflect the views of USAID.

Literature Cited Barrett, C., P. Little, D. Bailey, F. Chabari, and K. Smith. 1998. How might infrastructure improvements mitigate the risks faced by pastoralists in arid and semi-arid lands? In: S. Johnson (ed.) Ruminations—Newsletter of the SR/GL-CRSP. University of California, Davis. Fall Issue: 1, 10, 12-13 Coppock, D. L. 1994. The Borana Plateau of Southern Ethiopia: Synthesis of Pastoral Research, Development, and Change, 1980-91. Systems Study No. 5. International Livestock Center for Africa, Addis Ababa, Ethiopia. 374 p. Coppock, D.L., C. Valdivia, J. Yazman, C. Jetté, J. de Queiroz, L. Markowitz, and I. Ortega. 2001. Conclusions and recommendations. In: D.L. Coppock and C. Valdivia (eds.) Sustaining Agropastoralism on the Bolivian Altiplano: The Case of San José Llanga. Dept. of Rangeland Resources, Utah State University, Logan, Utah. 284 p.

32 de Queiroz, J., D.L. Coppock, H. Alzerréca, and B. Norton. 2001. Ecology and natural resources of San José Llanga. In: D.L. Coppock and C. Valdivia (eds.) Sustaining Agropastoralism on the Bolivian Altiplano: The Case of San José Llanga. Dept. of Rangeland Resources, Utah State University, Logan, Utah. 284 p. Desta, S., and D. L. Coppock. In press. Cattle population dynamics in the southern Ethiopian rangelands, 1980-97. J. Range Manage. Vol:pp??? Desta, S. 1999. Diversification of Livestock Assets for Risk Management in the Borana Pastoral System of Southern Ethiopia. PhD Dissertation. Dept. Rangeland Resources, Utah State Univ., Logan, Utah. 189 p. Ellis, J.E., and D.M. Swift. 1988. Stability of African pastoral systems: Alternate paradigms and implications for development. J. Range Manage. 41:450-459. Fernandez-Gimenez, M., and B. Allen-Diaz. 1999. Testing a non-equilibrium model of rangeland vegetation dynamics in Mongolia. J. Applied Ecology. 36: 871-885. Galaty, J. 1992. Social and economic factors in the privatization, sub-division and sale of Maasai ranches. Nomadic Peoples 30:26-40. Illius, A.W., and T.G. O’Connor. 1999. On the relevance of nonequilibrium concepts to arid and semi-arid grazing systems. Ecological Applications. 9(3):798-813. Jetté, C., H. Alzerréca, and D.L. Coppock. 2001. National, regional and local context. In: D.L. Coppock and C. Valdivia (eds) Sustaining Agropastoralism on the Bolivian Altiplano: The Case of San José Llanga. Dept. Rangeland Resources, Utah State Univ., Logan, Utah. 284 p. Kimani, K., and J. Pickard. 1998. Recent trends and implications of group ranch sub-division and fragmentation in Kajiado District, Kenya. Geographical Journal 164(2): 202-213. Little, P., K. Smith, B.A. Cellarius, D.L. Coppock, and C. Barrett. 2001. Avoiding disaster: diversification and risk management among East African herders. Development and Change 32:401-433. Moris, J. 1999. Under Three Flags: the Policy Environments for Pastoralists in Ethiopia and Kenya. SR/GL-CRSP Pastoral Risk Management Technical Report No. 03/98. Dept. of Rangeland Resources, Utah State University, Logan Utah. 119 p. Pratt, D., and M. Gwynne. 1977. Rangeland Management and Ecology in East Africa. Hodder and Stoughton, London. 310 p. Smith, K., C. Barrett, and P. Box. 2000. Participatory risk mapping for targeting research and assistance, with an example from East African pastoralists. World Dev. 28(11):1945-59. Valdivia, C., C. Jetté, L. Markowitz, J. Céspedes, J. de Queiroz, C. Murillo Quiroga, and E. Dunn. 2001. Household economy and community dynamics at San José Llanga. In: D.L. Coppock and C. Valdivia (eds) Sustaining Agropastoralism on the Bolivian Altiplano: The Case of San José Llanga. Dept. Rangeland Resources, Utah State Univ., Logan, Utah. 284 p. Westoby, M., B.H. Walker, and I. Noy-Meir. 1989. Opportunistic management for rangelands not at equilibrium. J. Range Management. 42:266-274.

33 Rangeland Policy in South Africa: Drivers and Passengers

DAVID GROSSMAN

Author is Private Consultant, P.O. Box 29038, Sandringham 2131, South Africa

Introduction Rangelands cover some eighty percent of the surface of South Africa, ranging from the desert in the northwest, through the arid succulent karoo, the winter rainfall region of the southwest Cape with its unique sclerophyllous (fynbos) vegetation, the central grasslands, patches of closed canopy forest, to the vast savanna woodlands which stretch northwards to central Africa. Rangelands contribute a wide array of goods and services to the local economy, including livestock production, game ranching, magico-medicinal plants and water, and form the basis of a tourism industry centered on national parks, game reserves and scenic landscapes. The purpose of this symposium is to debate the role of rangeland professionals in developing and implementing rangeland policies. In this paper I reflect on the South African experience in policy formulation, by means of a brief overview of government interventions over the past century or so, and analysis of the drivers behind these interventions. Obviously, any policy is informed and influenced by prevailing macro-policy of the government of the day. South Africa provides an example of a situation where, in essence, three different political ideologies and macro-policies prevailed since Union in 1910. These range from the post-colonial period, where the Union of South Africa was a member of and influenced by the British Commonwealth, through 40 years of institutionalized Apartheid (or separate development as it was termed by its proponents) and increasing international isolation, to the present era of a democratic and extremely diverse society, integrated into the international mainstream. Whereas this may imply that policy processes and any conclusions to be drawn from the South African situation may have limited application in other situations, I contend that there are certain underlying fundamentals that could apply to any or most situations internationally.

How are rangeland policies formed and how do rangeland professionals influence the process? An historical overview

From Union to Apartheid- 1910-1948 The Union of South Africa was formed by the amalgamation of the Boer (farmer) Republics of the Transvaal and Orange Free State, the Cape Colony and Natal, in 1910, nine years after the Anglo-Boer war. The new Union was a member of the British Commonwealth, and accordingly strongly influenced by colonial policies and expatriate experts.

34 Concern about the perceived degradation of the country’s rangelands is a recurring theme in the annals of South African rangeland and political history, and has been a major factor driving policy formulation. Before the arrival of European immigrants, rangelands were occupied by a wide array of indigenous wild herbivores, and, since at least 200 AD, by the domesticated herbivores of early indigenous nomadic Khoi-San pastoralists. Permanent settlement by Bantu people (African) and, after 1650, by Europeans, and the resultant general replacement of indigenous species with cattle, sheep and goats, was widely held to have resulted in changes in herbaceous species composition, accelerated soil erosion and phenomena such as bush encroachment (Hoffman et al 1999). Initial concerns were focused on the Karoo, a vast semi-arid basin in the center of the country covered by a combination of grasses and low scrub, where virtually the whole area was divided into individual farms, and extensive grazing by introduced sheep predominated. As early as 1875, Shaw stated “…the overstocking of farms has changed the flora, introduced and given undue influence to a worse herbage, and bids in fair time to change the climate.” He further stated that the plants of the Karoo commenced to travel northwards and added their energies to the extirpating of the indigenous and proper flora of the region” (i.e., the grasslands adjoining the semi-arid Karoo scrublands. Periodic droughts were considered to have exacerbated the situation. As early as 1914, it was reported that “the occupiers of the land, from the earliest times to the present day, have gradually destroyed the vegetation, whether trees, shrubs or grasses, which formed the natural protection of the soil” (Report of a Select Committee on Droughts, Rainfall and Soil Erosion 1914). According to the findings of this report, the denudation referred to was caused by “veld- burning, felling of trees for fuel or timber, and by the grazing of livestock”. During 1920, the then Government appointed a “Drought Investigation Committee”, which concluded that “the practice of overstocking farms is very prevalent, which leads to overgrazing and all its attendant evils” (Final Report, Drought Investigation Committee, 1923). The Committee seriously considered recommending legislation to eliminate overstocking, but concluded that guidance would be better than coercion (a recurring theme in rangeland policy). It is interesting to note that this series of investigations coincided with a period of below average rainfall in the early part of the twentieth century, and, importantly, with the period of influence of Jan C Smuts, former Boer General, Prime Minister, founder of the League of Nations (precursor to the UN of today) and botanist of note. Smuts was influenced by activities in other former colonies, by members of the British Commonwealth, and by such experts as Phillips, Bews, Pole-Evans and Marloth. Marloth had been Smuts botany professor and mentor at University, where he studied law but took courses in Botany, and a lifelong relationship of trust and respect developed. The Welshman, Pole-Evans, an expert in rangelands, became Smuts neighbor, advisor and confidante (Beukes 1996). However, the net result of these activities was a “period of inactivity," punctuated by a further severe drought during the 1930’s, and national focus on the Second World War. Shortly after the cessation of hostilities, under the Premiership of Smuts, a progressive Soil Conservation Act was promulgated in 1946. Smuts’ friend and neighbor, Dr I. Pole-Evans, was appointed head of the fledgling rangeland research division of the Department of Agriculture. A series of experimental farms was established in several bio-climatic regions, primarily to evaluate the effects and efficiency of grazing systems (Beukes 1996).

35 Although Smuts can be regarded as the main driver behind these policy interventions, it is equally clear that he was significantly influenced by international trends and by rangeland professionals of the day. Indeed he had ongoing contact and even accompanied many of them on field trips over much of South Africa, as far afield as present day Tanzania. The point is that in Smuts, we find an influential, senior politician with a lifelong passion for the rangelands and flora of the country, with close and mutually respectful associations with the most senior rangeland professionals of the day. Clearly this is a period where the rangeland professional can be regarded as a co-driver with the politician, a fortunate coincidence that undoubtedly impacted on policy formulation.

The Republican Era, the Farmers “Reign”, From Apartheid to Democracy- 1949-1994 After the 1948 referendum among the white electorate, 40 years of formal Apartheid rule ensued under an Afrikaner Nationalist Government. As a result of some creative gerrymandering of electoral divisions, the mainly Afrikaner rural land owning class were undoubtedly the most influential political group among the minority electorate. During this period, Prime Minister, Dr H F Verwoerd, drafted the grand plan for racially separate development. This led to a formal dichotomy in state policy towards rangelands. On the one hand entrenchment of an essentially white commercial ranching sector, on some 80% of the land. On the other, the development of “homelands” for the majority, black Africans, on some 14% of the land. Policy interventions in the commercial areas included subsidies for soil conservation works, farm dams, fencing, subsidized rail rates, special credit facilities, extension services and tax relief (Anon 1998). The establishment of rotational grazing systems was also subsidized, as the benefits of these systems became apparent on the rangeland research stations established around the country. In the homeland areas, a policy of “betterment planning” was followed. In essence, this involved land use planning and the large-scale resettlement of people in villages, with areas of rangeland designated for livestock grazing. These schemes ignored the indigenous systems of range and livestock management, developed over centuries. Verwoerd engaged the services of an agricultural expert, Professor Tomlinson, to develop the technical planning of the grand apartheid scheme, including use of the rangeland areas. Sometime after Verwoerd’s assassination, Tomlinson stated that he had tried to convince Verwoerd that the situation in the homeland areas was not sustainable from economic or environmental perspectives. It is apparent that Tomlinson recognized that rangelands of the homeland areas would simply not be able to support the numbers of livestock. However, the prevailing political ideology of the day simply overrode this reality. At the same time, several rangeland professionals were involved in the development and implementation of “betterment schemes” in the homelands, whilst many held the view that these were simply not sustainable. There was however, no formal concerted effort to challenge prevailing policy, indeed most rangeland professionals were employed in Government service and simply implemented prevailing policy. The odd dissenting voice usually emanated from academic circles, and had little influence on policy. For example, a leading critic of Apartheid policy, Eddie Roux, Professor of Botany and author of an influential book on Grasslands, was in fact listed as a communist and banned by the Nationalist Government, a common practice in dealing with those opposed to prevailing policy and ideology.

36 Despite the overriding influence of macro-policy, concern about rangeland degradation manifested itself by means of several further government interventions during this period. During the 1950’s, prominent scientist John Acocks had again warned about desertification and the advance of the semi-arid karoo scrub into adjacent grasslands (Acocks 1953). The publication of his maps showing the predicted advance of the Karoo influenced public opinion as well as government thinking. Interventions included the government appointed Commission of Enquiry into Agriculture 1968-1972, which once again followed a severe drought experienced during the 1960’s during which a prominent rangeland professional warned of irreversible desertification (Anon 1972). Rangeland degradation received prominent mention in the report of this Commission. Verwoerd led South Africa out of the Commonwealth and a Republic was proclaimed in 1961. During the 1960’s and 1970’s, international attention focused on South Africa’s domestic policies and a series of trade, travel and other sanctions were gradually imposed on the country. Government policy towards agriculture in general shifted to one verging on “produce at all costs”, but again, concern about rangelands continued. A Committee for the Control of Veld Deterioration in the Karoo and Adjoining Areas, reported in 1980 and a Committee on Pastures tabled its report in 1981. As a result of these initiatives, a radical stock reduction scheme was introduced between 1969-1978 (Baard 1978). Also, in 1983, a progressive Conservation of Agricultural Resources Act (Act 43 of 1983) was promulgated. The regulations passed under this Act demonstrate an unequivocal commitment to the conservation of the natural rangelands and soil, but excluded the so-called independent homeland areas. Also, perhaps because of the influential role of the farming community in government support, persuasion rather than coercion applied, despite clear provisions in the Act for punitive measures in the case of damage to the natural resource base through mismanagement. In 1985, the National Grazing Strategy was announced. Its main objective was to “use, develop and manage the natural and cultivated pastures in the Republic of South Africa in such a way that the present generation gains the greatest sustained benefit, while the production potential must be retained to satisfy the needs and aspirations of future generations” (Anon 1985). Once again, prominent rangeland professionals of the day were associated with the interventions mentioned above, with the drafting of the Act and with the development of the National Grazing Strategy (NGS). For example, the draft NGS was widely circulated amongst rangeland professionals who were afforded the opportunity of comment and input. Notably absent in the objective recorded above, is any reference to the by-then nominally independent homelands (thus excluded, by definition, from the rest of South Africa). By this time, South Africa was a deeply divided country, with laws in place to effectively silence critics of Apartheid policy, and incentives to reward supporters. Less attention was paid to international trends. Ministers and their advisers were all drawn from the ranks of the Afrikaner Broederbond (brotherhood), a supposedly secret organization that heavily influenced Government policy. Very few, if any, consultative or participatory policy processes took place, and where this did happen, as for example in a national effort to address bush encroachment, proceedings and ultimate formulation of policy recommendations were still dominated and controlled by government supporters. As such, quite simply, supporters of Apartheid, including the influential white farmer lobby, became the drivers of policy, and dissenters the passengers.

37

The Outbreak of Democracy, 1994- With the advent of a democratically elected, non-racial Government in 1994, and the formation of a new Cabinet by President Mandela, government priorities lay with redressing inequities and inherited problems. A comprehensive Reconstruction and Development Policy was formulated, with issues such as health, education, poverty alleviation, provision of essentials such as potable water etc receiving highest priority. At the same time, Government saw agriculture, including range use, as an important component of rural economic development, particularly in the former homeland areas. The new National Constitution guarantees the right to a safe environment and since 1994, South Africa has been admitted to international bodies from which it had been formerly excluded. The Government participated in and ratified several major international conventions, including the Convention on Biodiversity and the Convention to Combat Desertification The Government established a policy think-tank, the Land and Agriculture Policy Center, to assist in policy formulation. This center included several expatriate experts and, obviously, supporters of the majority African National Congress, many of whom had been in exile during the Apartheid period. A process of policy formulation was initiated in order to develop a policy on agriculture and land use that would at once build an efficient and internationally competitive agricultural sector, support the emergence of successful smallholder farmers (particularly in the former homeland areas), and at the same time conserve the natural resource base and put in place policies for sustainable resource use (Hanekom, 1998). A fully participatory process was adopted, on a scale not previously experienced in South Africa. Provincial Governments, each of which by now had a Provincial Department of Agriculture, organized agriculture (the various agricultural and farmers unions and associations), the parastatal Agricultural Research Council, Universities, non-governmental organizations, farmers, rural people and civil society at large were called on to participate in the process, either by participation in working groups, workshop attendance, submission of content or by reactive comment on drafts. Importantly, the “learned” scientific societies were formally included, with the Minister, for example, attending and participating in a seminar organized by the Grasslands Society of Southern Africa, as well as addressing the annual congress of the Society. This “experiment” in democratic process understandably resulted in a long, drawn out process, with growing impatience in some quarters (cynics coined the phrase “participation paralysis”). It also, for the first time since the period of international isolation, exposed South Africans to the direct inputs of foreign, donor-funded expatriates who assisted in drafting policy documents. Recognizing the identified rangeland problems, the government introduced two national programs aimed at reversing degradation, the Working for Water and LandCare programs. The first, aims to at control invasive alien species, the latter to promote sustainable rangeland use. Both programs were instigated after scientists conducted detailed preparatory work, especially for Working for Water. Studies were undertaken to quantify the nature and extent of the invasive alien problem, and to quantify the benefits that would accrue, particularly from increased quality and quantity of water production from catchments (e.g. MacDonald and Wissel, 1989). The resulting information, coupled with the politically attractive fact that thousands of jobs would be created and skills transferred to rural people, led to the implementation of the project. An example of a harmonious interaction between science and politics!

38 To date, the democratic era seems characterized by a situation where politicians remain the primary drivers of policy, but where rangeland professionals with credibility are certainly brought in to the policy formulation process.

The Roles of Science and Politics It is clear that South Africa has always been a highly politicized society and that macro- political considerations tend to override what are considered to be more parochial issues. Nevertheless, concern about rangelands has influenced policy throughout history. As we have seen, scientists were particularly prominent in policy formulation during the Smuts era, essentially because Smuts himself was passionate about the issue, and a scientist of note (Einstein is reputed to have said that he included Smuts as one of a handful of people who intuitively understood the theory of relativity at the time). However, scientific input has not always received such a warm reception, particularly where this was in conflict with macro-policy, as in the case of the homeland situation. Similarly, indigenous knowledge systems and customary law, which regulated the grazing of livestock in so-called communal areas, have been all but ignored in the formulation of policy in the past. In analyzing the roles of science and politics, a key factor that emerges is the question of credibility. In the first instance, this relates to the political response to “doomsday” scenarios. John Acocks’ dire scenarios about Karoo expansion have not really materialized to anywhere near the extent predicted. Predictions about desertification by prominent professionals during the prolonged drought of the 1960’s were quite simply wrong. Veld condition on farms where stock numbers were reduced under the large-scale stock reduction scheme was not significantly different from that on farms which had not participated in the scheme, when evaluated after a series of good rainfall years during the 1970’s (Hoffman et al 1999). Secondly, changing paradigms, including debate on non-equilibrium systems, as well as debate about the relative merits of rotational vs. non-rotational grazing management systems, whilst stimulating to the scientist, confuse the politician who needs simple epigrammatic answers to important questions. Historically, perceptions of rangeland degradation and the underlying dynamics have been rooted in an orthodox view of rangeland function, which view has been challenged by the “new thinking” espoused, amongst others, by Behnke et al 1993. Almost predictably, this “new orthodoxy” has in turn been challenged, particularly under conditions of higher rainfall. So the debate continues and politicians wait whilst answers are sought. The reality is that the time scale of scientific enquiry, the intervals between new paradigms or even “quantum leaps” in knowledge, rarely if ever matches the time horizon of political survival in the world of realpolitik. Political vagaries exacerbate the situation and, for example, there are cases of political administrations simply reneging on international protocols or local legislation where this is in conflict with national priorities. The net result of all this is that whilst there is often a benign political tolerance, there is only rarely fervent political support for, or reliance on, the rangeland professional insofar as immediate policy formulation is concerned. The Working for Water program mentioned in the previous section provides an example where the output of the scientist (quantified societal benefits to be derived from controlling invasive alien plants) happily coincides with the

39 requirements of the politician (job creation, fulfillment of legal obligations etc). This particular program also illustrates the importance of a scientific “champion” to lobby and bring to the attention of government the benefits to be derived. Working for Water was characterized by such an individual, a scientist who persistently lobbied until the program was effectively implemented. Also, the importance of “instant gratification”, where results of policy are seen immediately, not subject to a time lag between implementation and reaping rewards! However, it is not easy to find much other evidence of active lobbying by the individual scientist, nor the various professional bodies, many of whom apparently wait to be summoned rather than proffering services. In South Africa, the learned society representing rangeland professionals furthermore suffers from a minor crisis of identity. The Grasslands Society of Southern Africa has been home to rangeland professional since the 1960’s but has never really been formally used in an advisory capacity by the political sector. The very name has sown confusion in the ranks of political leadership, and there has been debate in the past as to whether the term “Grassland Society” conveys the correct image of this body. Indeed, many will have noticed that the name of the scientific journal has been changed to the African Journal of Range and Forage Science, more accurately reflecting the main spheres of activity. Perhaps such societies need to emulate the successful lobby and pressure groups, such as the international animal rights organizations, which completely professionalize the art of advocacy and lobbying. How many professional scientific organizations engage the services of full time professionals and actively enter the advocacy arena? A final thought. In South Africa, the invisible hand of market forces has caused dramatic changes in rangeland use in the past ten years or so, more dramatically than any explicit government policy. As subsidies for the white commercial ranching sector were phased out, in line with new policy, a dramatic increase in game ranching occurred, although this was never an explicit or tacit objective of government policy. Since the 1960’s, various professionals and academics extolled the relative advantages of multi-species indigenous herbivore production systems (game ranching) as compared to domestic livestock systems, but this form of land use only really took off some 30 years later, more in response to market demand than to scientific persuasion about the benefits from an ecological point of view. In fact, the current government in South Africa has been known to regard game ranching as the “hobbyist pursuit” of a somewhat elitist group of largely absentee landlords (views expressed privately rather than official policy).

The Roles of Centralized vs. Decentralized Institutions Historically, centralized institutions have dominated policy formulation, although implementation has been influenced by local peculiarities. With diverse bio-climatic regions, ranging from arid areas receiving 50 mm of rainfall to high altitude grassland and forest receiving in excess of 2 m of precipitation annually, local needs vary. However, the national Department of Agriculture has always had a centralized national headquarters, with regional structures in the former four provinces. Currently there are nine provinces in South Africa, each with a provincial Department responsible for implementing National Policy. While Provincial Departments cooperate in national policy formulation, there are some instances where Provinces are formulating certain aspects of policy that differ from central Government policy. However, it is too early to draw any firm conclusions, suffice it to say that the new political dispensation has resulted in nine Provincial Governments in a unitary rather than federal state, but that there are

40 signs of Provinces assuming greater levels of autonomy than originally envisaged. Perhaps this is a logical outcome of a tiered structure of Government.

Conclusion This brief overview of policy formulation has revealed that rangeland professionals can, to some degree, influence policy. To a greater degree, when their credibility and level of trust is high; when there is a “champion of the cause” within the body politic (e.g. a Smuts), and when the recommendations of rangeland professionals resonate with the needs of the politician and provide demonstrable short term benefits to society (e.g. the Working for Water Program). Credibility is in turn influenced by the extent to which professionals are able to present concise, easily understood statements reflecting the current state of knowledge and implications thereof. Credibility leads to trust, which presents opportunities to drive policy rather than watch it be driven. And if rangeland professionals wish to be heard, they must raise their voices in a language that the politician understands.

Literature Cited Acocks, J.P.H. 1953. The veld types of South Africa. Memoirs of the Botanical Society of South Africa 8: 1-128. Anon 1914. Report of Select Committee on Droughts, Rainfall and Soil Erosion. Report Sc 2- 1914, Govt Printer, Cape Town. Anon 1923. Drought Investigation Commission Final Report UG 49 Govt Printer, Cape Town. Anon 1972. Commission of Enquiry Into Agriculture 1968-1972. Second Report RP 84/1970. Final Report RP 19/1972. Govt. Printer, Pretoria. Anon 1985. The National Grazing Strategy. Dept Agric, Pretoria Baard, C.R. 1978. An evaluation of the stock reduction scheme. Dept Agric, Pretoria. Behnke, R.H., Scoones,I. And Kerven,C. 1993. Range Ecology at Disequilibrium. ODI, Regents College, London. Beukes, P. 1996. Smuts the Botanist. Human and Rousseau, Cape Town Hanekom, D. 1998. Agricultural policy in South Africa: A Discussion Document. CTP Printers, Cape Town. Hoffman, T., Todd, S. Ntshona, Z. and Turner, S 1999. Land Degradation in South Africa. IDRC report 1-245. MacDonald, I.A.W. and Wissel, C. 1989. Costing the clearance of alien Acacia species invading fynbos vegetation. S A J. Plant and Soil 6: 39-45. Shaw, J 1875. Changes in vegetation through the introduction of merino sheep. Bot. J. Linn Soc 14:202-208.

41 Effective Rangeland Policies for Implementing Global Conventions in Africa

GUFU OBA

Author is Senior Researcher, International Center for Environment and Development Studies, The Agricultural University of Norway, PO 5001, N-1432 Ås, Norway

Introduction Rangeland development in Africa mostly received international attention after the Sahelian drought disasters of the late 1960s and the early 1970s. The 1977 Nairobi-Desertification Conference and subsequent anti-desertification pilot projects of the 1980s were a part of the efforts to improve environmental and economic conditions on African rangelands. Through financial support from the World Bank, FAO, USAID, and the countries of the West, improved infrastructure and marketing and rangeland rehabilitation projects were funded. These projects were mostly aimed at transforming the traditional pastoral sector into market-oriented production. Nearly three decades later, the results have been disappointing. In some cases, the development programs actually accelerated rangeland degradation (Dodd 1994, Oba et al. 2000a). The Rio Conference of 1992 on sustainable development and environment broadened the debate about land degradation and threats to biodiversity. The United Nations Conventions on Control of Desertification (UNCCD) and the Convention on Biological Diversity (CBD), which the majority of the African countries ratified, are expected to increase awareness of the twin problems of land degradation and biodiversity loss in Africa (Hulme and Kelly 1993). The African countries, as part of their obligation to implement the CCD and CBD, established the National Environmental Secretariats (NES), which are cross-cutting ministerial and departmental national institutions, charged with the tasks of increasing capacity to develop environmental action plans and co-ordinating negotiations in international environmental conventions. The NES serves as an important entry point for rangeland professionals to pro- actively influence policies for implementing the United Nations Framework on Convention for Climate Change (UNFCCC). The UNFCCC and its sister, the Kyoto Protocol initiatives of 1997, will have important implications for the CCD and CBD. In comparison to other conventions, implementation of UNFCCC, although it might face a slow start in Africa, will nonetheless have important long- term policy contributions. Whereas, Africa’s contribution to industrial emissions of greenhouse gases is only 7% of the global emissions and emission of carbon dioxide is low, estimated at only 4%, the continent may offer opportunities to avoid future emissions (Sokona et al. 1998). Despite this, Africa in general, and the rangelands in particular, are vulnerable to climate variability (Nicholson 2001). Thus, for African rangelands, environmental uncertainties will reinforce the need to examine the prospects for developing effective policies for implementing global conventions. More specifically, rangeland professionals could play a crucial role in future

42 negotiations and in the pro-active development of effective policies. This is the second way in which rangeland professionals may make important long-term policy contributions. This paper offers a brief introduction to the prospects of funding implementation of the global conventions in Africa, and goes on to discuss the prospects for carbon trading in terms of developing carbon contracts between the host African countries and the utility companies from the developed countries. The role research should play in monitoring carbon sinks in grazing lands to establish baseline information for future negotiations is highlighted. Examples of carbon accumulation monitoring are provided to show how carbon sinks might benefit from funding mechanisms, as well as to illustrate opportunities for future carbon trade. The paper concludes by examining the policy implications linking rangeland research and development projects to implement the conventions in Africa.

Funding mechanisms Throughout sub-Saharan Africa and especially in the Sahel zone, foreign aid money was invested in range rehabilitation projects in the form of tree planting trials, range vegetation regeneration, and new grazing systems. The rural reforestation projects, which were popular with development agencies, were aimed at solving environmental problems and meeting fuel wood shortages. Lack of evaluation of anti-desertification projects reduced prospects for using them for certification of implementation of the climate convention. However, they will yield baseline information needed for certifying carbon sinks in degraded rangelands in Africa.

The Clean Development Mechanism The Kyoto Protocol initiatives2 offer provisions for funding implementation of the UNCCD and the CBD. Article 12 of the Protocol deals with certification of emission reductions and financial mechanisms for implementation (Grubb et al. 1999). In particular, Paragraph 5(b) of Article 12 of the Protocol includes passages such as, “real measurable, and long-term benefits related to mitigation of climate change;" while Paragraph 5(c) states, “…reduction in emission that are additional to any that would occur in the absence of the certified project activity." Article 12 and its sections are, however, less clear as to whether the certification is limited to carbon sinks or will also include avoidance of future emissions (Sokona et al. 1998). The sinks are addressed under the articles related to the so-called “joint implementation” which…

…is a system in which a country subject to a quantified emission commitment, or a company from that country, invests in projects that reduce emissions in another country. The sponsor company and/or country can claim the emission reduction, as a ‘credit’ against its own required reduced emissions. The host country has an incentive to allow…the investment on the basis of its perceived contribution to economic and/ or environmental objectives. The incentive for the participating host companies/country is the profit potential of the investment and related stakeholder benefits (Grubb et al. 1999: 88).

2 The paper will restrict itself to discussing the prospects of the CDM for funding range rehabilitation projects in Africa. The details on negotiation of the Kyoto Protocol is presented in Grubb et al. (1999) for global perspectives and Mwandosya (2000) for the developing world perspectives.

43

The funding mechanism through the “Clean Development Mechanisms” (CDM) is to assist non-Annex 1 (i.e., developing countries) to achieve sustainable development, while encouraging the Annex 1 Parties (i.e. developed countries and their companies) to reduce emissions and increase carbon storage above- and below-ground. An additional opportunity is that the carbon sinks above- and below-ground (for which the Non-Annex 1 countries will not qualify for consideration until after the first commitment period of 2008-2020), can be used for trading in future carbon markets, which experts predict will account for 2% of total global financial transactions by 2010 (François Falloux, pers. comm.). The host countries will benefit in terms of the perceived technology transfer or socio-economic development. The Kyoto Protocol provisions do not include use of rangelands for sinking carbon, even though the rangeland soils worldwide may contain as much as 10-30% of the global soil organic carbon (Schuman et al. 2002). Rangeland professionals working in Africa will, therefore, need to understand carbon sinks for renegotiations after the first commitment period (i.e. 2020). The Protocol compliance through monitoring requires that “investments in carbon sinks” above the baseline or business as usual scenario be used for verification, using some standards to be established.

Carbon trading by utility companies The ultimate goal of UNFCCC is reduction of greenhouse gases, but the Convention also seeks to increase the prospects for rewarding those Parties that meet emission reductions by transforming carbon into a commodity used in international trade. In Africa, there is little expertise on developing carbon contracts between the host countries and the parties in Annex 1. Based on limited experience (from Tanzania and Uganda), questions have been raised regarding “equity” and “transparency” of the carbon contracts between the host countries and utility companies. For example, two Norwegian Forestry-utility Companies acquired land for planting fast growing exotic tree species for carbon sinks in Uganda. Harald Eraker (2000) of NorWatch who wrote the report called “Carbon Colonialism” showed how the Norwegian Tree-Utility companies “lacked transparency,” while the Ugandan counterparts lacked technical skills to negotiate carbon contracts. The Norwegian Forestry companies (hereafter called Company A and Company B) used grants from NORAD to lease a total of 80-100,000 ha of public land. The Lease fees (for over 50 year period) annually amounted to about USD 3 per ha. The balance sheet in terms of long-term benefits is overwhelmingly in favour of the Norwegian companies, with potential earnings after maturity of the plantations to the tune of USD 28.7 million for Company A against rental expenses of USD 319,500, while Company B will potentially earn USD 13.5 million compared to the rental expenses of USD 210,000. Moreover, the contracts did not take into account land use conflicts with the local farmers and the costs of displacement. This evidence suggests that technical expertise in carbon contracts and carbon trading will be needed to explore ways that the collaborative ventures in carbon sinks between local land users and utility companies can be improved in the future (see examples later sections).

44 The GEF funding incentives Among the funding incentives for implementing the CCD and CBD as well as UFCCC is the GEF financial initiative, which serves as a mechanism for international co-operation for the purpose of providing new and additional grants to finance programmes and projects that will achieve global benefits (Sokona et al. 1998). The GEF funding offers prospects for sinking carbon in degraded rangelands worldwide, but will mostly benefit the developing countries (GEF 1994). The funding covers incremental costs conditional to the developing countries' fulfilment of their obligations under the terms of the conventions. The funding is for a single project phase on the assumption that institutional capacities have been established, while the partner countries are expected to duplicate the results through improved management. Only a few projects aimed at sequestering carbon have been implemented in African rangelands. The immediate and direct benefits from the GEF incentives are net gains in carbon sequestration above the baseline. Additional benefits of the pilot projects are increased vegetation cover, reduced soil erosion, and conservation of biodiversity. The GEF model assumes benefits to be incremental. -2 The methods for calculating incremental benefits are straightforward. Let C g m (t0) represent the baseline carbon stock in the vegetation or soil at the start of the project (or in -2 control degraded lands) and let C g m (t1) represent carbon sequestered as result of the project -1 in year (t1). Total carbon sink is expressed in terms of Mg C or t ha (Eq. 1) and the net gains as ratio of the difference between carbon sequestered in the final year and the baseline and the incremental period (i.e. carbon sequestered as a result of the project) (Eq. 2).

-1 Total carbon sinks = Ct0 + Ct1 t C ha Eq. 1 -1 -1 Net gains of carbon = Ct1 – Ct0 ha yr Eq. 2

Ct1

Potential carbon sinks are determined by multiplying the total area under management by net carbon gains. Alternatively, if the project is concerned with building shelterbelts for sand dune stabilization, carbon may be calculated in terms of individual trees by taking total area of the dunes stabilized and vegetation biomass regenerated following the project. Carbon content of wood is usually taken as 45-50% of the total fresh weight, and that of the herbaceous layer is estimated as 50% of fresh weight. The same method is used for determining carbon sequestration in degraded ecosystems. Thus, the GEF funding model for managing carbon sinks can be applied to traditional land use systems. Although rangelands in Africa are used for crop production and grazing, the examples given relate to land use systems practiced under pastoral management. The pastoral land use model might be used for monitoring effects of management on carbon sinks.

The traditional land use model In pastoral systems, land use is at the scales of landscapes (Oba et al. 2000b). Among the pastoralists in East Africa, landscapes vulnerable to degradation and those that are highly

45 resilient3 are distinguished. Different landscape patches are used by different livestock species during different seasons of the year. Turkana pastoralists in Northwest Kenya, for example, identified 15 landscape patch types within 5,000 km2, each serving as a grazing resource for multi-species livestock. The indigenous ecological knowledge of pastoralists thus provides an important ecological and management basis for monitoring carbon sinks at landscape scales.

Monitoring carbon sinks in rangeland landscapes The rangeland professionals working in Africa, to accomplish the tasks of establishing verifiable methodologies for measuring carbon sinks and developing carbon contracts, would need to be familiar with the complicated rules of the Kyoto Protocol, such as establishing standards for making verifications and certification of carbon sinks. Baseline data for carbon stocks above- and below-ground, as well as turnover time for different carbon pools, should be understood before the renegotiations during the second commitment period in 2020. Carbon stock data from African rangelands is sketchy and values vary from 50.8 x 1015g (Gaston et al. 1998) to 275 Gt (Cao et al. 2001). African soils reportedly contain only 12% of the total global carbon pools (Batjes 2001), and their carbon sink potentials appear to be lower due to threats from land degradation (Lal 1999). Most importantly, data on the effects of grazing on carbon sinks is sparse. Yet, data from past grazing experiments could be used to calculate carbon sinks and to derive baseline information under a business as usual scenario. The examples below show how monitoring of carbon sinks on rangelands may be accomplished.

Example 1 The role livestock grazing plays in carbon sinks in African grazing lands is little documented, while misperception about rangeland overgrazing associated with livestock grazing and browsing gives the impression that the effects on carbon sinks will be negative (see review, Oba et al. 2000c). On the contrary, grazing stimulates carbon accumulation in the world’s rangelands (Schuman et al. 2002). In the arid central Turkana (northwestern Kenya), between 1986-1990, effects of free- ranging goats on biomass production in Acacia tortilis woodlands were monitored using 4-ha exclosures that received browsed and control treatments (each 2-ha in size). The pre-study tree density was 458±119 trees ha-1 in the control treatment and 446±116 trees ha-1 in the browsed (Oba 1998). In the browsed treatments, average goat-stocking densities varied from 16.6 to 25 TLU km2 (Oba and Post 1999). In each treatment, 10 trees that varied in height from <1.5 m heights to 2.0 m heights were selected randomly. Tree stem diameters were marked at about 100 cm aboveground. At six-month intervals from 1986 to 1990 the marked trees were measured repeatedly for stem diameter, height, and crown diameter (mean of the short and the longer diameters). Wood volume (Volume = ½pr2h/ m3) was converted into biomass by multiplying wood specific gravity (which is 1 for Acacia sp.). The wood biomass data was used to calculate carbon accumulation above-ground, which was estimated by multiplying wood biomass per tree x 0.45 (the wood carbon content coefficient) x 3.66 (the conversion coefficient from organic

2 I have used the concept of resilience here as understood by the pastoralists. It is a term that describes capacity of the land to recover after heavy grazing as well as resistance to degradation (Oba and Kotile 2001).

46 carbon to CO2). The results showed greater carbon accumulation in the browsed trees compared to the controls (Fig. 1). When interpreting these data, however, goat-browsing stimulation alone could not explain the differences between treatments. Rather, field observations suggested that goat browsing and grazing indirectly improved rainwater infiltration through removal of herbaceous vegetation in contrast to the control, where soil water infiltration was reduced by soil crusts4 and by competition between the herbaceous and tree layers. The improved growth conditions favored greater accumulation of carbon in the browsed plots relative to the controls.

50000

# Browsed 40000 # ( Unbrowsed #

30000 # # 20000 #

# # 10000 ( ( # ( ( ( ( ( ( 0 Sep86 Feb87 Sep87 Feb88 Sep88 Feb89 Sep89 Feb90 Sampling dates

Fig. 1. Effects of goat herbivory on carbon accumulation in young Acacia tortilis woodlands in arid central Turkana, NW Kenya (Gufu Oba unpubl. Data)

Example 2 In the lower sections of the Turkwel River, 14.5 km2 of the floodplain woodlands were mapped (scale 1: 120,000) and wood volume and litter biomass in individual ngikwarin measured in 1990 before dam impoundment in 1991 and repeated again in 1998. Wood volume and litter biomass data were converted into carbon stocks for individual ngikwarin (Fig. 2a,b). Carbon stocks in the woodland and litter between 1990 and 1998 showed a decline that probably reflected the combined effects of changes in the flood regimes and human exploitation of the woodland.

4 Soil crusts according to Van Bremen & Kinyanjui (1991) in the study area were a common condition in areas that have not been trampled by livestock after the rains.

47 3500

3000 a) 1990 1998 2500

2000

1500

1000

500

0 1 2 3 4 5 6 7 8 9 1011121314151617181920212223 Ekwar

3500

3000 b)

2500

2000

1500

1000

500

0 1 2 3 4 5 6 7 8 9 1011121314151617181920212223 Ekwar

Fig. 2. Carbon accumulation in (a) woodlands and (b) litter layer in floodplain of the Turkwel River, measured between 1990 and 1998 (Gufu Oba unpubl. Data). Note: numbers 1, 2, 3…23 represents woodland patches owned by individual pastoralists.

48 Implications for global conventions These two examples show how rangeland professionals might use data from past research to understand the effects of land use and develop verifications for carbon sinks and emissions. In the first example, grazing was shown to have positive effects on carbon accumulation, while in the second; the effects of development seemed to be adverse to carbon accumulation. The results established a baseline for carbon certification. Further, the findings have implications for the roles rangeland professionals might play in improving knowledge of how land use systems contribute to carbon emissions. For the utility companies, the model of the floodplain might explain the risks of investing in carbon sinks without addressing problems associated with development policies related to building dams in drylands. Under normal conditions of natural floods, the floodplain woodland might have offered local land users the opportunity to trade carbon instead of exploiting forest resources. The risk of the investment, however, becomes greater if building dams induces carbon emissions from the floodplain.

Effective rangeland policies During future negotiations, rangeland professionals need to emphasize the use of carbon sinks in the world’s rangelands. Through research and monitoring, the criteria for evaluating carbon sink projects will be established. Given the paucity of range research in Africa, rangeland professionals might collate the sparse data from experimental stations and archives to build up baseline information. Moreover, whereas the socio-economic issues related to carbon sinks in rangelands remain to be clarified, rangeland professionals should pro-actively be involved in future negotiations of the conventions to promote carbon sinks in the rangelands, establish baselines for the business as usual scenarios, develop criteria for monitoring and certifying carbon sinks, and develop guidelines for carbon contracts. Policy makers should be aware that, although African countries are presently not required to offset carbon emissions, they are faced with conflicting responsibilities. By allowing utility companies to invest in their land to sink carbon they cannot use the same land to earn emission credits during future negotiations when specific conditions and terms are worked into the protocol (Eraker 2000). Other potential risks such as insecurity, climate unpredictability, poor land tenure, and lack of resource accounting for carbon credits will constrain rangeland policies.

Conclusion Developing countries in general and those of Africa in particular are signatories of the CCD and CBD but are not yet required to reduce emissions according to the Articles of the Kyoto Protocol. The developed countries under the provisions of the CDM are willing to invest in forestry projects in the developing countries to earn carbon credits and participate in global carbon trade. Alternatively, there are provisions for allowing local farmers and pastoralists during the second commitment period to manage rangeland resources for carbon sinks (e.g. Shroader, 1994) and to use carbon stocks for trading, just as coffee farmers trade their commodity. The issues that are presently undefined should be the focus of future rangeland policy in Africa. The policy for promoting implementation of global conventions should be

49 based on systems of monitoring and certification of baseline carbon sinks, and the development of guidelines for carbon contracts. The policy initiatives are likely to reactivate rangeland rehabilitation projects in terms of implementing carbon sinks.

Acknowledgements The author thanks the International Affairs Committee for Society of Range Management; especially Dr. Maria E. Fernandez-Gimenez for invitation to the Symposium and the International Centre for Environment and Development Studies (Noragric) is thanked for supporting the preparation of the paper.

Literature Cited Batjes, N.H., 2001. Options for increasing carbon sequestration in West African soils: An exploratory study with special focus on Senegal. Land Degradation & Development 12:131- 142. Cao, M.K., Zhang, Q. F. and Shugart, H. H. 2001. Dynamic responses of African ecosystem carbon cycling to climate change. Climate Research 17:183-193. Dodd, J.L. 1994. Desertification and degradation in sub-Saharan Africa. BioScience 44:28-34

Eraker, H. 2000. CO2lonialism-Norwegian Tree Plantations, Carbon Credits and Land Conflicts in Uganda. Norwatch, Oslo. Gaston, G., Brown, S., Lorenzini, M. and Singh, K. D. 1998. State and change in carbon pools in the forests of tropical Africa. Global Change Biology 4:97-114. Global Environmental Facility 1994. Sudan: Community-based rangeland rehabilitation for carbon sequestered and biodiversity. UNDP, New York Grubb, M., Vrolijk, C. and Brack, D. 1999. The Kyoto Protocol: A guide to Assessment. The Royal Institute of International Affairs. London. Hulme, M. and Kelly, M. 1993. Exploring the links between desertification and climate change. Environment 35:39-46. Lal, R. 1999. Soil management for carbon sequestration. Pp. 108. In: Drylands, Poverty and Development. Proceedings of the June 15 and 16, 1999 World Bank Round Table. The World Bank Washington, DC. Mwandosya, M. J. 2000. Survival emissions: A perspective from the south on global climate change negotiations. CEEST, Dar-es-Salaam. Nicholson, S. E. 2001. Climate and environmental change in Africa during the last two centuries. Climate Research 17:123-144. Oba, G. 1998. Effects of excluding goat herbivory on Acacia tortilis woodland around pastoralist settlements in northwest Kenya. Acta Oecologia 19:395-404. Oba, G. and Kotile, D. G. 2001. Assessments of landscape level degradation in southern Ethiopia: Pastoralists versus ecologists. Land Degradation & Development 12:461-475.

50 Oba, G. and Post, E. 1999. Browse production and offtake by free-ranging goats in an arid zone, Kenya. J. of Arid Environment 43:183-195. Oba, G, Stenseth, N.C. & Lusigi, W. J. 2000a. New perspectives on sustainable grazing management in arid zones of sub-Saharan Africa. BioScience 50:35-51. Oba, G., Post, E., Syvertsen and Stenseth, N. C. 2000b. Bush cover and range condition assessments in relation to landscape and grazing in southern Ethiopia. Landscape Ecology 15:535-546. Oba, G., Post, E., Stenseth, C.N. and Lusigi, W.J. 2000c. The role of small ruminants in arid zone environments: A review of research perspectives. Annals of Arid Zone 39:305-332. Schroader, P. 1994. Carbon storage benefits of Agroforestry systems. Agroforestry Systems 27:89-97. Schuman, G. E., Janzen, H. H. and Herrick, J. E. 2002. Soil carbon dynamics and potential carbon sequestration by rangelands. Environmental Pollution 116:391-396. Sokona, Y., Humphreys, S. and Thoams, J-P. 1998. What prospects for Africa? Pp. 109-118. In: Goldemberg, J. (ed.). Issues and options. The Clean Development Mechanism. UNDP, New York. Stave, J., Oba, G., and Stenseth, N. C. 2001. Temporal changes in woody-plant use and the ekwar indigenous tree management system along the Turkwel River, Kenya. Environmental Conservation 28:150-159. Van Bremen, H and Kinyanjui, H. C. K. 1992. Soils of the Lodwar area: an inventory of soils, an evaluation of present land use and recommendation for future land use. Reconnaissance Soil Survey Report No. R17. Kenya Soil Survey, Nairobi.

51 Taking Stock: Policy, Practice, and Professionalism in Rangeland Development

ROBIN MEARNS

Author is Senior Natural Resource Management Specialist, Rural Development and Natural Resources Sector Unit, East Asia and Pacific Region, World Bank, 1818 H Street NW, Washington D.C. 20433.

Abstract Support for range management interventions by the World Bank and other international development institutions has a somewhat checkered history. In certain respects, this has mirrored theoretical developments and shifts in thinking among rangeland professionals, but the relationship between rangeland science and development policy and practice is far from direct. It is mediated by contextual factors such as the relative importance of livestock production as a livelihood source and the relative political power of pastoralists in the countries in question; and influenced by broader debates surrounding property rights, globalization, and sovereignty in international development. This paper takes stock of some of the lessons learned in recent years from attempts to engage with rangeland tenure and policy in arid and semi-arid areas of Africa and Asia, with a particular focus on the experience of the World Bank, most recently in China and Mongolia. It concludes with reflections on the role of rangeland science and rangeland professionals in the development process.

Introduction As an institution committed to the goal of eradicating global poverty, the World Bank has long been engaged in supporting rangeland management in developing countries. The context and rationale for supporting such interventions, however, has changed considerably over recent years. This paper aims to take stock of some of the lessons learned in engaging in rangeland tenure and policy in developing countries, drawing particularly on experience in Africa and Asia. Several undercurrents underlie this ongoing experience. Some appear to strengthen the case for supporting environmentally sustainable and socially inclusive rangeland management interventions in developing countries, while others appear more contradictory. These undercurrents concern: the broadening of the Bank’s development mandate and the ways the Bank has positioned itself to respond; changes in thinking within the range management profession at the international level; and a growing appreciation that the practical opportunities afforded by these encouraging developments remain heavily circumscribed by prevailing socio- political and historical contexts at the country level. The challenge for rangeland professionals lies in being able to chart a course through this complex policy terrain that is theoretically and empirically informed yet practically feasible in particular country contexts. The rationale for World Bank support in rangeland tenure and policy has broadened and deepened over the past decade or so with a re-assertion of poverty eradication as the Bank’s core

52 mission (World Bank 2001a), and with a new emphasis on environmental stewardship as both an end in itself and an important means of contributing to sustained poverty reduction (World Bank 2001b). Poverty is commonly associated with heavy dependence on natural resources, and those who depend upon these resources include indigenous peoples, whose cultural survival is tied to their use of the resource. The sharper focus on reducing poverty has directed attention towards these resource areas, and towards solutions that meet the needs of those in poverty, rather than solutions that give highest priority to the needs of the larger national economy. Nowhere is this shift in emphasis clearer than in the Bank’s work on livestock development (de Haan et al. 2001). An earlier generation of projects focused on commercial production, often for export and with foreign exchange needs very much in mind, but more recent projects give priority to meeting local livelihood needs by working with local production systems, and are concerned with interactions between livestock production and the environment (Mearns 1997a, 1997b, de Haan et al. 1998). The interests of groups beyond the local also come into play, such as those at a national or international level who favor biodiversity conservation. The need to balance competing claims among multiple users and uses at multiple scales then becomes even more challenging. The broadening of the Bank’s mandate has also called into question the utility of the Bank’s conventional lending instruments, particularly for supporting more complex and risky interventions that require engagement over the longer term. Specific investment loans, for example, typically limited to four to six years in duration, are not well suited to interventions that combine often contentious policy reforms with innovative investments on the ground that require prior pilot-testing, yet it is now recognized that rangeland development projects must usually take this form. The last five years has seen the Bank experiment with new, more flexible lending instruments, including small ‘learning and innovation loans’ (LILs), often suitable for pilot- testing new ideas before they can be scaled-up operationally; and ‘adaptable program loans’ (APLs), which entail several phases over a longer period (say, 12 years), and are designed to allow for greater flexibility in implementation, with specific triggers such as key policy reforms marking the transition between phases. These new lending instruments are often suitable for operations that seek more explicitly to achieve environmental and social outcomes, including those in rangeland management. The Bank has also considerably increased its professional cadre of technical specialists with skills in the environmental and social sciences. Over a similar period, rangeland management as a profession has also seen significant changes, particularly those engendered by the growing acceptance of the ‘new’ thinking in range ecology. This paradigm shift has focused attention on rangeland dynamics, which in turn has significant implications for policy and management (Scoones 1994, Lane and Moorehead 1994). Attempts to support more adaptive approaches to rangeland management are still in their infancy in the field of international development, but some recent efforts have tried to bring together these new theoretical insights, aided by the more supportive attitude towards adaptive management within the World Bank and certain other international development institutions. For example, a stronger operational focus on addressing vulnerability in particular, as well as poverty in general, has led to institutional innovation in the area of drought preparedness and other forms of pastoral risk management. Currently, however, there are also some signs that practical progress towards achieving more environmentally sustainable and socially inclusive outcomes in important areas of the world’s rangelands may be in danger of being derailed by broader political obstacles. As the

53 World Bank’s mandate has broadened, and as it has begun to engage more directly in activities involving indigenous peoples or ethnic minorities and their customary natural resource base, controversies have begun to emerge in particular country contexts, usually surrounding the Bank’s operational directives concerning environmental and social safeguards. International NGOs and human rights groups as well as national governments are actively engaged in dialogue on these issues which could impact pastoral livelihoods. This may make it more difficult to tackle the very real challenges of rangeland management in some of the countries in which they most urgently require attention. In such situations rangeland professionals and other technical specialists should be aware of the broader context within which they operate, beyond the narrower issues of pastoral ecology and pastoral livelihoods alone. These parallel undercurrents come into sharp focus in those empirical contexts in which the disconnects between science/theory, management and policy, and empirical realities on the ground are at their widest. The "new" thinking in range ecology emerged principally from theoretical and empirical research in sub-Saharan Africa. New work seeks to test its applicability in other dryland regions of the world, for example in Inner Asia, but it has often proved difficult to conduct the detailed field research that is required to understand the effects of particular policies and actions in particular places. Whatever the insights from the ‘new’ range ecology, and however strong the technical justification for certain policy and investment interventions may be, development practitioners are reluctantly coming to the realization that ‘second-best’ approaches may need to be adopted where these offer more promise of practical progress on the ground (Mearns 2001). This pragmatic acceptance that ‘the ideal may be the enemy of the good’ is also influencing other fields, such as biodiversity conservation (Margules and Pressey 2000, Whitten et al. 2001). Against this background, the case material in this paper takes stock of some of the lessons learned in efforts to engage in rangeland management in Africa and, more recently, Inner Asia, focusing particularly on their implications for pastoral land tenure and resource access. Although the issues being confronted may appear strikingly similar, and inter-regional comparative analysis undoubtedly has an important role to play, policy choices need to be made according to what is judged to be feasible under different ecological, socio-political, and historical circumstances. The paper concludes with reflections on some appropriate roles for rangeland science and rangeland professionals in contributing to pastoral land tenure policy and practice in developing countries.

Pastoral land tenure and resource access: a brief review of operational experience The World Bank has supported pastoral and rangeland development in arid and semi-arid areas, particularly in Africa and the Middle East, for several decades. The manner in which land tenure and policy have been approached in this field, however, has often been found wanting. The literature is replete with examples of misguided Bank and other donor support for, among other failures, borehole development in Botswana and group ranches in Kenya, that have variously exacerbated conflict over resources and contributed to elite capture of higher-value, better-watered land; uncontrolled privatization of common pastures; and severe environmental degradation. Over the past 15 years or so, there has been a fundamental questioning of the conventional wisdom that underlay earlier, failed pastoral development efforts (Sandford 1983,

54 Behnke et al. 1993, Scoones 1994, Leach and Mearns 1996, Niamir-Fuller 1999). This has also led to a more nuanced approach among some donor agencies and national governments to rangeland tenure and policy (Lane and Moorehead 1994, Grell and Kirk 2000). The learning process has been incremental rather than lending itself to discrete shifts in policy and practice; nonetheless, it is possible broadly to characterize the distinguishing features of these approaches. Conventional, property-rights approaches tend to treat all extensive livestock production systems as if they were essentially similar. The central issue to be addressed, in this view, is to achieve sustainability by balancing the number of grazing livestock against the long-run carrying capacity of the range. On the assumption that groups are unable internally to control access to and use of resources among their members, it is taken for granted that the appropriate management solution is to assign property rights to individuals (or to the state). The resource is assumed to be perfectly divisible in this respect, and conflicts among resource users are assumed to be minimized through sub-division and individual appropriation. The fundamental premises behind such reasoning are weakened if ecological variability and the possibility of uncertainty are acknowledged (see Box 1). Livestock producers in such systems need to be able to ‘track’ available forage or browse for their animals, which usually requires that they have access to large areas that encompass a diverse range of landscape niches. This calls for livestock mobility and flexibility in access to resources to the maximum extent possible. Sub-division of the resource and assigning property rights at insufficiently large a scale creates that risk of introducing rigidities that preclude opportunistic, feed-tracking strategies on the part of resource users. This could worsen their vulnerability to drought, worsening asset/income distribution among resource users (since higher quality resource patches tend to be captured by those with higher bargaining power), and exacerbating any existing tendencies towards overgrazing under current management practices, resulting in more not less conflict.

Box 1. The equilibrial - non-equilibrial continuum in grazing ecosystems

The "new" thinking on range ecology highlights the distinction between equilibrial and non-equilibrial ecological systems (Behnke, Scoones and Kerven 1993). This is not an either/or distinction, but rather a continuum. At one extreme, relatively equilibrial systems are those in which the density of grazing livestock explains a significant amount of the variation in vegetation dynamics over time, and in which conventional range management techniques such as maintaining appropriate average stocking rates are thought to remain most suitable for sustainable grassland management. At the other extreme, relatively disequilibrial systems are those in which livestock populations and vegetation dynamics are only loosely coupled, and density-independent factors such as precipitation explain a higher amount of variation in vegetation dynamics. Opportunistic means of "tracking" available nutrition from natural grazing and browse, usually through mobility, are typical pastoral management adaptations to such spatial and temporal variability. Annual rainfall totals and/or the coefficient of variation in annual precipitation is often taken to be a proxy for this continuum, given that more arid areas tend to experience greater inter-annual variability in precipitation. While many of these insights are derived from empirical work in African savanna ecosystems, a steadily growing body of work shows that they are also relevant to pastoral production systems in other regions of the world, such as Inner Asia (Fernandez-Gimenez and Allen-Diaz 1999, Humphrey and Sneath 1999, Dunlop 2000, WRI 2000). What emerges from this work is a more nuanced view of the importance of mobility and flexibility in pastoral land tenure and resource access, without rejecting outright the need to consider density-dependent grazing pressure as well.

Property rights theorists have not been unaware of these challenges. Recognizing that certain resources have distinct attributes (such as a high degree of ecological variation over time

55 and space) that lend themselves to use in common, common-property approaches are also widely advocated as appropriate for pasture land tenure. This often implies a need for strengthening, reconstituting, or adapting customary rules governing land tenure and resource access. It is assumed, first, that the resource unit can be defined at a sufficiently broad level that the full range of desired landscape niches is included, and natural variability thereby internalized. Second, it is assumed that a group of users can be identified that is commensurate in scale with this resource unit. Property rights over the resource may therefore be vested in the group as a corporate entity, leaving more detailed questions of resource allocation, access, use, and management to be decided by governance structures internal to the group. While attractive in theory, there has been little success with such "fixed-boundary" common-property approaches to pasture land tenure in practice. On the one hand, they often aim at a moving target: external pressures on pastoral production systems may confound even the best-laid plans for their internal management. Pressures typically arise from the conversion of better-watered land to crop agriculture and its loss as dry-season grazing or browse to pastoralists or competition from other forms of land use such as peri-urban development or environmental conservation (Fratkin and Mearns forthcoming). Under such pressures, formerly sustainable pastoral production systems can quickly become so compromised that it is extremely difficult to reconstitute them. The limits of what is feasible may be closely circumscribed by prevailing population densities, patterns of migration into or out of pastoral areas, and the availability of supplementary or alternative sources of livelihood, which necessitates a case-by-case, context- sensitive approach between and within particular countries. On the other hand, social groups large enough to internalize the necessary degree of landscape diversity are often too heterogeneous to overcome the challenges of collective action. Not only are there multiple uses made of the resources in question by multiple users, so that their claims overlap, but these claims are frequently contested and subject to intense negotiation and re-negotiation over time. It rarely proves possible to ‘resolve’ such conflicts once and for all; indeed, conflict is seen to be an integral feature of many pastoral and agro-pastoral production systems (Hendrickson et al. 1998). Where common property regimes are "strengthened" with external support so as to create more or less fixed resource-unit and social boundaries, such conflicts may become unmanageable. In recognition of the need for a more flexible approach to land and resource tenure under such conditions, "negotiated-tenure" approaches are increasingly being tried. The focus of these process-orientated approaches is to provide a framework for fair and transparent consultation among competing resource user groups, such that resource boundaries and access options are subject to ongoing contractual negotiation (Behnke 1994). Still the problem remains of specifying in a sufficiently inclusive manner who are "legitimate" resource claimants, and who has the right to decide (Leach, Mearns and Scoones 1999). These approaches are in the forefront of international thinking on pastoral land tenure and resource access, and are still at a highly experimental level. Making allowance for the usual lead-lag effect in the way conceptual advances filter into development practice on the ground, the learning process described here has begun to exert a powerful influence over the Bank’s evolving (if shrinking) pastoral development portfolio in the past decade. Many of the "new" generation projects under implementation or in preparation incorporate some of the latest thinking and lessons learned from past experience. The current portfolio is therefore diverse, and reflects the accretion of ideas concerning pastoral development

56 over recent years. No single ideology or "model" for pastoral land tenure and policy prevails, as is appropriate given the wide variety of ecological, social and economic contexts in which pastoralists and agro-pastoralists pursue their livelihoods. While there are hopeful signs in certain countries, meaningful progress remains hampered by the relative political weakness of pastoralists in most countries and the breadth and complexity of the reform agenda. Earlier reviews of Bank-supported experience in livestock development have not chosen to focus particularly on its implications for land tenure and policy, although relevant lessons can be drawn from them. Aside from the errors of commission that hindsight now allows us to recognize, as in the case of Kenya’s group ranches, there have also been errors of omission: arguably, more could and should have been done to protect pastoral land rights in national policy dialogue. Bank-supported approaches to pastoral development in Africa and the Middle East have evolved from support for capital investment in ranching through fencing, water point development and introduction of exotic breeds of livestock (1960s-1970s, e.g. Botswana, Kenya, Yemen); through development of infrastructure (water, roads, marketing) in communal rangelands (e.g. Eastern Senegal, Somalia), including adjudication of grazing rights under the concept of group ranches (1970s-1980s); to support for pastoral associations and integrated, participatory approaches to natural resource management (1980s-ongoing, e.g. Burkina Faso, Mali, Mauritania) (de Haan et al. 1994). While the need to facilitate herder mobility through more flexible approaches to land tenure and access to grazing was recognized from the late 1980s, less practical progress was made in this respect than in the organization of herder- managed services (notably animal health and water-point rehabilitation) through pastoral associations (Shanmugaratnam et al. 1992, Pratt et al. 1997, Øygard et al. 1999). More recently, with the Bank’s renewed emphasis on poverty reduction, the Bank’s evolving livestock strategy reflects the growing consensus that pastoral development is motivated as much by social and environmental objectives as by production- or output-orientated objectives alone (de Haan et al. 1998, de Haan et al. 2001). Attention is shifting towards a broader concern with the sustainability of grassland ecosystems and with livelihood security and sustainability for those who rely on them. Risk management through drought contingency planning and investment in community-based infrastructure has become a major focus of current Bank support for pastoral development (notably in Kenya and Mongolia), and it is acknowledged that land tenure remains an important dimension of the relevant policy reform agenda in this respect. Important and challenging opportunities to support poverty reduction through grassland tenure and management interventions are also beginning to open up, for example in China with the central government’s recently announced priority on development of the western region. A recent review of Bank-supported drylands interventions by the World Bank’s Operations Evaluation Department (OED) found that levels of lending in this field have been lower than is often claimed. Over FY90-FY98, around $1.73 billion was lent for a total of 54 projects that aimed to reduce land degradation in dryland areas, with a marked decline over FY94-FY98. This represented 6 per cent of total lending in agriculture over the same period. By no means did all of this lending address pastoral development or pastoral land tenure directly. OED judged that annual lending volumes specifically for pastoral development had declined tenfold from $200 million in the late 1970s to $20 million by 1999. It is likely, however, that a disproportionate share of overall lending for drylands development had some implications for pastoral land tenure and resource access, whether positive or negative.

57 Recent assessments suggest that the quality of Bank-supported work focusing on land tenure in dryland areas has been only marginally satisfactory. The Bank’s Rural Development Department found that of 39 projects since 1985 related to pastoral land tenure, task teams identified land tenure, access to resources, and conflict resolution as key issues in 27 cases, but in only 16 of them were relevant actions undertaken. Similarly, only six of the total of 18 projects in the OED review were judged to have made a practical contribution through studies or pilot schemes that may lead to purposive change in land tenure and resource access arrangements on the ground. The Bank’s performance in supporting policy reforms that affect drylands, including land policy, was also found to have been less than satisfactory. Policy reform was considered substantially relevant for 11 of the total of 18 drylands projects reviewed by OED, yet in only three was significant progress made. This mixed experience begs the question, should the Bank have attempted to intervene more in pastoral land tenure and resource access? OED drew several conclusions from past experience in answer to this question. First, land tenure in drylands is highly complex, not least owing to conflict between customary and statutory legislated rules. Towards the drier end of the precipitation gradient, the need is more for fair, efficient and equitable rules of resource access rather than titled ownership of land. Conflict management is recognized to be an especially important area for future attention, particularly given changing livelihoods in drylands (transitions towards mixed farming rather than livestock production alone; "de-agrarianization" in farming areas (Ellis 1998, Bryceson 2000). Second, this complexity carries high reputational risks for the Bank in attempting to support purposive change in land tenure and resource access. And third, whether or not the Bank should have attempted to play a more active role in policy and practice, it should certainly have done more to support multi-disciplinary analytical work as a means to reduce the risks through greater understanding of the issues involved. Rangeland professionals are among those whom the Bank ought to have involved a great deal more in such analytical work in the past, and there are now encouraging signs that current project teams are paying greater attention to including the relevant skills. The following examples are illustrative of this shifting policy agenda and how Bank operations have attempted to respond in practice. These experiences are described in more detail elsewhere (Bruce and Mearns 2002). In Chad and Mauritania, Bank-supported initiatives in the late 1980s to organize herders into pastoral associations based on existing tribal fractions achieved a measure of decentralization in animal health service delivery and water point rehabilitation in otherwise highly centralized administrative settings. Policy reforms around land and water tenure failed to materialize, however, in part owing to the misalignment of project goals (improved NRM) and those of pastoralists and policy makers (water and animal health). Subsequent Bank-supported efforts in Mauritania tended to address the needs of agro-pastoralists and sedentary villagers rather than those of transhumant pastoralists, encouraged by a village- territory oriented approach. Some of these interventions have had the unintended outcome of strengthening the land claims of sedentary agriculturalists at the expense of transhumant pastoralists. Even ostensibly participatory approaches to land use planning, such as the gestion des terroirs approaches that are widely applied in West Africa, risk exacerbating such conflicts, since they tend to focus on village communities that can easily be spatially delineated, rather than at a broader, landscape scale which would also encompass transhumant pastoralists and the landscape niches they value.

58 Building on the lessons learned under this earlier generation of projects in West Africa, the Bank is supporting the pilot-testing of a "holistic resource management" approach to pastoral land use in a number of West African countries including Chad, Burkina Faso, Mali, Mauritania, Senegal and Niger. Adopting a consensual approach to goal-setting, taking account of local livelihoods and landscapes as a whole, the program aims to support pastoral and agro-pastoral communities in developing grazing management plans that also take account of transhumant herders as secondary resource users, together with the complementary provision of agricultural services. An independent evaluation of program achievements in 1999 found strong support for the initiatives among participating agro-pastoralists, who perceived benefits in vegetation cover, animal health, calving rates, milk production, higher income from livestock, and reduced labor demands. These benefits could not be confirmed by field observations, however, or by the system for monitoring and evaluation of development outcomes. M&E systems are therefore being strengthened, in terms of local capacity to adapt and re-adapt grazing management plans according to changing conditions, external support to assess the livelihood benefits for participating communities, and the wider environmental benefits of the program. Pending the outcome of those evaluations, the program is poised to scale up these pilot initiatives to national and regional levels. Significant challenges lie ahead in maintaining the momentum achieved to date in local-level conflict resolution. In international experience to date, there are very few countries in which a permissive legislative framework exists at national-level to underwrite the principles of mobility and flexibility in pastoral land tenure and resource access. Where such a framework does exist, its central provisions are often weak and difficult to implement, in the absence of implementing regulations and sound institutions to realize the vision in practice. As a result, the Bank and other donors are increasingly adopting a twin-track strategy, focusing attention both on supporting or scaling-up local-level, pilot initiatives in adaptive co-management and, often at the same time, on national policy dialogue. In Mauritania, for example, GTZ is supporting pilot schemes for "focal-point management" of the natural wetlands vital to the sustainability of pastoral livestock production and, in parallel, the development of national policy reforms to create the legal basis for a Côde Pastorale. Central among the principles being followed are efforts to give formal legal recognition of customary resource management practices, in particular those of transhumant pastoralists. Involving a fundamental decentralization of power, the process is still in its infancy, but in June 2000 a succinct law was enacted that aims to preserve pastoral mobility and rights to common grazing for transhumant herders. A similar national policy reform process began in Niger in the early 1990s but has since stalled at the political level. Burkina Faso offers another case in which GTZ is supporting local-level, collaborative approaches to natural resource management and conflict mediation between transhumant pastoralists and agro-pastoralists, again by creating platforms for multi-stakeholder consultation and negotiation around land tenure and resource access (Banzhaf et al. 2000). In Kenya, having learned lessons from the failed group ranch approach of the 1980s which disintegrated owing to elite capture of benefits and ad hoc land privatization (Rutten 1992, Lane 1998), the Bank is also supporting focal-point management in the context of a broader drought mitigation strategy. Recognizing the vital importance of key water and grazing resources during times of seasonal and episodic stress, and building on those customary land and resource tenure systems that remain effective, the approach has provided space for Government-NGO

59 partnerships to address inter-tribal conflicts over land and access to key resources. A detailed pastoral land tenure study laid the basis for policy dialogue with Government, although here, as in West Africa, the political context demands a slow, cautious approach to land policy reform. Against a background of armed and often violent conflict, political favoritism of some tribal groups over others is a major obstacle to a negotiated approach to focal-point management, since different stakeholder groups are perceived to enter these negotiations on an unequal footing. Pastoral risk management is also a major focus of the Bank-supported program in Mongolia, justified by a broader concern to support the Government’s efforts to eliminate poverty and ensure sustainable livelihoods for all. Here, severe winter weather conditions frequently compound the effects of drought to threaten livelihood security for those – around a third of the total population – who rely on livestock for a living. A central priority of the Government is to rebuild a pastoral risk management strategy adapted to the needs of a market- oriented economy, including attention to pasture land tenure and management (Box 2). By comparison with the African examples discussed above, Mongolia’s national policy framework is actually remarkably supportive, in principle, of mobile livestock production. In large part the more supportive policy environment is a reflection of the fact that in recent years pastoralists have been a rising share of total population in Mongolia, and have accounted for a growing proportion of GDP, in stark contrast to most other countries in which pastoralism is significant. Nonetheless, economic transition in the 1990s gave rise to new pressures within Mongolia’s extensive livestock sector which, combined with growing poverty, have increased the proportion of the population that is vulnerable to risk in pastoral livestock production. While the Constitution protects state-owned grazing land as ‘common land’, community-based initiatives are essential to restore previously effective systems of seasonal pasture rotation and grazing reserves for emergency use. Many of the pressures in the livestock sector, and rising vulnerability among herding households, can be attributed to the lack of alternative livelihood sources. The proposed Mongolia Sustainable Livelihoods Project currently under preparation therefore also aims to support community-driven investments in local infrastructure and access to rural micro-finance services in order to improve livestock marketing and help to foster more diverse livelihood sources and strategies. In China, the Bank has had little involvement in matters of pasture land tenure and management to date. This looks set to change, however, as central Government places growing emphasis on environmental management, poverty reduction, and development of the western provinces in which the vast majority of China’s pastoralists live. In this case, perhaps to an even greater extent than elsewhere, major challenges turn on tailoring policies and investments to the highly diverse ecological, social, and economic conditions prevailing throughout western China, rather than in following a ‘one-size-fits-all’ approach. For example, the promotion of "grassland contracting" on an individual household basis is unlikely in all circumstances to meet the stated objectives of enhanced grassland productivity and sustainability, as revealed by a recent Bank- supported grassland management study (Box 3). The challenge, however, is knowing when and where it is appropriate to try alternative approaches to pasture land tenure and management. Under some circumstances, such as in large parts of Inner Mongolia, for example, mobile livestock production systems have come under such sustained threat from rising human and animal population densities and from ‘land reclamation’ that there may now be little alternative but to support investments to intensify livestock production based on some level of household contracting of grassland. On no account should this trend be assumed to be inevitable everywhere, however, nor should it be artificially accelerated where more extensive forms of

60 livestock production can be shown to offer greater promise for livelihood security and sustainability. In many other parts of western China, such as large parts of northern Xinjiang or on much of the Tibetan Plateau, mobile livestock production warrants support in its own right, with particular attention paid to pastoral risk management (Mearns 2001).

Conclusion What role for rangeland professionals? Several themes emerge from this review of the Bank’s experience in the field of pasture land tenure and policy in Africa and Asia, which take the relevant policy agenda well beyond the terrain usually thought to be the concern of rangeland science and rangeland professionals. First, ‘natural resource management’ per se is rarely identified by pastoralists themselves as a major priority, in spite of the fact that many recent pastoral development interventions have been motivated by a concern for sustainable resource management. At best, pastoralists may go along with certain land-use restrictions in return for a share of other benefits that pastoral development projects bring, such as water point development or herder-managed animal health services. At root are often misconceptions concerning the types of "natural resource management" required to ensure sustainability of the resource base as well as local livelihoods. Often, a variety of indirect or complementary policy or investment approaches is required to provide the enabling conditions for continued mobility and flexibility in livestock production. This broader framing of the problem may be thought of as “tenure +”, where the additional actions could take a range of different forms depending on local circumstances. In some cases, access to alternative or supplementary livelihood sources may be sufficient to reduce the proportion of the population primarily dependent on grazing resources; a goal which could be pursued through promoting access to alternative forms of capital, and social service provision (education, human and animal health care) tailored to mobile populations so as to expand opportunities outside the livestock sector. Second, the weight of emphasis has shifted from a focus on boundaries towards the empowerment of groups. At a broad level, the two are not incompatible. Land-use zoning needs to be an integral aspect of unified land information systems, so that the boundaries between zones (pastoral/ agro-pastoral, agriculture/ forestry, agriculture/ peri-urban, etc) may be revised periodically, based on transparent criteria and procedures agreed by all concerned stakeholders. Within such "contractually flexible boundaries", ongoing land allocation and management decisions governing grazing lands should be devolved to groups constituted at the appropriate level. Detailed specification of tenure rules at the level of national policy will in most instances be unnecessary and even counter-productive. Rather, enabling legislation may simply provide for the rights and responsibilities of user groups to regulate resource access in the areas within their jurisdiction. Of utmost importance, however, is the principle that groups should be self- identified. South Africa’s Communal Property Associations Act provides a useful example of such enabling legislation from which other countries could learn. Third, actors in multiple layers are involved in community-based approaches to pasture land management. Vesting control of the use of the resource at one level of community does not preclude delegation of actual management of the resource to smaller groups or families under contract. There also remain important roles for national and sub-national levels of government, within a nested approach to land tenure and resource access that are commonly referred to as “co-management”. User groups for some activities, such as livestock marketing, may be located

61 at a relatively small scale level, and may federate upwards for other functions such as periodic land use allocation or conflict management. Lingering suspicion of collective approaches in certain transition economies (e.g. Mongolia, Central Asian republics, and to a growing extent China) may lead Governments to consider an unduly narrow range of institutional and policy options, and rangeland professionals have an important role to play in offering lessons from international experience to help keep all possible lines of communication open. Finally, this paper has touched on a number of themes that highlight the importance of a theoretically and empirically informed approach to rangeland policy. The relevant agenda of issues that need to be taken into account in understanding the interplay between science, policy, and practice has broadened considerably over recent years, and now includes attention not only to range ecology, but also to debates concerning land tenure and policy; poverty and vulnerability; biodiversity conservation; indigenous peoples’ and ethnic minority rights; and the political economy of international development, globalization and national sovereignty. Rangeland professionals have a vital role to play in contributing to the formulation of theoretically and empirically informed policy and its practical realization on the ground. The degree to which they are successful in doing so, however, is likely to depend increasingly on their ability to position themselves within this broader and shifting set of policy debates. Table 1 suggests some appropriate roles for international donor support in contributing to the development of rangeland tenure and policy, virtually all of which would benefit considerably from input from rangeland science and rangeland professionals. Among these, particular attention is drawn to: (i) assisting in the technical design of community-based pilots in adaptive co-management, in order to test what will and what will not work in practice, together with monitoring of outcomes on the ground, and disseminating information on good practices, focusing on the lessons of broader relevance; (ii) detailed, empirical research to evaluate the impact of policies in practice. This is particularly important where common policies appear to be implemented across widely varying ecological conditions and rural livelihood systems, in order to identify opportunities to tailor policies to suit particular “recommendation domains”; and (iii) integrating lessons from international experience into national research programs on livestock and pastoral development in developing countries, and in the preparation of appropriate teaching and training materials in national agricultural universities. Underlying these efforts, rangeland professionals may also need to offer technical advice to development practitioners to help identify possible thresholds between “recommendation domains.” Beyond certain thresholds, common property regimes in rangelands may be so compromised that there may be little alternative but to invest in the external inputs required to ensure that individual appropriation of the resource can work, and to accept that additional interventions may be necessary to deal with the consequences for risk management, equity and sustainability of such "second-best" outcomes. The challenge in this case is in knowing when and where such thresholds are being approached.

62 Table 1. Appropriate roles for donor agencies and rangeland professionals in support of rangeland tenure and policy. (Source: adapted from Grell and Kirk (2000)) Issues and levels Policy formulation and Applied research Adaptive approaches adoption International and Networking among donor Demystifying links Disseminating information regional agencies and development between livestock and the on ‘good practice’ case practitioners on pastoral environment studies development Comparative analysis of Influencing international the evolution of pastoral negotiations to benefit land tenure systems pastoralists National National-level policy Analyzing policy and Support for pastoral dialogue on land policy evaluating past experiences organizations in lobbying and advocacy Support for drafting and Encourage integration of implementation of lessons learned in national Promotion of procedural appropriate tenure policies research programs on approaches in law making livestock and pastoral Technical assistance and Technical assistance in development financial support for preparation of teaching and pastoral development professional training administration materials Local Incorporate principles of Selection of pilot areas for Facilitate learning through mobility and reciprocity in action-research following roundtables and strategies for decentralized setting of broad policy participatory approaches rural development directions Concentrate on ‘focal point Support appropriate tenure Identify minimum management’ and rights of frameworks and conditions for access to key resources management institutions strengthening of existing Support to ensure where they are shown to be common property regimes economic viability of effective pastoral institutions Support pastoral groups in negotiation and advocacy

63

Box 2. Vulnerability to dzud in Mongolia

For two years in a row, Mongolia has suffered winters so harsh that the livelihoods of up to a third of the population have been placed in jeopardy through the loss of the livestock on which they depend. Such events are known in Mongolian as "dzud": a range of winter weather-related conditions, typically involving heavy accumulations of snow or ice crusts covering pasture, that prevent domestic animals from obtaining forage from open grazing. Dzud conditions are exacerbated by drought in the preceding summer, which occurred both in 1999 and 2000, since less forage is then available for over-wintering animals. It is tempting to view recent dzud events simply as "natural disasters", but this would be inaccurate; their severe consequences owe as much to institutional failures as to Mongolia’s harsh environment. Mongolia’s economic transition in the 1990s has so far resulted in a higher proportion of the overall population being exposed to dzud risk, and a higher level of vulnerability to dzud for those exposed to such risks. Changing patterns of pastoral land use and land tenure are associated with increasing vulnerability to dzud, and the Bank is assisting Government of Mongolia to address the issue as part of its broader poverty reduction strategy. The share of Mongolia’s total population of 2.3 million engaged in livestock production doubled from around 17% in 1989 to around 35% by the mid-1990s, largely owing to the lack of alternative employment opportunities for those who lost public-sector jobs in the early stages of economic transition. The livestock sector effectively acted as an economy-wide safety net in absorbing many of those who would otherwise have been unemployed. The incidence of poverty increased sharply over the same period, from virtually no officially recorded poverty at the end of the 1980s to around 36% of the population by 1995. Meanwhile, formal marketing systems collapsed, so that urban dwellers relied mainly on their rural relatives as a source of animal food products. Herders preferred to increase their herds rather than participate in the market on unfavorable terms. Annual off-take rates declined sharply so that many more animals were carried over the harsh winter/spring season. While herd sizes steadily grew (from a steady 25 million head until the mid-1990s to 33 million head by 1999), and the number of herding families increased, available feed supplies dwindled with the near-collapse of formerly subsidized crop production. Economic transition also brought new pressures within the pastoral livestock sector itself, which were manifested in the form of reduced pastoral mobility. This in turn contributed to overgrazing around settlements and close to major transport routes, thereby threatening the sustainability of pastoral livestock production overall. Some of the most important factors included: the removal of subsidies in the provision of health and education services, leading to a decline in their spatial coverage; rising inequality in flock and herd sizes among herding households following privatization, owing to differences in labor endowments, levels of skill and experience in herding, and extent of social networks (particularly with local officials) that could bring privileged access to inputs and good grazing; constraints faced by poorer herders in gaining access to vehicles or draught animals with which to move camp; reluctance of more established herders to move camp for fear that their customary pastures may be grazed by others in their absence; and a general decline in the observance and effectiveness of customary norms regarding common grazing, such as respecting others’ customary claims on winter camp sites and associated pastures. This has led to a decline in seasonal separation of pasture use, and a corresponding rise in year-round grazing of particular pastures. The result has been rising congestion, overlapping and contested claims over pastures, leading to open conflict in areas of higher population density. Pasture privatization is prohibited under Mongolia’s 1992 Constitution, and the 1995 Land Law upholds the principle that all pasture land (80% of national territory) should continue to remain in public ownership and held in "common", while customary patterns of seasonal pasture rotation should be observed. Although in principle the legislative framework is highly permissive of continued mobility, the major weaknesses relate to its lax practical implementation (Fernandez-Gimenez and Batbuyan 2000). In part this is owed to ambiguities in the law itself (e.g. it is not clear whether designation of pasture land as "commons" is intended to suggest open access or true, controlled- access commons held by particular herding communities), and in part to the new pressures within the pastoral livestock sector identified above.

64 Box 2. Continued. The Bank’s Country Assistance Strategy for Mongolia outlines a selective approach, focusing on the conditions needed to foster private sector-led growth (macro-economic policy, banking and enterprise reform, energy and transport), and supporting a National Poverty Alleviation Program (NPAP, 1996-2000). NPAP, a type of social fund, aimed to improve the provision of health and education services in rural and urban communities, and promoted income generation for the poor through micro-credit and labor-intensive public works. Over 1999/2000, the Bank undertook several activities intended to re-focus its program of poverty-targeted intervention in Mongolia. First, an evaluation of NPAP supported by Government, UNDP and World Bank in 1999 concluded that while much had been achieved under the program, its impact in reaching the rural poor had fallen short. Second, the Bank supported analytical and advisory work to assist the Government Working Group on Land Reform in identifying key constraints in land tenure policy, including constraints in access to grazing that worsen vulnerability among poorer herding households. Third, the Bank (with co-financing from DFID) supported a Participatory Living Standards Assessment (PLSA) in 2000 as an input to the preparation of Mongolia’s Poverty Reduction Strategy. The PLSA has considerably deepened understanding of vulnerability and other dimensions of poverty in Mongolia, and documented the perceptions and experiences of around 2,000 rural and urban community members with respect to changing living standards and livelihood strategies in the 1990s. These activities have helped to re-focus the Bank’s strategy in Mongolia, and to shape a follow-on program currently being prepared for possible Bank support. The proposed Sustainable Livelihoods Project (SLP), intended as the first phase of a 12-year Adaptable Program Loan, would provide support in three, inter-related areas: pastoral risk management, rural micro-finance services, and community-driven investments in small-scale infrastructure. Under the proposed SLP, efforts are envisaged to assist Mongolia to restore institutional capacity to manage risk in pastoral livestock production, with a primary focus on risk preparedness. A wide range of measures are being considered for support in pursuit of this goal, several of which concern pasture land tenure and management: ?? early-warning systems for predicting drought and dzud incidence; ?? clarification of institutional roles and responsibilities in dzud response; ?? marketing arrangements for livestock and livestock products, in part to permit rapid destocking at supported prices prior to dzud and drought; ?? emergency grazing reserves and fodder banks; ?? community-based pasture land tenure and management arrangements to ensure the equitable implementation of key provisions of the existing land law, including ?? capacity-building for multi-purpose grazing associations, and for negotiated dispute and conflict resolution; ?? developing sustainable approaches to livestock insurance based on district-level indexing of trigger events using weather and/or livestock mortality data. In addition to these efforts to support pastoral risk management, the SLP would aim to broaden the capital assets of the rural poor so as to expand opportunities for livelihood diversification beyond livestock production. Lessons learned point to the importance of investment in ways to make micro-finance services, including savings and insurance products as well as credit, available to the rural poor; and support to enable rural communities themselves to prioritize and manage needed investments in small-scale infrastructure such as wells, rural access roads, and facilities for health and education services. The PLSA confirmed, among other issues, that rural communities in Mongolia place high priority on ensuring that patterns of investment and forms of service provision remain compatible with mobile livestock production on pasture land held and managed as controlled-access commons, which is widely recognized as the key to sustainability. Sources: Mearns (1996); Fernandez-Gimenez and Batbuyan (2000); Hanstad and Duncan (2001); Skees and Enkh-Amgalan (2001); NSO and World Bank (2001); project documentation

65 Box 3. The case for locally differentiated approaches to grassland management in western China China has been very successful in reducing the poverty it embarked on market-orientated reforms. The number of people living below the official poverty line fell from 260 million in 1978 to 42 million (5% of rural population) in 1998. Rural poverty is now concentrated in isolated pockets, many of which coincide with the grasslands of western China. They include some of the most degraded grasslands in China: an overlap of around 80% has been observed between ecologically sensitive areas and poverty counties located in the same regions, and poverty rates in grassland provinces remain well above the national average of 6.3% (e.g. Xinjiang 27%, Gansu 23%, Yunnan 23%, Tibet 10%, Inner Mongolia 9%, and Sichuan 7%). The challenge in reducing poverty still further in rural China depends critically on being able to sustain and enhance the livelihoods of livestock keepers in these grassland provinces, while at the same time ensuring the sustainable management of the grasslands on which they rely for a living. This twin challenge is to be tackled through China’s recently announced "Great Western Development Plan". Mobile pastoral production systems have come under increasing threat. While true pastoral systems may still be found in northern Xinjiang or on parts of the Tibetan Plateau, and are usually practiced by ethnic minorities such as Kazakhs, Tibetans, and Mongols, the trend over the last 15-20 years has been towards semi-pastoral and mixed farming systems. In earlier decades, policies were heavily geared towards land "reclamation", involving the conversion of natural grassland to arable cropland, with the common result that large areas were left salinized or vulnerable to wind erosion. The higher quality, better-watered grasslands were the first to be converted with a disproportionately adverse effect on the sustainability of pastoral production overall. This policy-induced process was commonly accompanied by in-migration, particularly by agriculturalists, which fragmented the area available for open grazing further. Grassland ecosystems in Inner Mongolia were particularly severely affected. More recently, considerable investment is being made to intensify livestock production in western China through the so-called "Four-Way Plan" of building winter shelters, improving water supply (drilling wells and gravity-flow irrigation), growing fodder crops, and fencing pastures allocated to individual households. Known as the cao kulun system in Inner Mongolia, it has other regional variants. In principle, such investments permit higher output per ha, but the implications for sustainable grassland management remain unclear; the approach does little to ensure the inclusion of poor livestock keepers who cannot access capital for the required investments; and it could increase production risk. In spite of some regional adaptation of the technical package, the same general approach is being promoted more or less uniformly throughout formerly pastoral areas of western China, regardless of the wide diversity in underlying ecological conditions, livelihood opportunities, and socio-cultural systems. In spite of this “uniform” approach at the national level, significant and often creative adaptations occur in the ways centrally planned policies are implemented within particular provinces. This accounts for greater continuity in grassland management practices than might be imagined from official accounts of management intensification. A recent Bank-supported study of grassland management in Inner Mongolia, Xinjiang, and Gansu provinces, revealed a wide range of institutional innovations around grassland tenure and management, often building on customary antecedents. Important lessons can be learned from such examples, which suggest that an immediate priority is for provincial and local governments to pilot-test and tailor policies and interventions to encourage sustainable grassland management to local ecological, social, and economic conditions. Pasture privatization and enclosure is encouraged under the 1985 Grassland Law. A gradual approach was envisaged, beginning with the allocation of pastures to individual households under the Household Responsibility System, followed by the assignment of pasture carrying capacities, and finally the introduction of incentives and sanctions to comply with the assessed stocking limits. Grassland use contracts have a 50-year term, and fees are based on estimated grassland productivity. Around 80% of natural pastures were reportedly contracted to households in Inner Mongolia by 1990, and as much as 95% in Xinjiang by 1998, but in practice group tenure remains widespread. Where individuated grassland allocation has been strictly implemented, especially in drier areas, it has typically increased conflict within communities over grazing rights, and reduced the capacity to respond to risk through mobility and flexibility of access to grazing resources, with considerable cost to long-term sustainability. Significant ambiguities exist in China’s Grasslands Law, notably in defining state versus collective ownership. The law conflicts with other statutes on the question of sub-contracting and the sale of grassland use rights, which the Grassland Law prohibits but the 1983 Agriculture Law and 1991 Land Management Law allow. Evidence from Gansu and elsewhere shows that sub-contracting or leasing of grazing rights can ensure equity under contracting, by allowing poorer households to lease out spare pasture to households with larger herds, thereby offering livelihood options to a wider range of households while maintaining sustainable stocking rates on natural grasslands.

66 Box 3. Continued. Sources: Banks (1999, 2001), (Banks and Sheehy 2000), Brandenburg (2000), Dunlop (2000), Ho (2000, 2001), Williams (1996), World Bank (2000, 2001c), Mearns (2001)

Literature Cited Banks, T. 1999. State, community and common property in Xinjiang: synergy or strife? Development Policy Review 17(3): 293-313. Banks, T. 2001. Property rights and the environment in pastoral China: evidence from he field. Development and Change 32(4): 717-740. Banks, T., and D. Sheehy. 2000. Executive Summary Report of China Grasslands Study. Unpublished report to the World Bank. Banzhaf, M., B. Drabo, and H. Grell. 2000. From Conflict to Consensus: Towards Joint Management of Natural Resources by Pastoralists and Agro-Pastoralists in the Zone of Kishi Beiga, Burkina Faso. Securing the Commons No.3. IIED, SOS Sahel, and GTZ, London. Behnke, R. 1994. Natural resource management in pastoral Africa. Development Policy Review 12: 5-27. Behnke, R., I. Scoones, and C. Kerven (eds). 1993. Range Ecology at Disequilibrium: New Models of Natural Variability and Pastoral Adaptation in African Savannas. ODI, London. Bruce, J. W., and R. Mearns 2002 (forthcoming). Natural Resource Management and Land Policy in Developing Countries: Lessons Learned and New Challenges for the World Bank. IIED, London. Bryceson, D. 2000. Rural Africa at the crossroads: livelihood practices and policies. Natural Resource Perspectives 52. ODI, London. de Haan, C., et al. 1994. An overview of the World Bank’s involvement in pastoral development. Pastoral Development Network Papers No. 36b. ODI, London. de Haan, C., H. Steinfeld, and H. Blackburn. 1998. Livestock and the Environment: Finding a Balance. European Commission, FAO and World Bank, Brussels. de Haan, C., T. van Veen, B. Brandenburg, J. Gauthier, F. Le Gall, R. Mearns, and M. Siméon. 2001. Livestock Development: Implications for Rural Poverty, the Environment, and Global Food Security. Directions in Development Series. World Bank, Washington, DC. Dunlop, S. 2000. Conquest and Change: Mongol Herding in Xilingol, Inner Mongolia. PhD thesis. Australian National University, Canberra. Ellis, F. 1998. Household strategies and rural livelihood diversification. J. of Development Studies 35(1): 1-38. Fernandez-Gimenez, M., and B. Allen-Diaz. 1999. Testing a non-equilibrium model of rangeland vegetation dynamics in Mongolia. J. of Applied Ecology 36: 871-885. Fernandez-Gimenez, M., and B. Batbuyan. 2000. Law and disorder in Mongolia: local implementation of Mongolia’s land law. Paper presented at Eighth Biennial Conference of

67 the International Association for the Study of Common Property, Bloomington, Indiana, May 31-June 4. Fratkin, E., and R. Mearns (forthcoming). Sustainability and pastoral production: lessons from East Africa and Mongolia. Human Organization. Grell, H. and M. Kirk. 2000. The role of the donors in influencing property rights over pastoral resources in sub-Saharan Africa, In: N. McCarthy, B. Swallow, M. Kirk, and P. Hazell (eds), Property Rights, Risk and Livestock Development in Africa. IFPRI and ILRI, Washington, DC and Nairobi:. Hanstad, T., and J. Duncan. 2001. Land Reform in Mongolia: Observations and Recommendations. RDI Reports on Foreign Aid and Development No. 109. Seattle: Rural Development Institute for the World Bank. Hendrickson, D., R. Mearns, and J. Armon. 1998. The changing nature of drought and famine vulnerability: the case of livestock raiding in Turkana district, Kenya. Disasters 22(3): 185- 199. Ho, P. 2000. The clash over state and collective property: the making of the Rangeland Law. The China Quarterly 161: 240-263. Ho, P. 2001. Who owns China’s land? Policies, property rights and deliberate institutional ambiguity. The China Quarterly 162: 394-421. Humphrey, C. and D. Sneath, 1999. The End of Nomadism? Society, State and the Environment in Inner Asia. Duke University Press, Durham, NC. Knox, A., R. Meinzen-Dick, and P. Hazell. 1998. Property Rights, Collective Action and Technologies for Natural Resource Management. CAPRi Working Paper 1. IFPRI, Washington DC. Lane, C (ed). 1998. Custodians of the Commons: pastoral land tenure in East and West Africa. Earthscan Publications, London. Lane, C., and R. Moorehead. 1994. New directions in rangeland and resource tenure and policy, in I. Scoones (ed.), Living with Uncertainty. IT Publications, London. Leach, M., and R. Mearns (eds). 1996. The Lie of the Land: Challenging Received Wisdom on the African Environment. James Currey, Oxford. Leach, M., R. Mearns, and I. Scoones. 1999. Environmental entitlements: dynamics and institutions in community-based natural resource management. World Development 27(2): 225-247. Margules, C.R., and R.L. Pressey. 2000. Systematic conservation planning. Nature 405: 243- 253. Mearns, R. 1996. Community, collective action and common grazing: the case of post socialist Mongolia. J. of Development Studies 32(3): 297-33. Mearns, R. 1997a. Livestock and environment: potential for complementarity. World Animal Review 88: 2-14. Mearns, R. 1997b. Balancing livestock production and environmental goals. World Animal Review 89: 24-33.

68 Mearns, R. 2001. Contextual factors in the management of common grazing lands: Lessons from Mongolia and northwestern China, In: Proceedings of the XIX International Grassland Congress, São Pedro, Brazil, February 11-21, 2001. Brazilian Society of Animal Husbandry, Piracicaba. Niamir-Fuller, M. (ed). 1999. Managing Mobility in African Rangelands: The Legitimization of Transhumance. IT Publications, London. NSO and World Bank (2001). Mongolia Participatory Living Standards Assessment 2000. Mongolian National Statistical Office, Ulaanbaatar. Øygard, R., T. Vedeld, and J. Aune. 1999. Good Practices in Drylands Management. World Bank, Washington, DC. Pratt, D., F. Le Gall, and C. de Haan. 1997. Investing in Pastoralism: sustainable natural resource use in arid Africa and the Middle East. World Bank Technical Papers No. 365. World Bank, Washington, DC. Rutten, M. 1992. Selling Wealth to Buy Poverty: the process of individualization of landownership among the Maasai pastoralists of Kajiado District, Kenya, 1890 1990. Verlag breitenbach, Saarbrücken and Fort Lauderdale. Sandford, S. 1983. Management of Pastoral Development in the Third World. John Wiley & Sons, Chichester. Scoones, I (ed). 1994. Living with Uncertainty: New Directions in Pastoral Development in Africa. IT Publications, London. Shanmugaratnam, N., T. Vedeld, A. Mossige, and M. Bovin. 1992. Resource management and pastoral institution building in the West African Sahel. World Bank Discussion Papers No. 175. World Bank, Washington, DC. Skees, J., and A. Enkh-Amgalan. 2001. Examining the feasibility of livestock insurance in Mongolia. Draft Policy Research Working Paper. World Bank, Washington, DC. Whitten, T., D. Holmes, and K. MacKinnon. 2001. Conservation bio logy: a displacement behavior for academia? Conservation Biology 15(1): 1-3. Williams, D. M. 1996. Grassland enclosures: catalyst of land degradation in Inner Mongolia. Human Organization 55(3): 307-313. World Bank. 2000. China: Overcoming Rural Poverty. World Bank, Washington, DC. World Bank. 2001a. World Development Report 2000/2001: Attacking Poverty. World Bank, Washington, DC. World Bank. 2001b. Making Sustainable Commitments: an Environment Strategy for the World Bank. World Bank, Washington, DC. World Bank. 2001c. China: Air, Land, and Water – Environmental Priorities for a New Millenium. World Bank, Washington, DC. WRI. 2000. Grassland ecosystems: sustaining the steppe – the future of Mongolia’s grasslands, pp212-224 In: World Resources Report 2000-2001. Washington, DC: World Resources Institute with UNDP, UNEP, and World Bank.

69