section 2
PERSPECTIVES ON AGROFORESTRY

Chapter 4
Institutional aspects of agroforestry research and development

Bjorn O. Lundgren
Director
International Council for Research in Agroforestry (ICRAF)
P.O. Box30677, Nairobi, Kenya

Introduction

The history of agroforestry as a science and as a focus for systematic development efforts is very short—fifteen years at the most (see King, this volume). In 1982—the "middle ages" of this short history—the present writer was asked by the Office of Technology Assessment of the United States Congress to make an evaluation of the role of agroforestry in improving tropical lands. The forecasts contained in that report (Lundgren, 1982a) regarding the likely developments in and constraints to agroforestry over the coming five to ten years generally seem to have been correct. Interest in agroforestry is increasing rapidly among scientists, land-use experts and development professionals; resources for research and development are being made available from donors and national institutions at an unprecedented level (although they are still modest in absolute terms); concrete results from R & D programmes are just starting to emerge on a significant scale; and the next three to five years will see an information explosion in agroforestry. These developments in general, and the progress made in specific fields and regions, are highlighted in other contributions to this volume.

Another assessment contained in the report mentioned above was that the main constraints to a full realization of the potential of agroforestry were of an institutional nature and related to the rigid disciplinary compartmentalization which characterizes institutions working in the field of land use. There have been very few signs in the last five years of this situation changing for the better. On the contrary, it is more urgent than ever that these institutional questions are addressed at the highest possible levels, both in individual countries and internationally. If effective and relevant institutional arrangements are not developed for implementing agroforestry R & D programmes on a large scale within the next five to ten years, the risk is very real that the potential of agroforestry will never be fully realized.

This article deals with the institutional aspects of agroforestry. It presents the writer's personal thoughts and should certainly not be seen as an ICRAF policy statement.

Land-use institutions today

The basic institutional structures established to deal with the use of land in virtually all the countries of the world today originate from temperate Europe and North America. There, in the late nineteenth and early twentieth centuries, the modernization of agriculture and forestry, which was necessitated by and dependent upon the rapid industrialization, led to the gradual emergence of government and private institutions to support the land users. Crop production and industrial wood production, which were carried out on separate types of land, required different professional skills, had different aims, and very often were managed by different owners (farmers versus governments or private companies). It was entirely rational, therefore, that agricultural and forestry institutions developed independently of each other. In the few cases where it was deliberately planned that trees, crops and/ or animals should interact in specific technologies or land-use practices, for example in windbreaks, shelterbelts, hedges, grazing in fruit orchards, or game management for meat production in forests, there was never any difficulty in establishing which institutional sector was "responsible" for the technology or practice. With very few exceptions, anything done on designated farmland, even if it involved tree growing, was (and is) the responsibility of the agricultural/horticultural sector, and any use of forest land, including game management, rational utilization of wild berries, mushrooms, etc., falls under the forestry sector.

As a result of these separate institutional developments, there are today different laws and policies governing agricultural and forest land use; there are separate training, education and research institutions; advice to land users is provided through separate extension services; agriculture and forestry normally fall under different ministries or, if they are under the same ministry, under separate departments.

Another important aspect of the land-use legacy from the industrialized countries is that all policies and disciplinary R & D efforts are aimed at maximizing, in a sustainable way, the output of products per unit of land — this applies as much to wheat, maize, milk and meat as it does to timber and pulpwood. Commercially oriented monocropping dominates the use of land and has been seen as very successful as markets for agricultural and forest products have continuously increased in volume over the last century. Subsistence use of land, in the sense of people being dependent on their own land for food, has virtually disappeared in industrialized countries.

When the European colonial powers established their administrations in Africa and elsewhere in tropical and subtropical regions, the institutional structures, policies and aims related to land use and development used in the home countries were simply copied in the colonies. This applied also to those countries which were not colonies, e.g., in Latin America, where governments chose to adopt the industrialized countries' institutional structure in land use. The model has been continued after independence in all tropical countries and the post-war international organizations that were set up to assist the emerging nations in improving and rationalizing the use of their land resources are all oriented on conventional-discipline lines. There is little doubt that the concentration of R & D efforts on the particular commodities, technologies and practices which the mono-disciplinary approach leads to, has resulted in some remarkable success stories in tropical countries during this century. Cultivation of export crops such as coffee, tea, oil palm, fruits and spices forms the mainstay of many developing-country economies. Some countries have achieved self-sufficiency in industrial wood production by systematically building up plantations of exotic and indigenous trees, and, most remarkable of all, the "green revolution" of the last two to three decades has turned previously food-deficit countries into major grain exporters. Although all these developments have had their share of criticism (some of it justified, but most based on ignorance) from economists, social scientists and environmentalists, their technical and economic success is an undisputed credit to all the horticultural, silvicultural and agronomic scientists and R & D institutions behind their development.

In spite of the relative successes achieved in agriculture, forestry and other disciplines in the tropics and subtropics (developing countries), there are, quite obviously, many aspects of land use which are not successful. Food production per capita has been decreasing in most of Africa over the last 25 years; man-made desert conditions are spreading at an alarming rate; erosion and flooding, largely as a consequence of defective land use following deforestation, cause unprecedented loss of farmland; rural populations in more and more areas are affected by existing or imminent energy crises due to lack of fuelwood; and the effects of naturally occurring droughts, in terms of loss of human life and livestock, become more and more devastating with time. It is open to discussion whether the main causes of these conditions are demographic, political, technological, economic or environmental. It is generally agreed, however, that there is no single cause but rather a complex interaction among several factors — interactions which are different from place to place and country to country.

I have become increasingly convinced that a significant contributor to the failure to solve many important land-use problems is the inappropriateness of conventional-discipline-oriented institutions for identifying and addressing real problems in land-use systems in most tropical and subtropical (developing) countries. This particularly applies to the multitude of subsistence or mixed subsistence/ cash farming and pastoral systems in which the vast majority of rural land users live and from which they eke out a living. They differ, of course, with ecological and socio-cultural conditions, from the purely nomadic pastoral systems of the arid to semi-arid zones to the sedentary mixed-farming systems on upland soils in subhumid areas and the shifting cultivation in the humid zones. Some features are shared by most of them, e.g., low cash incomes and, hence, little ability to invest in improvements requiring money, and marginal ecologies, such as infertile or erosive soils, or marginal climates. Minimizing the risk of crop failure or animal loss in these situations is a much more important concern than increasing yields. Often, land tenure is insecure or non-existent. Common lands, which often serve as important sources for fuelwood, building material, grazing, etc., are diminishing in area or are being degraded through overuse as a result of increasing populations.

With few exceptions, subsistence.or near-subsistence farming systems are mixed in the sense that the fanner produces not only the bulk food crop from the land but also specialized food crops (vegetables, fruits, spices, etc.), animals for meat, milk or draught power, and, very often, trees and shrubs for fuel, fodder and building material. Cash income sometimes comes from specialized crops intended for sale but normally from surpluses of the "subsistence" crops and animals. Obviously, the relative importance of each of these components of the system varies, but they all serve to satisfy basic needs of the land user (food, shelter, energy, cash, etc.), and they all interact economically and/or ecologically in that they are managed by the same limited labour resource and they share the same farm environment (soil, water, topography). The subsistence land user's strategy and aims are to use his labour and land resources to optimize, with minimum risk, the production of various products and services required to satisfy all his basic needs.

Why are today's institutions inadequate?

The fundamental inadequacy of conventional-discipline-oriented institutions lies in the failure to acknowledge and understand these basic facts, strategies and aims, and in the inability to adapt to them. The aims, infrastructure, rationale and philosophy of these institutions, as well as the training of their experts, are geared to the maximization of individual components, be they food crops, cash crops, animals or trees. There is little understanding that the land user needs to share out his resources for the production of other commodities or services. Even if such an understanding appears to exist—the forester may generously agree that food production is essential and the agronomist may not disagree that fuelwood is needed — there is very little technical comprehension of the requirements for producing commodities outside one's own discipline.

The inability of technical institutions and experts to understand how social, religious, cultural and traditional beliefs and preferences can nullify a convincing cost/benefit analysis on, say, increased fertilizer use, upgrading of cattle or establishment of fuelwood plantations, has been so well documented by social scientists that there is no need to elaborate the point.

It is, quite obviously, this lack of understanding 01 the complexities of many land-use systems, which is built in to the very foundation of conventional institutions, that is the basic cause of the many failures and frustrations in trying to solve the land-use problems of developing countries today. Land users are generally not conservative or "primitive"; they are not adamantly unwilling to improve their lot; they are not opposed to increasing their yields or cash incomes; nor are they hostile to planting and caring for trees on their land — provided we understand how technology changes and improvements fit into their problems, priorities, beliefs and aims. The situation today can be summarized as in Figure 1, which is derived from an idea by R. Chambers (personal communication).

There are a large number of more or less narrowly specialized sectors, disciplines, institutions, scientists and experts, laws, policies, etc., all dedicated to maximizing (within limits known to and identified by themselves) "their" product, i.e., which look at one segment of the land-use system only.

There is no institution today which has both the mandate and the competence to identify solutions to land-use problems based on an interdisciplinary analysis of interactive constraints and potentials within land-use systems, and the power to assign resources in a way that will cut across institutional boundaries in order to implement such solutions.

Figure 1 Institutional dilemma of land development? Disciplinary institutions look at a specific compartment of a land-use system only and normally do not consider the complex interactions within the system

There are, nevertheless, a few positive signs that awareness of the need to address the institutional constraints to real problem-solving is increasing. For example, the recently published report of the World Commission on Environment and Development (the "Brundtland Commission") (1987) concluded:

The integrated and interdependent nature of the new challenges and issues contrasts sharply with the nature of the institutions that exist today. These institutions tend to be independent, fragmented, and working to relatively narrow mandates with closed decision processes.

Less encouraging, however, is the fact that virtually all recent policy and planning documents from international institutions, even if they pay lip service to the need for multidisciplinarity, integrated approaches and holistic views, end up making conventional recommendations such as increased use of fertilizers, irrigation and genetically improved crop varieties, increased tree planting, etc. None contain any critical analysis of the adequacy of existing institutions for addressing the totality of the problems and for contributing to their solution. This applies as much to agricultural (TAC, 1987) and forestry (Adams and Dixon, 1986; Carlson and Shea, 1986) research, as to development statements, again both agricultural (FAO, 1986a, 1986b, 1987; OAU, 1985) and forestry (FAO, 1985; WRI, 1987).

One reason for the slowness with which the institutional problems are being addressed has probably been the lure for the statistics on "global problems" which have emerged over the past 15 years, e.g., on environmental and developmental problems such as desertification, deforestation, the fuelwood crisis and declining food production. As a result of improved techniques of survey, monitoring and other statistical methods, it has been possible to break down an almost infinite number of complex, local problems into their component effects and to add up these effects into global perspectives. Since it is easy to express the magnitude of these problems in conspicuous and alarming global figures, such as the numbers of hectares of forests that are lost, or land that is turned into desert, or the distance people will have to walk every day to collect their fuelwood, it becomes imperative and even very attractive to politicians, decision-makers and institutions to look for simple solutions to these problems and to extrapolate the likely benefits to a global scale. Disciplinary-sector institutions have been very successful in exploiting this situation to increase resource allocation to themselves by promising simple solutions to what appear to be straightforward problems — more tree plantations will solve the fuelwood and deforestation problems; more fertilizers and irrigation will increase food production, and so on.

The fact, however unappealing and complex it may be, is that just as the "global problems" are the sum of the effects of a large number of local problems, the solutions can only be achieved by adopting an equal number of sound land-use practices and political and economic measures (Lundgren, 1985). Problems must be identified, diagnosed and solved where they occur. It does not help small-scale mixed farmers in district x of country y, to know that 300 million people in the developing world do not have adequate supplies to meet their fuelwood or protein requirements. There is an urgent need to re-think and re-evaluate the situation. The conclusions arrived at must direct the relevant institutions from the discipline-oriented maximization thinking that is fuelled by global statistics to the multidisciplinary optimization thinking that is geared to solving local problems.

Agroforestry as a catalyst for change

When agroforestry was institutionalized through the creation of ICRAF in 1977, there were very few people who thought of the subject as anything but an off-shoot from the forestry sector. Indeed, the early ideas and concepts originated with tropical foresters who were concerned about the poor contribution that the forestry sector made to the well-being of rural populations other than those directly involved in forestry operations. The long and basically positive experience of taungya-type agrisilvicultural systems on forest land had demonstrated to foresters that timber and food-crop production from the same land was possible (King, this volume). In the early stages, agroforestry was seen as the forestry sector's contribution to agriculture and many foresters still think of it in that way. There were no serious efforts to integrate, forestry, or rather tree growing, into agricultural practices, let alone any critical analysis of whether the existing forestry institutions were competent to take trees outside the forests.

It was only in the early 1980s that agroforestry developed into a truly integrated and interdisciplinary approach to land improvement, mainly through ICRAF's conceptual and methodological work (Lundgren, 1987b; Steppler, this volume). A more objective definition of agroforestry than previous ones, and one that ICRAF has used since the early 1980s (Lundgren, 1982b), is as follows:

Agroforestry is a collective name for all land-use systems and practices in which woody perennials are deliberately grown on the same land management unit as crops and/or animals. This can be either in some form of spatial arrangement or in a time sequence. To qualify as agroforestry, a given land-use system or practice must permit significant economic and ecological interactions between the woody and non-woody components.

This definition clearly underlines the integrated nature of the approach.

The aim of agroforestry is (or should be) to optimize the positive interactions between components in order to achieve a more productive, sustainable and/ or diversified (in relation to the land users' need) output from-the land than is possible with other forms of land use. It is obvious that, with this definition and aim, agroforestry as a science and practice must cut across conventional institutional areas and draw upon several disciplines in the social, production and environmental sectors (see Figure 1) if its full potential for improving land use is to be realized.

ICRAF has built up a truly multidisciplinary team of experts and scientists, representing all the relevant disciplines deemed necessary to study all aspects of a land-use system. Through this team, objective analytical methods of identifying problems in land-use (farming) systems, and potentials for their solution that are not restricted to agroforestry solutions, have been developed (Raintree, 1987). Using an analogy from Figure 1, this diagnostic and design (D&D) methodology means that instead of the individual experts standing outside the land-use system and observing their own disciplinary components of the system, the whole team, without any preconceived ideas about the nature of the problems and their potential solutions, "parachutes" into the middle of the complex system and tries to diagnose causal mechanisms behind the problems and interactions between the components of the system. From this diagnosis, technologies with a potential for solving the problems are designed. It is only at this stage that the role of different disciplinary sectors and institutions in developing these technologies can be defined.

When putting this methodology into practice in collaborative research programmes with national and international institutions, ICRAF has been faced with a host of problems related to the compartmentalization of conventional disciplinary institutions and professions, ranging from direct mistrust and lack of appreciation of each other's expertise to the enormous difficulty of arriving at decisions about resource allocation in programmes involving different institutions from different ministries.

In the Agroforestry Research Networks for Africa (AFRENA) programme, ICRAF has developed a model for inter-institutional collaboration at national and regional levels which, at least in its early stages, has been very successful (Torres, 1986). The key elements of the approach are to stimulate and assist (technically and, if necessary, financially) national forestry, agriculture and other institutions to work together in analysing land-use problems and designing research programmes to solve them, and to define their exact role in implementing such programmes. It is essential that participation in these activities is voluntary for the different institutions, that it is seen as professionally .stimulating and enriching and that it does not initially "threaten" existing institutional power structures. The technically sound results across disciplinary boundaries that are starting to be produced from such semi-formal programmes show that this approach certainly has potential. Once their usefulness has been confirmed we will be ready to take the next steps required to achieve more appropriate institutional functions and structures for addressing the problems of the small-scale subsistence farmers and land users of the tropics and subtropics.

Agroforestry as a discipline has the potential for taking a leading and catalytic role in this process of change, because of its inherent integrative and multidisciplinary nature, its optimization rather than component-maximization aims, and because of the great interest shown in it today.

Land-use institutions tomorrow

It would be presumptuous to end this article by proposing an "ideal" structure and set of objectives for tomorrow's land-use-related institutions. No such ideal institution will ever exist because of the enormous variety in conditions, policies and institutions among regions, countries, economies, etc. Some general thoughts on directions may, however, produce some food for thought.

1. In the short and medium term, and within existing institutional structures, collaborative programmes which cut across disciplinary boundaries and address concrete land-use problems must be encouraged and given more support, both at the national level and at the level of international agencies such as donors, UN bodies, and International Agricultural Research Centres (lARCs). It would be appropriate if the international institutions took the initiative in this. Unfortunately, such is not the case today: most international bodies and institutions are firmly entrenched in disciplinary thinking and actions. Although there are a few very encouraging exceptions, integrated approaches still seem far away. With the risk of over-generalizing facts and of stepping on some toes, I feel that it would be encouraging to see the following positive developments on the international land-uses scene:

1. That the institutions within the Consultative Group of International Agricultural Research (CGIAR) accept the fact that trees and shrubs form integral parts of most small-scale food-producing farming systems in the tropical world and that their rational development can enhance both productivity, sustainability and diversity of food production;

2. That FAO's Agriculture and Forestry Departments develop joint projects to show that the UN's main land-use-related agency has the interest, mandate and competence to look at crops and trees, and their interactions, in the same place and in the same land-use systems;

3. That the forestry departments and advisors of bilateral and multi-lateral donor agencies, development banks, research and policy institutes, etc., accept the fact that tree growing on farmland (agroforestry) is not an exclusive forest-sector activity, and that the same donors and institutions ask themselves why they channel all then- support to agroforestry through forestry, energy or environmental programmes rather than through agricultural programmes.

2. It must be strongly emphasized that in the future we will still need specialized institutions and experts — the plea for multidisciplinary, inter-institutional approaches and integrated thinking must not lead to the formation of super institutions manned exclusively by generalists. Undoubtedly there will be a need for crop physiologists, maize breeders, tree geneticists, fertilizer experts, insect ecologists, and so on, and for specialized institutions that provide a working environment for such experts. Even the most "perfect" interdisciplinarily-derived research or development programme must, when being implemented, be broken up into its component parts, and then resynthesized. If there is inadequate expertise on the structure and function of the component parts, then the whole entity (the improved land-use system, or whatever) will never be fully operational and optimized.

3. What will be needed are new institutional functions for problem identification, priority setting and resource allocation, without necessarily making fundamental changes of structure. Such new functions can be created within or between existing land-use ministries and departments, or as independent bodies with existing institutions subordinate to them. It would probably be best to start by creating inter-institutional committees for planning but which will become more and more executive as experience is gained. Depending on the problems to be addressed, these committees can then create appropriate task forces of existing disciplinary institutions and resources.

These may not sound very fundamental changes but they will require fundamental rethinking among disciplinary scientists, institutions and decision-makers. The sooner this process starts, the faster can some of the key land-use problems of the world be solved.

REFERENCES

Adams, N. and R.K. Dixon (eds.). 1986. Forestry networks. Proceedings of the first network workshop of the forestry/fuelwood research and development project (F/FRED), 24-27 September 1986, Bangkok. Washington D.C.: Winrock International.

Carlson, L.W. and K.R. Shea (eds.). 1986. Increasing productivity of multipurpose lands. Proceedings of IUFRO research planning workshop for Sahelian arid North Sudanian zones, Nairobi, January 1986. Vienna: IUFRO.

FAO. 1985. Tropical forestry action plan. Committee on Forest Development in the Tropics. Rome: FAO

——.1986a. A programme of action for African agriculture proposed by the Director-General. Rome: FAO

——.1986b. African agriculture: the next 25 years. Main report (ARC/86/3). Rome: FAO.

——.1987. Summary programme of work and budget 1988-1989. Document COAG/87/5 presented to 9^th Session of the Committee on Agriculture, 23 March-1 April 1987. Rome: FAO.

Lundgren, B. 1982a. The use of agroforestry to improve the productivity of converted tropical land. Report prepared for the Office of Technology Assessment of the United States Congress(mimeo).

——.1982b. Introduction. Agroforestry Systems 1:3-6.

——.1985. Global deforestation, its causes and suggested remedies. Agroforestry Systems 3:91-95.

——.1987a. Agroforestry in third world countries. Paper presented to the IUFRO workshop on agroforestry for rural needs. New Delhi, India, 22-26 February 1987.

——.1987b. ICRAF's first ten years. Agroforestry Systems 5: 197-218.

Organization of African Unity. 1985. Africa's priority programme for economic recovery 1986-1990. Addis Ababa: OAU.

Raintree, J.B. 1987. The state of the art of agroforestry diagnosis and design. Agroforestry Systems 5: 219-250.

Technical Advisory Committee (TAC). 1987. CGIAR priorities and future strategies. Rome:TAC/FAO

Torres, F. 1986. Agroforestry research networks in tropical Africa: an ecozone approach. Paper presented to the First International Conference of Agricultural Research Systems, IFARD, Brasilia, Brazil, 6-11 September 1986.

World Commission on Environment and Development. 1987. Our common future. London: Oxford University Press.

World Resources Institute. 1985. Tropical forests: a call for action. Parts I-III. Washington, D.C.: WRI.