Citation
Journal of farming systems research-extension

Material Information

Title:
Journal of farming systems research-extension
Running title:
Journal for farming systems research-extension
Abbreviated Title:
J. farming syst. res.-ext.
Creator:
Association of Farming Systems Research-Extension
Place of Publication:
Tucson Ariz. USA
Publisher:
Association of Farming Systems Research-Extension
Publication Date:
Language:
English
Physical Description:
v. : ill. ; 23 cm.

Subjects

Subjects / Keywords:
Agricultural systems -- Periodicals -- Developing countries ( lcsh )
Agricultural extension work -- Research -- Periodicals ( lcsh )
Sustainable agriculture -- Periodicals -- Developing countries ( lcsh )
Genre:
serial ( sobekcm )
periodical ( marcgt )

Notes

Dates or Sequential Designation:
Vol. 1, no. 1-
General Note:
Title varies slightly.
General Note:
Title from cover.
General Note:
Latest issue consulted: Vol. 1, no. 2, published in 1990.
Funding:
Electronic resources created as part of a prototype UF Institutional Repository and Faculty Papers project by the University of Florida.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
The University of Florida George A. Smathers Libraries respect the intellectual property rights of others and do not claim any copyright interest in this item. This item may be protected by copyright but is made available here under a claim of fair use (17 U.S.C. §107) for non-profit research and educational purposes. Users of this work have responsibility for determining copyright status prior to reusing, publishing or reproducing this item for purposes other than what is allowed by fair use or other copyright exemptions. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder. The Smathers Libraries would like to learn more about this item and invite individuals or organizations to contact Digital Services (UFDC@uflib.ufl.edu) with any additional information they can provide.
Resource Identifier:
22044949 ( OCLC )
sn 90001812 ( LCCN )
1051-6786 ( ISSN )

Full Text

� Volunic 5, Number


___ 1995














for Farming Systems Research-Extension








Jui A . adcu 0n 00nEsoa

31 Thor' a tc nFR:Cniein h oeo oeln




Tcu 9,d, Joh S. 'advel an ie i Yky A . Hujsa .

910 Sytm 0prace to St aCnrli SbShrnArc
Hemu Aler an Aru Rune-etge
107~~~~~~~~ 4neepodTe rp:Cmpnnso rdcieadMr





119 In inul Deelpe Pond Aqa r in Banglades
Wac Ali Shhn P hii TowI I





F.S 0edus F. K mj . .tinr M. Grfo, .Fdai l Mai and L . mp0le 155A aseof Patcpor ri ng in Gede Aais AnPt L Fro rad er .EquzadElsL a9








Journal
for Farming Systems Research -Extension


Volume 5, Number 1, 1995


Published by
the Association for Farming Systems Research-Extension








Journal for Farming Systems Research-Extension


Editor
Timothy R. Frankenberger Office of Arid Lands Studies The University of Arizona, Tucson

Associate Editors
Jennifer Manthei, M. Katherine McCaston, Karen Coelho and Claude P. Bart Office of Arid Lands Studies
The University of Arizona, Tucson

Production and Layout
Jennifer Manthei and Claude P. Bart Arid Lands Design, Office of Arid Lands Studies The University of Arizona, Tucson

Sponsors
Ford Foundation
The University of Arizona




The Journalfor Farming Systems Research-Extension is published by the Association for Farming Systems Research-Extension (AFSRE), an international society organized to promote the development and dissemination of methods and results of participatory on-farm systems research and extension. The objectives of such research are the development and adoption through participation by farm household members of improved and appropriate technologies and management strategies to meet the socioeconomic and nutritional needs of farm families; to foster the efficient and sustainable use of natural resources; and to contribute toward meeting global requirements for food, feed, and fiber.
The purpose of the Journal is to present multidisciplinary reports of on-farm research-extension work completed in the field, and discussions on methodology and other issues of interest to farming systems practitioners, administrators, and trainers. The Journal serves as a proceedings for the annual international Farming Systems Symposium from which selected and refereed papers are included. It also welcomes contributed articles from members of the AFSRE who were unable to attend the symposium. Contributed articles will be judged by the same review process as invited articles.


ISSN: 1051-6786









Journal for Farming Systems Research-Extension
Volume 5, Number 1, 1995


CONTENTS

1 New Directions of the Systems Approach for the Modernization of
Latin American Peasant Agriculture
Julio A. Berdegul and Germ4n Escobar

31 Theory & Practice in FSRE: Consideration of the Role of Modelling
Barry Dent

45 New Resources for International Agricultural Cooperation:
Village-Based Self-Help and Agricultural Research in Japan
Teruo Wada, John S. Caldwell, and Shigeki Yokoyama

79 Toward the Concerted Management of Agro-Ecosystems
A. Huijsman

91 Management of Natural Resources: Systems Approaches to
Striga Control in Sub-Saharan Africa
Helmut Albert and Artur Runge-Metzger

107 Underexploited Tree Crops:
Components of Productive and More Sustainable Farming Systems
Robert Moss

119 An Indigenously Developed Pond Aquaculture System in Bangladesh
Wajed Ali Shah and Philip Townsley

129 A Systems Approach for Better Understanding of Policy Impact:
The Vulnerability of Family Farms in Western Europe
Karlheinz Knickel

143 Agricultural Price Policy and Export and Food Production in
Cameroon: A Farming Systems Analysis of Pricing Policies
F. Heidhues, F. Kamajou, J. Chataigner, M. Griffon, A. Fadani,
Ali Madi, and L. Temple

155 The Role of Training in Women and Development:
A Case of Participatory Training in Gender Analysis
Anita L. Frio, Armando Jerry S. Erguiza, and Ellis L. Matheny









New Directions of the Systems Approach for

the Modernization of Latin American

Peasant Agriculture'

Julio A. Berdegue and German Escobar2



ABSTRACT
Latin American agriculture faces fundamentally new economic, cultural, political, and social scenarios. The conditions under which the systems approach took form in our Continent 25 years ago, are no longer valid nor are the objectives that agricultural scientists and rural
development workers are asked to meet.
This paper examines these changes, discusses the current vulnerability of the conventional application of the systems approach, looks at the new directions being explored by numerous research and development teams in different countries, and finally attempts to suggest some concepts and criteria that could be useful to adapt the systems approach to the new circumstances and new objectives of Latin
American peasant agriculture.


INTRODUCTION

Latin American agriculture faces fundamentally new economic, cultural, political, and social scenarios. The conditions under which the systems approach took form in our Continent 25 years ago, are no longer valid nor are the objectives that agricultural scientists and rural development workers are asked to meet.
This paper examines these changes, discusses the current vulnerability of the conventional application of the systems approach, looks at the new directions being explored by numerous research and development teams in different countries, and finally attempts to suggest some concepts and criteria that could be useful to adapt the systems approach to the new circumstances and new objectives of Latin American peasant agriculture.



1 Keynote paper presented at the Thirteenth Annual Farming Systems Research-Extension
Symposium, Montpellier, France, November 21-25, 1994.
2 RIMISP, Casilla 228 - Correo 22, Santiago, Chile


Vol. 5, No. 1, 1995






BERDEGUt AND ESCOBAR


A TURNING POINT: THE CRISIS OF THE 1980'S
The crisis of the decade of the 1980s can now be seen as a turning point for a large number of Latin American countries. That the crisis had the effect of putting an end to the old paradigms that had been in effect since the post-war period, was due, to a large extent, to its magnitude. The per capita Gross National Product (GNP) in 1989 for the region as a whole, was equal to that of 1976, with several of the major countries, such as Argentina, Peru or Venezuela, showing average rates of growth of the per capita GNP of 20% or greater for the 1981-1989 period (CEPAL, 1990).
The inflation rate reached the highest levels in history, climbing to an annual average of 1 023% in 1989 (CEPAL, 1989), an increase of 18 times when compared with the 1980 rate. The value of exports dropped in 11 of the main 19 countries, from a price index of 100 in 1980, to only 72.3 in 1989 (CEPAL, 1990). Over 70% of our exports are primary products, which show an annual rate of growth of only 1.6% in the 1962-1985 period, as compared with an annual increment of 8.1% for the technology intensive manufactured goods (CEPAL, 1990). In other words, by the end of the decade, it had become apparent that there was a fundamental discrepancy between the structure of international demand and the composition of Latin American exports (CEPAL, 1990).
The crisis hit hardest the poorest sectors ofsociety. In 1988,61% of the rural population (76 million people) lived below the poverty line (FIDA, 1993). Small-scale farmers and landless peasants are among the poorest social groups in countries such as Mexico, Brazil, Colombia, Ecuador, Peru and Guatemala (FIDA, 1993). For example, in Mexico the rural poor include I million smallscale farmers, 1.5 million minifundistas (peasant with extremely small plots, of
1.7 ha on the average) and 1.2 million landless peasants (FIDA, 1993).
According to de Janvry (1994a), the qualitative nature of rural poverty in Latin America has changed in the last decade. The smallholder sector is larger in absolute numbers as well as in terms of its share of total agricultural employment; non-agricultural income has become more important for agricultural households; households headed by women represent a larger fraction of the poor; and ethnic groups have been differentially affected by the crisis as a result of such factors as racial discrimination and language disadvantages.


A NEW PARADIGM
The economic issues of the 1980s and the pressure originated by the size of the external debt ofsome Latin American countries, marked the collapse of the prevalent import-substitution model. All the major countries and many of the smaller ones, implemented structural adjustment programs oriented, in the short run, to deal with the external debt issue and, immediately after that, to the reorientation of the economy under a fundamentally new paradigm.


Journal for Farming Systems Research-Extension






NEW DIRECTIONS


The major characteristics of the new economic model included the following:
The principle was accepted that economic decisions were to follow free market rules and conditions. The role of Government in society was redefined in the direction of transferring resources and attributions to the private sector. All economic factors and all production and consumption activities must be inserted in the market regardless of the economic sector or the individual firm.
The axis of economic development was shifted from the industrialization and import-substitution policy that had been in effect for several decades, to an export-oriented strategy. The national economies were rapidly opened to international markets, through the reduction of tariffs and custom duties, the dismantling of protection policies in most economic sectors, and the promotion of exports through different mechanisms. In order to stimulate a minimum level of competitiveness of the economy, liberal monetary policies and new exchange rates policies were implemented.
Most sectorial (e.g., agriculture) compensation and regulation instruments were removed or downsized in order to let market forces operate in the allocation of resources among the different sectors of the national economy. Most internal prices were let free to vary with the international market tendencies.
Regional and international trade agreements and economic integration mechanisms were adopted by most countries in the region, to increase their market size and as an answer to the economic blocks created by Europe, the Asian Pacific region, and North America.


THE NEW SCENARIO OF LATIN AMERICAN AGRICULTURE
There is a consensus that Latin American agriculture is already immersed in a new scenario. In general terms, this means that Latin American agriculture is being called upon by our societies to fulfill new objectives under new economic and political conditions.
Although the new scenario places important and difficult challenges in front of Latin America's agricultural sector, it is also true that it has opened new opportunities, as testified by the fact that in the 1980s and for the first time since the post-war period, agriculture grew at significantly faster rates than industry (2% and 0.5% per year, respectively, as an average for the period between 1980 and 1990; Kay, 1994).
In the past 30 or 40 years, the essential objective of Latin American agriculture, was to provide basic food staples at a low cost to sustain the growth of the urban-industrial sector, considered the main engine of our economies.
Today, the least that can be said is that this objective has been qualified: agricultural development must be based on its own the competitiveness in the international markets; it must be ecologically sustainable; and it must lead to greater social equity (Calder6n et al., 1992).


Vol. 5, No. 1, 1995






BERDEGUE AND ESCOBAR


There are subsectors of agriculture in each country that have long been oriented towards the international markets and that have enjoyed a competitive aptitude for many years and even decades (e.g., fresh vegetables in North East Mexico; coffee in the Caldas region of Colombia;. banana in the Ecuadorian region of Guayas; table grapes in the Central valley of Chile; small grains in the Argentinean Pampa; or citrus in Southern Brazil).
Also, in almost all countries one can find regions where important programs have been developed to manage and restore the quality of important watersheds, to promote soil conservation, to reduce deforestation, to regulate pesticide use or to insure water quality. Examples of these major efforts are the Plan Sierra in the Dominican Republic (de Janvry, 1994b) and the microwatershed management program for soil erosion control in the Brazilian state of Parani (Fialho et al., 1992).
The emphasis under the new scenario, however, is that all of agriculture must be competitive and sustainable. Thus, the new problem is to integrate what up to now have been dispersed and fractioned issues. This is a major shift in the formulation of the objectives of Latin American agriculture.
On the other hand, the economic and political conditions under which these new objectives must be achieved, have also changed profoundly. First and foremost, the new scenario means that the viability of Latin America's agricultural sector rests on its ability to compete with price and quality standards set by the international markets. This is obviously and immediately true in the for-export sector, but it will become an increasingly valid condition for that fraction of agriculture which is oriented towards the internal markets and that is being forced to turn out products at a similar cost and with comparable quality than those which could be bought in the open international market (see Box 1).
In general, prices of agricultural products have been adjusted to international levels as a direct consequence of an open economy and a revaluated currency that facilitate imports (e.g,. Chile has changed over 4000 agricultural custom positions from an average tariff of 95% to 11% of the CIF value).
This has occasioned at least two additional problems: the profitability of several agricultural products (mainly those considered as traditional) has decreased in the last five years, since international prices of these products have notoriously diminished in the same period. In Chile, the participation of salaries and revenues in the total value of agricultural products, dropped from 24.3% in 1980 to 18.7% in 1989, corresponding with the period in which Chile made the strongest changes in the economy (Larrariaga, 1991).
At the same time, exports of traditional agricultural products have decreased from 51% to 39% in the ALADI countries (L6pez Cordovez, 1993). Although Latin American agricultural exports grew twice as fast as internal market products (4% and 2%, respectively) in the period between 1964 and 1984 (Kay, 1994).


Journal for Farming Systems Research-Extension







NEW DIRECTIONS


A number of subsidies and direct transfer expenditures have been discontinued, due to the termination of compensation policies that favored the agricultural sector. Minimum support prices for agricultural products, favorable interest rates for agricultural credit, direct subsidies to particular inputs, land reform programs, free irrigation water through public investment, subsidized technical assistance and direct government market interventions are but a few of the subsidies that are being discontinued (e.g., tradable water rights for irrigation were created in Chile in 1981; direct subsidies to fertilizers and other agricultural inputs were eliminated in Venezuela in the early 1990s).
Some support services to agricultural production and marketing that used to be supplied by the government are being privatized or downsized. Activities that were usually run by Government such as production ofagricultural inputs; banking and financial intermediation; investment on production services like technology transfer, agricultural research and marketing centers, have been turned over to the private sector or have suffered enormous budget reduction when still managed by the government (e.g., the government agricultural marketing agency in Colombia, IDEMA, has reduced direct purchasing of products from 39 to 8, which have been designated as critical for small farmers, and which will continue to receive government support for only three



Box 1: The New Scenario and Milk Production in Costa Rica
Estrada (1992) has documented the effect of micro- and macroeconomic variables on the technological and production decisions of small-scale dairy farmers in the regions of Rio Frio and Sonafluca, in Costa Rica's low humid tropics. It had been observed that the decisions of farmers were at variance with the recommendations that originated in the research of CATIE (Tropical Agronomy Research and Education Center), with respect to such variables as size of the dairy operation, type of pastures, use of fertilizers, and production levels.
The studies concluded that the interaction between biological and economic variables could explain the farmers' behavior. For example, CATIE recommended the application of 250 kg of nitrogen/ha-year, but this technology was not adopted at all in Rio Frio and only to a very limited extent in Sonafluca. It was found that in low-fertility soils, infested with a low quality pasture (Ischaemun indicum), the response to nitrogen fertilizers was lower than expected. On the other hand, the price ratio between nitrogen and milk in 1979-84, when the farmers started their dairy operation, was the best in 20 years (1 kg N2:2 kg milk). However, the oil crisis forced farmers to pay 40% more in terms of milk to obtain the same amount of nitrogen fertilizer. This in turn stimulated farmers to turn to nonrecommended, native pastures, that had low nitrogen requirements.
On the other hand, high international meat prices in the 1970's convinced the Government of Costa Rica to implement support programs for cattle-based systems and in particular, to stimulate dual-purpose (milk-meat) operations. Large amounts of public funds were channelled to implement subsidized credit programs and road-building and rural electrification projects, which provided a favorable environment for dairy-improvement projects such as that .evaluated by Estrada. For example, US$ 3,595 were made available between 1970 and 1989, as subsidized credit for cattle-related investments.
The outlook changed substantially in the 1980's, due to shifting consumer preferences in the developed countries and because of the economic crisis that forced Costa Rica to drastically curtail all kinds of price-support and other subsidized programs. After 15 years of work, the end result is that the net farm income per hectare has dropped 68% in Rio Frio and 84% in Sonafluca.


Vol. 5, No. 1, 1995






BERDEGUt AND ESCOBAR


additional years. Mexico has initiated the process to regularize a free land market by privatizing the public land assigned to small farmers in the form of ejidos).
The integration of national economies into free-trade blocks is causing and will continue to force major shifts in the agrarian structure of the participating countries. For example, Figueroa and Muchnik (1993) foresee that the Free Trade Agreement that is being discussed between the United States and Chile would cause a drop of employment in Chile in the sector of basic agricultural products (cereals, dairy products, sugar and oils) of 8,000 jobs per year. However, 2,000 jobs would be created annually in the sector of fresh forexport products, and an additional 8 500 jobs per year in the agroindustrial export sector. Small-scale farming in Chile is related basically to those products that face a more serious challenge.
This single fact ought to be enough to recommend a major reorientation of systems-oriented research and development projects in Chile, in particular if one considers the geographic distribution of the future economic dynamics: The job-losing areas are in the Central-Southern and Southern provinces, where peasant farms are most abundant, while employment net gains are concentrated in the Central-North regions, dominated by modern, capitalist farms.
As a result of the North American Free Trade Agreement with the USA and Canada, Mexican agriculture will have to undergo major adjustments within the next 15 years. Essentially, at the end of this period, all production, transport and marketing costs, in all agricultural and agroindustrial products, will have to be covered directly by producers in a context of zero barriers to trade with the US and Canada. A detailed study of nine major crops (L6pez and M~ndez, 1994), found that 57.6% of their harvested area could not expect to compete under the new conditions. Since these crops account for 81% of the national harvested area, 47% of Mexican agriculture (in terms of acreage) will face a serious and probably terminal challenge as a consequence of the new scenario.
Corn and field beans (the two major crops of small-scale farmers in Mexico), are considered largely non-competitive (63% and 62% of the corn and bean acreage, respectively). If new on-the-shelf technology is applied, about 305,000 ha could be reconverted to the other seven crops included in the study, out of a total of 5.7 million ha of corn and beans which are considered non-competitive. It is estimated that about 947,000 jobs can be lost in corn and 170,000 in beans, which are equivalent to 77% and 67% of the total of each crop, respectively. Finally, it should be noticed that an additional 3.5 million ha (31.4% of the total for the nine crops, or 25% of the agricultural area) are only marginally competitive, so that a 10% decline in prices would render them non-competitive.
As can be seen from the examples of Chile and Mexico, the changes in the economic and political context of the agricultural sector, are producing


Journal for Farming Systems Research-Extension






NEW DIRECTIONS


differential effects across social groups and across regions. These effects often disfavor small-scale farmers. Peasant farms, linked as they usually are to the traditional agricultural subsectors, have faced a deterioration of their production structure and of their capacity to compete in a relatively free market economy. This undesired condition is aggravated by the reduction of government expenditure on physical and production infrastructure in areas of smallscale producers and by the downsizing of investment in rural development programs.
The agricultural context in the case of small-scale farmers is marked by the fact that the crisis aggravated the bimodal characteristic of the agrarian structures of most Latin American countries, in which there is a segment of capitalists agricultural firms, more or less modernized, and an important and differentiated contingent of peasant production-consumption units. Schejtman (1994) has argued that the bimodal characteristic of our agrarian structures is a major impediment for the diffusion of technological progress. Technological patterns in an homogeneous agrarian structure tend to be valid for the majority or a large fraction of the production units. However, when the agrarian structure is highly heterogeneous, options which are valid for the larger, modern, capitalist farms, are usually not appropriate for the large contingent of peasant family farms. It follows that this bimodal characteristic of Latin American agrarian structures is a major impediment for the advancement of social equity, for the transformation of production patterns, and for the objective of increasing the systemic competitiveness of the agricultural sector as a whole.
But again, the new scenario, while challenging, is also a source of new opportunities. To begin with, despite the major crisis of the 1980s and the unfavorable political environment of the 1970s with the prevalence of rightwing dictatorships, peasant agriculture continues to be a major economic reality in Latin America. In the 1980s, it controlled over 33% of the arable land and over 40% of the harvested area (Kay, 1994). Peasant farms employed two thirds of the labor force, and supplied 40% of the internal market products and 33% of the export goods (Kay, 1994). In Guatemala, close to 150,000 peasant households are involved in non-traditional for-export agriculture; one million small-scale farmers (less than 5 ha) form the backbone of the Colombian export coffee industry; in Mexico, property of state-owned agroindustries is being transferred to innovative associations of small farmers (see Box 2).
In a recent article, de Janvry (1994a) discusses several new phenomena which ought to facilitate the task of achieving more equitable growth in the Latin American rural areas: (a) Real exchange rate depreciation and trade liberalization not only reduce the bias against agriculture in general, but also tend to favor small-scale farms which are more intensive in the use of nontradable inputs; (b) Public expenditure in agriculture increased slightly in the last years of the last decade, as compared with a strong decline in the previous period; (c) Improved technologies are being developed for peasant


Vol. 5, No. 1, 1995






BERDEGUE AND ESCOBAR


00 Box 2: Pineapple Agroindustry in Loma Bonita, Mexico
According to Castillo Cuazcle (undated), in the 1940's pineapple was introduced to the region of Loma Bonita, in the Papaloapan watershed, in the state of Oaxaca. In the last 15 years, the pineapple production of Loma Bonita has represented about 20% of the national total, with about 1,000 ha. Due to accumulated technical, organizational, and economic problems, only a small fraction of the potential area was planted with pineapple.
In 1989, the Mexican government decided to transfer property of the agroindustry COFRINSA to the pineapple producers. Three Farmers' organizations joined to form a new firm: AgroindustriasLoma Bonita, which received a US$0.5 million loan to start operations. The main product was juice concentrate for the national market. However, a 5,000-ton contract was negotiated with the one of the Mexican juice-producing firms (Jumex), which allowed the factory to operate at close to full capacity and was a critical factor in offering a more stable price to producers throughout the season.
The new agroindustry is now attempting to organize and rationalize production by contracting specific acreages with the different farming communities. Also, plans are underway to improve specific technical problems at the farm, factory, and marketing levels. Finally, the agroindustry is studying ways to form a Credit Union in order to be able to intervene in the financing ofproducers.
farming systems and new partnerships are being developed between government agencies and non-governmental organizations (Bebbington et al., 1993); (d) Democracy has returned to almost all countries in the region; (e) NGOs and Grass Roots Organizations (GROs) have proliferated in direct relation to the downsizing of the public apparatus; and (f) There are rising concerns for environmental protection.
In summary, beginning in the 1980s a whole new scenario has developed for Latin American agriculture in general and for small-scale agriculture in particular. Systems-oriented research and development in Latin America, with its strong emphasis on small-scale farmers, must make fundamental adjustments to stay in tune with this new scenario, or else it will become irrelevant. The starting point is to be able to evaluate our achievements and, above all, our shortcomings.


SHORTCOMINGS OF THE CONVENTIONAL APPLICATION OF THE SYSTEMS APPROACH
In the early 1970s, organizations for international development (notably the Rockefeller Foundation with the pioneer Puebla Project in Mexico; US AID in Guatemala with ICTA; or Canada's International Development Research Center with the Caqueza project in Colombia) had the largest influence in the dissemination of the systems approach in Latin America. Since the early 1970s, a number of major systems-oriented projects, involving tens of millions of dollars, were implemented in the region.
While the money lasted, many of these Latin American projects shared the label of "Farming Systems Research and Extension" (FSRE, or IESA in Spanish). However, aside from the name, what is important is that many projects did have a number of characteristics in common.


Journal for Farming Systems Research-Extension






NEW DIRECTIONS


THE CONVENTIONAL FORMULATION OF THE SYSTEMS APPROACH
Since the issue at stake was that small-scale, resource-poor farmers had not benefited from the Green Revolution, the projects implemented under this conventional formulation had the objective of adapting and disseminating appropriate technologies that could be adopted by these peasant farmers (Byrnes, 1993; Escobar and Berdegu6, 1990).
To obtain this type of product, research and extension activities were conducted at the farm system level, since this was recognized as the key conceptual and operational unit (or "system of reference").
Interactions between the farm subsystems, and between the farm and the socioeconomic and biophysical environments, were presumed to condition the farmer's decisions concerning technology and production, and thus were to be taken into consideration when designing new alternatives.
Usually, the method was organized as a set of five stages: characterization or diagnosis, design, on-farm research, validation, and dissemination.
Given the multidimensional nature of the farm and its environment, interdisciplinary work was considered to be an essential characteristic of the method. With time, new concepts were added to respond to new issues, such as that of farmer participation, gender, or ecological sustainability.
There is no doubt that this conventional formulation had important positive impacts in Latin America. Perhaps one of the most significant contributions is that it started the idea that agricultural researchers had to get out of their experiment stations and confront the actual conditions under which their technologies were to function, by working closer and more systematically with farmers, under real field conditions.
Also, small-scale farmers were recognized as important customers of agricultural research and extension. Working at the farm level, it was rapidly understood that technology was not neutral and that social, economic, and cultural variables played a decisive role in starting or frustrating effective adoption processes.
Finally, these farming systems projects were instrumental in the organization and strengthening of several national and regional agricultural research and rural development institutions, as in the case of Guatemala, Colombia, or Mexico.
In short, the first generation of Latin American systems-oriented projects, made an important contribution in redefining agricultural research and extension in the region and in opening new opportunities for small-scale farmers to be integrated into these activities.
However, with some exceptions, the balance indicates that the projects developed under this conventional formulation of the systems approach, fell short of achieving their objectives (Berdegu6, 1993a; Byrnes, 1990, 1993; Hibon, 1993). This was a fact recognized by most practitioners, and as Hibon


Vol. 5, No. 1, 1995






BERDEGUt AND ESCOBAR


(1993) recently wrote: "A growing disconcert is noticeable among the practitioners of Farming Systems Research and Extension."
There are several factors that help explain this limited result in terms of actual impact at the farmer level. Since the specific operational problems have been well documented (Tripp, 1991; Byrnes, 1993; Berdegu6, 1993b), forthe purposes of this paper we will emphasize four issues which in our opinion are of a more fundamental nature.


FOCUS ON INCREASING SUPPLY
A large number of the research and development projects that responded to the conventional formulation of the systems approach, were designed from the almost exclusive point ofview of increasing the output of the farm system. The relevant problem usually was defined as some variation of the idea of ". increasing adoption rates by adapting, validating and disseminating improved appropriate technologies ."
It was normally forgotten that the farmer's objectives are almost always defined in terms of their participation in one or more markets, i.e., are defined and achieved outside the farm gate. Technology adoption is not a farmer's objective per se, but, in the last analysis, only a dependent variable, a "condition for."
This is true even in the case of subsistence agriculture, in which the farmer's decisions are greatly influenced by such factors as regional and seasonal opportunities for off-farm employment, availability of transportation, possibility to trade goods with farmers from the same or other villages, nonagricultural income, and many other variables related to different markets for goods and services (Berdegu6 et al., 1990; Espinosa et al., 1990; Martinez et al., 1990; Miranda, 1990).
This problem was aggravated in those cases in which it was proclaimed a priori ".the option of rescuing traditional, low productivity technologies, in a peculiar interpretation of what is adequate ." (Schejtman, 1994, p. 150). There is no substantial difference between this approach and that of the "neutral technology" school, since both negate the farmer's objectives and place the technological artifact or process as the central issue (Berdegu6 and Nazif, 1988).
The well-known argument that farming systems projects should concentrate on promoting small, cumulative and gradual changes in the existing system structures is probably at the base of this neglect for the demand variables (Baker, 1993; Berdegu6, 1993b).
While farming systems practitioners went around building projects around this dogma, Guatemalan Indian farmers increased the acreage of non-traditional agricultural exports by almost 100% in only 14 years, in a strong linkage with over 300 new agroindustries (see Box 3).


Journal for Farming Systems Research-Extension







NEW DIRECTIONS 11

Box 3: Non Traditional Export Crops in Guatemala's Highlands

The Guatemalan highland is a region traditionally dominated by small- and medium-size crop and beans markets (CEPAL, 1994b). However, between 1979 and 1993, nontraditional export grew 90.9% (an annual average of 6.5%, which more thandoubles the rate for the agricultural sector as a whole), to reach a total area of almost 70,000 ha. Macadamia nuts, mango, okra, broccoli, minivegetables, and strawberries are among the emerging crops planted mainly by approximately 140,000 small-scale farmers in plots of only 0.5 ha on the average. Yields have consistently improved in the last 14 years, with increases that usually range between 80% and 112%. A recent study concludes that
Even through this new generation of farmers have maintained their minifundista production system, their capacity to innovate made them modernize their production marketing processes. the small and medium size of the production units was no impediment to their rapidly achieving high levels of competitiveness in the in the international markets. Their ability consisted in specializing in a given crop to obtain high productivity and qualitylevels. (CEPAL, 1994b,p. 26).
The case of the "Cooperativa Cuatro Pinos" in the state of Sacatep~quez (altitude 2,000 m) is considered a paradigmatic example of how small-scale farmers can achieve high levels of international competitiveness. The Cooperative was formed in 1979 by 1,600 Indian minifundistas to help in the reconstruction of their village after a major earthquakes. In 1981, with support from Swiss agencies and from the Guatemalan government, they started a small dehydrating plant to improve the marketing of their traditional summer vegetables. By 1983, they were exporting fresh vegetables, having built extensive cold storage facilities. The Cooperative today runs complementary operations, including their own research, technical assistance, credit and rural infrastructure programs.

Neglecting the issue of determining the structure of market demand for agricultural products is always a problem, but it becomes a cardinal sin in the context of open market economies because in the new scenario a key to success is the ability to identify new market opportunities.


MISSING MACRO-MICRO RELATIONS

A large number of projects simply did not pay attention to the basic systems concept that the functions and dynamics of the farm system are conditioned to a large extent by the perceptions of the household members about the context in which their productive activities take place (Berdegu6, 1992).
"Macro" analyses were frequently limited to a compilation of secondary data, and very few cases are documented when projects tried to deal with the important issue of the interaction of "micro," "macro," and sectorial variables (Estrada, 1992).
This is true in the field of economics, when the most frequent role of the specialist was to prepare partial budget analyses or other forms of documentation of the effect of specific technologies on gross and net margins.
However, the situation is even more dismal in relation to other types of nonfarm variables. For example, there are few efforts to systematically research the role that can be played by intermediate agrarian institutions in the articulation of the farm, the farmer, and the household, with their environments in general


Vol. 5, No. 1, 1995






BERDEGUt AND ESCOBAR


and in the dissemination of technical progress in particular. Also, the social and cultural elements that determine the gender dynamics of the farm system, were rarely understood even when they had a direct and evident relationship with the technological issues being addressed by the projects (Campaha, 1992; Flora, 1994; Feldstein et al., 1991; Feldstein et al., 1989).
Off-farm variables and the issues of policy and support systems, were defined as external parameters (Baker, 1993; Berdegu6, 1992). Of course, neither assumption holds true, since these phenomena are highly variable (and increasingly so in the new scenario) and are also fundamental criteria in the farmers' decision-making processes (Ashby, 1989; Benito, 1976; Gladwin, 1980, 1976; Low et al., 1991).
For example, Gladwin (1980) found that land-poor, subsistence agriculture farmers in the highlands (Altiplano) of Guatemala, consider market conditions, labor availability, and access to capital, as key criteria in deciding which of eight crop and crop mixtures to plant.
It has been common for researchers and rural development workers to blame peasant risk-aversion for the shortcomings of their projects in the case of land-poor households. However, it has been found that many times these projects are based on the promotion of new technologies that are cashintensive, in a context of highly variable output prices and poorly-developed marketing infrastructure (Reinhardt, 1987) and that often the new technologies cannot claim to yield increased net returns (de Janvry, 1983). Development strategies that are unreasonable from an economic point ofview, actually deserve to fail.
Setting the off-farm variables as external parameters, allowed practitioners to think the expected or desired outputs of their projects in terms of optimal trajectories, which were themselves held constant at least for the duration of a given project.
The very nature of the innovation process in a context first of a major crisis (the 1980s) and then of global markets and reduced state intervention, increasingly questioned the appropriateness of this major assumption. Farmers reacted to the new contexts by developing livelihood strategies that usually confounded the efforts of the technical staff of many of these projects. The changes of direction were very pronounced in the case of small-scale farmers, since often they meant significant shifts in the combination of farm and offfarm activities and in the very structure of the farm systems (Estrada, 1992).
The false assumption that off-farm and cultural variables are external parameters, has many times cornered systems-oriented projects in an impossible contradiction between the true "battle field," which is the development and transfer of improved technologies for the crop and animal subsystems, and the off-farm environment, perceived basically as a source of problems, a source of constraints, and rarely as a source of opportunities.
New or emerging societal concerns, such as that of sustainability, added complexity to this picture, for in general they meant the obligation to


Journal for Farming Systems Research-Extension






NEW DIRECTIONS


incorporate and act upon new dimensions, such as those of time and cultural variables, that were at the root of the new issues. It is not surprising that many projects reacted with a combination of generic conceptual acceptance of the new questions and effective incompetence to deal with them in practical terms. In Latin America we are still far from having designed efficient and practical ways to operationalize the concept ofsustainability or even to agree upon basic indicators to deal with it (Harrington et al., 1994).
The sustainability issue implies the operationalization of the time variable at least over one generation, and this is an impossible task when the usual analytical and operational horizon of the conventional systems-oriented projects usually do not extend beyond three to six years.
Incorporating the dimension ofsustainability also means that projects need to be able to deal with trade offs in farmers' objectives. There are trade offs over time (e.g., short term versus long term family income), over space (e.g., upstream versus downstream benefits of a watershed management project), and across objectives at any one time and space (e.g., reducing soil erosion and family income in the short run through less intensive crop rotations). By overconcentrating on the farm as the reference system, by defining expected products in terms of technology adoption, and by reducing the importance of "macro" and sectorial variables, the conventional formulation of the systems approach places additional conceptual and operational constraints to the integration of the new attribute of sustainability.


NARROWING THE OPERATIONAL FIELD TO THE FARM
In many of the conventional systems-oriented research and development projects, the correct definition of the farm as the primary scenario of peasant farmer development, was fatally distorted when it was interpreted as meaning that the project's action could be limited to the farm. We forgot that practicing agriculture means more than growing crops or raising cattle.
In the five-step methodological protocol (characterization, design, onfarm testing, validation, dissemination) that most projects usually followed, the actual, real problem was defined in terms of taking a component X of the system, and modifying or replacing it with a component Z. This protocol corresponds to a strategy designed to "modify technologies rather than modifying farmer circumstances" (Baker, 1993). We were happy thinking that our method was superior because we were able to recognize constraints to this process, even if we actually did little about them.
We can now see that the important issue is not to understand that, for example, lack of credit will impair the adoption of a given technology; this knowledge is of little use if we do not act to surmount this bottleneck. The lesson to be learned is that constraints are there to be removed, and not simply to be analyzed.


Vol. 5, No. 1, 1995






BERDEGUt AND ESCOBAR


Even when projects are successful in adapting new appropriate technologies and in stimulating their adoption, many technical and institutional obstacles usually still remain to be solved before the increased output can get to the consumers at the appropriate time, with the required quality and at a competitive price.
Considering that FSRE teams are in a unique position to provide farm-level knowledge about how agricultural support systems work, in particular in the case of the input and output markets, it is a great puzzle why many farming systems projects restrained themselves from acting beyond the farm gate (Baker, 1993).
In the new scenario, small-scale farmers need to be able to innovate all along the chain that leads from their field to the buyer of their product, and moreover, many times they will need to adjust their production, their postharvest processes, their local organizations, their marketing channels, their input supply systems, and their access to numerous other services, in order to reach those markets that are more dynamic and more profitable. In other words, new farming systems projects need to be based on a mid- and long-term strategy for developing farmer competitiveness.
Comparative advantages turn into competitive advantages when organizations, information and management systems, and top quality technical expertise, are put in place on a permanent and self-reproducing basis.
The issue of developing farmers' organizations deserves to be addressed in particular. Successful and stable technical change in small-scale agriculture is always a collective enterprise, involving local communities and microregions rather than isolated, individual farms.
Farming system projects rarely understood that together with achieving a specific technological or production objective, it is necessary to put in place mechanisms to strengthen the permanent capacities of the local communities to innovate and to administer and expand the results of the innovation. The issue of local and microregional community institutions of which functional farmers' organizations are the most important, was not a fundamental preoccupation of the conventional farming systems projects in Latin America. To a large extent, this explains why the positive results of many projects dissipated after the external sources of funding and technical support were terminated.
The process of political and administrative descentralization and ofstrengthening of municipal governments that is taking place in many countries, can help in opening wider opportunities for organized participation of the local communities in the decision-making processes of agricultural research and rural development. Colombia is perhaps the leading country in this area, having designed decision-making and funding mechanisms that give a greater voice to the municipal governments and the local communities, in the definition of priorities and approval of specific research and development programs and projects. As local governments grow stronger, it will become


Journal for Farming Systems Research-Extension






NEW DIRECTIONS


more difficult for the agricultural research and rural developmet institutions to design and implement centrally-controlled programs and projects.


THE INABILITY TO BUILD STRATEGIC PARTNERSHIPS
Many of the limitations of the conventional systems-oriented research and development projects can be traced to the fact that most were implemented by a single institution or by one or two institutions with an analogous mandate (e.g., an international and a national agricultural research agency).
Baker (1993) has argued convincingly that in the early FSRE analytical models, agricultural policies and institutions were prominently included, and that "the most important reason" why the primary function of the methodological approach was reduced to technology design and testing, was that "the early FSRE methodology development took place mainly in the International Agricultural Research system."
Few cases are documented in which complementary institutions, public and private, established active partnerships to provide the required expertise to adequately cover the relevant links in the chain between the farm fields and the marketplace. Interdisciplinary dialogue--in itself a rare and evasive occurrence-lacked an executive, action-oriented dimension, in which farmers and production experts got together with people that could actually plan and work on the financial, marketing, infrastructure, or organizational issues relevant to the farmers.
In the new scenario, the development of the farmers' ability to compete in the national and international markets, will require to develop strategic partnerships between complementary organizations, always including at the very center, those of the farmers themselves.
In summary, the conventional formulation of the systems approach in Latin American agriculture, has implicitly led to an unnecessary reductionist position, in contradiction with all the central tenets of the systems concept: Reductionist because it overemphasizes the farm and its subsystems; reductionist because of its overemphasis on technology adoption; and reductionist because it was quite incapable of dealing operationally with macro-microeconomic relations and with social and cultural variables and with new societal concerns.
The end result was that the farm was isolated from its environments in general and from the markets in particular. This was the cardinal failure of the conventional application of the systems approach in Latin America between the early 1970s and the late 1980s.
This issue of the reductionist nature of the conventional formulation of the systems approach, needs to be emphasized because addressing it must become the number one conceptualand methodologicalpriority in the new scenario with its dynamic conditions, its new objectives and its new concerns.


Vol. 5, No. 1, 1995






BERDEGUt AND ESCOBAR


THE SYSTEMS CONCEPT AND THE FARMING SYSTEMS METHODOLOGY
It has been argued (Berdegu6, 1993a, 1993b) that the conceptual framework of the systems approach has comparative advantages in the new scenario. The notion that the interactions and dynamics among the components of a system are not expressed additively in defining the end result of a process, and the idea that systems are hierarchically related "upwards" and "downwards" so that agricultural processes have global and local manifestations, are two important concepts of the systems approach that are particularly pertinent under the new conditions of Latin American agriculture.
Additionally, the practical work of the past two decades has trained hundreds of Latin American farming systems practitioners, so that there is now a significant regional expertise in the use of methodological approaches which are very suitable to the new scenario; in particular, the ability to work in the field with small-scale farmers, is an invaluable capitol.
Therefore, in our opinion the systems conceptual framework and many of the conventional methodological developments of the first generation of systems-oriented projects, will remain fully valid in the years to come.
In fact, numerous governmental institutions that are undergoing modernization processes, have explicitly adopted the systems approach as their basic conceptual framework for dealing with the current problems of Latin American agriculture. Such is the case, for example, of INTA (National Institute for Agricultural Technology) in Argentina, EMBRAPA (Brazilian Organization for Agricultural Research), INIAP in Ecuador (National Institute for Agricultural Research), CORPOICA (Colombian Corporation for Agricultural Research), or INDAP in Chile (Agricultural Development Institute). In the 1990s, the use of the systems approach has responded to a national option rather than to the influence of multilateral or bilateral donor agencies.
However, it is important to recognize that our conventional formulation cannot be projected linearly into the future, because it showed significant weaknesses under the old scenario and also because there are fundamental discrepancies with the demands of the new conditions and the new objectives of Latin American small-scale agriculture.
The present challenge is to develop new strategies, new methods, new combinations of the already proven instruments, and first and foremost, to foster a new outlook and a new mentality of those who are in charge of directing and applying the systems concept in specific agricultural research and development institutions, programs, and projects.
The task is to retool the systems approach, to gain a new efficiency and a new efficacy in the research and development of competitive, sustainable, and equitable agricultural systems.


Journal for Farming Systems Research-Extension






NEW DIRECTIONS 17

RETOOLING THE SYSTEMS APPROACH
The new scenario imposes new constraints to peasant development, but it is also offers new opportunities. Already a large number of systems-oriented projects in Latin America have been designed precisely to take advantage of the opportunities offered by the new scenario. The examples inserted in the text boxes of this document, are only an illustrative sampling of a widespread reality. In dozens of research and development experiences in Latin America, a new generation of systems-oriented projects is gradually starting to take shape, which is addressing many of the shortcomings of the conventional methodological formulation of the systems approach. By analyzing what these projects are doing, it is possible to highlight some common characteristics of this second generation ofsystems-orien ted research and development projects.


DEVELOPING A NEW MENTALITY
All successful projects of the second generation are engined by a new mental outlook on the part of their central actors (see Box 4). There is no hope to move ahead if we remain anchored in the old paradigms. A famous British economist once said that it is relatively easy to find new ideas, but that the difficult problem is to get rid of the old ones.
BOX
To act aggressively in the search of new sources of competitiveness; to think and operate in terms of broad systems such as agroindustrial chains and regions; to focus on the quality of processes and not only on products; to integrate dynamic and often contradictory attributes and objectives of development, resisting the temptation to move them out of the picture because of the inherent complexity of this task; to foster functional and efficient farmers' organizations and other such permanent capacities of the local communities; and to build strategic alliances and partnerships (see Boxes 5 and 6), are six key

Box 4: Natural Foods of Irupana, Bolivia
Natural Foods of Irupana was formed in 1987 by a group of professionals linked to NGOs who were interested in experiencing with a new type of microenterprise, capable of working according to market rule (Hurtado, 1994).
Initially, the firm tried to compete in the regular coffee market, but found that
. this was the first lesson: Microenterprises cannot and should not compete in the mass products market, on the contrary, they must take advantage of the quality of their labor to make high quality products, with greater added value and higher price. Paradoxically, our market is in the rich social classes." (Hurtado, 1994, p. 149).
Following this rule, Irupana started making sugar-free toasted coffee, introducing after thatnew types of natural foods, produced from over 25 kinds of agricultural products bought from smallscale farmers. For example, Irupana buys coffee from peasant farmers with a scheme that is different from that used by the traditional intermediaries in that it pays a quality bonus of up to 15% and provides technical assistance. Thanks to this kind of measures, Irupana now has a network of peasant suppliers that process their coffee with special attention to quality.

Vol. 5, No. 1, 1995






BERDEGUE AND ESCOBAR


concepts that define the new mental outlook that to a greater or lesser extent characterize the numerous research and development projects that in Latin America are already exploring new ways to deal with the conditions imposed by the new scenario.


IDENTIFYING MARKET OPPORTUNITIES AND ACHIEVING COMPETITIVENESS
As de Janvry (1994a, p. 90) has very well pointed out, "identifying market opportunities and achieving competitiveness has become the hallmark of rural development."
Even where transitional periods have been established to allow for the adjustment of the more disadvantaged sectors of the rural economy (e.g., Fifteen years for some basic staples in Mexico under the new Free Trade Agreement with the US and Canada), it is a hard fact that under the new rules of the game, small-scale farmers will eventually need to compete under more or less the same rules that apply to all producers.
New generation systems-oriented projects have in common that they have been able to construct a clear and precise idea of the market demands that they are trying to fulfill. The new methodological approaches must incorporate top-quality market studies to be able to define in the most precise terms the

Box 5: Rice Processing and Marketing in Santa Cruz, Bolivia
"La Campana," a rice processing and marketing firm, was formed in Santa Cruz, Bolivia, in 1991 (Matzuzaki, 1994). It is defined as a mixed corporation for risk-sharing, involving a large, commercial farmer (50%), an NGO (36%), and an organization of 243 small-scale farmers (14%). There is a private contract between the NGO and the farmers' organization that establishes that eventually the latter will gain ownership ofan additional 26% of the society, for a total participation of 40%.
La Campana provides different services to the rice farmers: storage, grading, drying, processing, and marketing, all with the direct participation and control of the peasants.

Box 6: The Vaquerias Contract of Association in Participation, Mexico
One of the best known Mexican Asociacionesen Participacion involves 400 farmers and the large aroindustrial complex GAMESA (Suarez, undated). GAMESA contributed US$6 million to finance large investments in irrigation works, machinery and technology. The Mexican government, through a debt swap scheme, placed another US$6 million to finance the farmers' capital contribution to this contract. A specific corporation, DICAMEX, was formed to manage the project started under this contract.
At the beginning of the project, the farmers contracted 2,500 ha of beans (first crop season) and 3,500 ha wheat (second crop season). The contract was signed for 12 years or 24 cropping cycles and of this period the farmers will have an option to buy back the equipment at 25% of its original value.
Profits are divided in equal parts and farmers can obtain cash advancements during the season. If no profits are obtained in one season, the farmers are still guaranteed 8% of the gross value of their crop.


Journal for Farming Systems Research-Extension






NEW DIRECTIONS


products and varieties that are demanded by buyers and what are the specifications of such products in terms of quality, packaging, presentation, labeling, volume, and opportunity.
For example, the yuca project being implemented in the Guayas province of Ecuador by such organizations as CIAT and FUNDAGRO, has had to test a number of varieties to find those that are best adapted to different uses in a dynamic market that fluctuates between fresh consumption and industrial uses in the extensive shrimp farms that characterize that region. At the same time, they have had to innovate permanently to improve the postharvest processes in order to obtain a top-quality product for each market.


ARTICULATING THE GOALS OF COMPETITIVENESS,
SUSTAINABILITY, AND SOCIAL EQUITY
Many agricultural research and rural development projects have become trapped in a false dispute between the search for economic, environmental, or social objectives (competitiveness, sustainability, and social equity, respectively).
Second-generation systems-oriented projects are showing in a practical manner that these three objectives not only are not contradictory, but that in effect reinforce each other and are mutually necessary for the sustainability of each of them in the long run.
The Economic Commission for Latin America (CEPAL, 1990) has indicated that long-term competitiveness can only be achieved if the increasing participation in the markets is achieved simultaneously with an improvement in the levels of wellbeing of the population. In other words, in the long run economic competitiveness cannot be based on such factors as low wages or declining investment on educational, housing, or health systems. This is so because, increasingly, economic development is based on such factors as information-intensive production processes, permanent technological innovation, and high savings rates to finance investment, which are not compatible with an impoverished and marginalized population. As CEPAL (1990, p. 71) has stated: "The dividing line between the successful and unsuccessful experiences of international insertion, apparently is linked with the efficient use of resources at a given time and with the capacity to undertake activities which require a growing intellectual added value."
On the other hand, it is increasingly improbable that competitiveness can be achieved and maintained in the long run if it is based on production systems that are not environmentally sustainable. First of all, because "environmental quality" is being appreciated by consumers as a desired attribute of agricultural products, not only in terms of the absence of contaminants in the product itself, but also in relation to the effect that the production process had on the quality of the physical and biological resources; the debate in Brazil about


Vol. 5, No. 1, 1995






BERDEGUt AND ESCOBAR


sugarcane production, or the worldwide concern about livestock systems in deforested tropical rain forests, or the standards set by the European Union to the packaging materials used by Chilean fruit exporters, are examples of this factor.
Second, competitiveness and sustainability are increasingly interlinked attributes because inter-sectorial environmental conflicts are being recognized as important sources of inefficiency. The conflict in the coastal region of Ecuador between banana growers and shrimp producers over the use of pesticides that apparently affect the environment in which shrimp larvae develop, is an example of this type of issues.
Third, numerous cases have been documented that show that environmental degradation has become the main cause of decreasing productivity and/or increasing production costs (de Camino and Miller, 1993). For example, Ramirez and Martinez (1994) used multiobjective mathematical programming to analyze the trade off between the farmer's objectives of increasing gross margins and decreasing the erosion rates in small farms in the Andean piedmont in Chile. They found that in the very short run, more erosive systems yielded higher incomes, but that after only two to six years (depending on the erosion rate), the farmer's income was higher in the less erosive alternatives.
However, it is easier to recognize conceptually the positive and mutually reinforcing linkage between competitiveness, sustainability, and social equity, than to operationalize this approach to the problem. Several institutions are exploring research and development projects designed and implemented at the scale of micro-regions, to attempt to bridge this gap, because it is at this level that it is possible to operate simultaneously on variables that have an influence on all three objectives (INDAP, 1994; Namdar-Irani and Quezada, 1994). In particular, the discussions about the appropriateness of agricultural development (economic emphasis) versus rural development (social emphasis) projects, and between the incorporation or not of "viable" versus "inviable" peasant families, lack meaning if placed in the context of projects that aim at developing the agricultural systems at the level of micro-regions rather than farms.


DEVELOPING BROAD SYSTEMIC COMPETITIVENESS
Most of the ongoing Latin American second generation systems-oriented projects, design research, and development plans to cover all the relevant links in the chain that is required to produce the product and take it to the intended buyer at the appropriate time, in the required volumes, with the demanded quality specifications, and at a competitive price.
Thus, for example, Chile's PROCAMPO S.A. is implementing programs to provide its small-scale vegetable suppliers with a range of different top-quality technologies, from seedlings to transplanters to portable drip irrigation systems to new packaging materials to low-cost cold storage. The issue at hand,


Journal for Farming Systems Research-Extension






NEW DIRECTIONS


according to the General Manager (P. Rioseco, personal communication, August 1994) of this peasant-owned firm, is that no single variable can be left out if their products are to maintain the quality that allows them to hold a privileged position in a demanding and sophisticated market.
As de Janvry (1994a) has pointed out, contract agriculture, linking smallscale farmers with specific agroindustries is a promising channel to have access to international markets which are highly competitive and demanding in product specification and quality.
It should be noticed that by the early 1990s, Latin American agroindustrial exports had reached the annual sum of US$ 27 billion, equivalent to 22% of the total regional exports (CEPAL, 1994a). In addition, it is interesting to note that the most important group of agroindustrial exports is that of high added-value goods, which account for about one third of the total value (CEPAL, 1994a).
Schejtman (1994) states that agroindustry presents three characteristics which make it particularly valuable for the purpose of strengthening smallscale, family-based agriculture: (a) Greater flexibility than other industries in the scales or magnitudes of the required fixed-capital investments, which allows for a greater range of options of size adjustments to "fit" with the size and resources of the local units with which it must become articulated; (b) It allows for the integration of capital-intensive and labor-intensive processes, especially in the agricultural activities that supply the raw materials or primary inputs for the agroindustry; (c) Agroindustries impose production calendars, volumes of production and quality specifications, and, through them, they can play an integrative and organizing role of those units which are its direct or indirect suppliers.
Production systems which integrate agroindustries with small-scale agriculture, have a greater competitive advantage in those areas in which the primary processes are very labor intensive per unit of land, and in which mechanization is not an efficient option (Schejtman, 1994). This is because the opportunity cost of the small farmer tends to be lower than that of the capitalist sector.
According to Schejtman (1994), the association of agroindustry and smallscale agriculture has a greater competitive advantage when the agricultural product is also highly perishable so that it cannot be easily obtained in the open market; and when the raw input has a high cost per unit of volume so that transaction and transportation costs are of less relative importance. In Schejtman's (1994) typology, the modern agroexport agroindustries (e.g., vegetable and fruit products, flowers), followed by the modern agroindustries of basic agricultural products (e.g., milk products, sugar from sugarbeets, animal feed), show the greater advantage from the point of view of their potential to generate technical progress in small-scale, family-based agriculture (see Box 7).
However, it is important to underline that the search for export-oriented and agroindustrial opportunities for small-scale farmers, does not preclude the


Vol. 5, No. 1, 1995






BERDEGUt AND ESCOBAR


Box 7: Asparagus Production by Small-Scale Farmers in Peru
In terms of value, asparagus in 1990 was the most important vegetable crop of Peru and, by far, the most important nontraditional agricultural export (CEPAL, 1993). In that year, 17,930 ha were planted with this crop, up from 1,512 ha only ten years before. In 1970, there were only three agroindustries that processed asparagus, but the number had increased to 14 by 1993. IQF and fresh exports have diversified the traditional markets of canned product.
In one of the most dynamic asparagus regions, Piura 47% of the crop acreage is under the control of small-scale farmers, with less than 3 ha each. The majority of the minifundista peasants used to plant between 0.5 and 3 ha ofbasic crops, such as cotton, rice, corn and grain legumes, facing severe problems of shortage of irrigation, obsolete technology and lack of access to technical assistance and credit. For these farmers, the new crop meant new development alternatives based on the substitution or complementation of their traditional crops. Despite the fact that asparagus is a foreign crop in their farm systems, it represents a safer market and more stable income. Also, asparagus production implies access to financing, training and technical assistance.
In 1986, a group of businessmen linked to the mining industry decided to explore new investment options in the agroindustrial sector in Piura. Two corporations were formed to set up the project, which started with an adaptive experimentation project in 2 ha, with the cooperation of CIPA, an agency of the Peruvian Ministry of Agriculture. A scheme was setup in which the Agrarian Bank of Peru (BAP) would make long-term loans to start new asparagus plantations to all those persons that had previously signed productions contracts with the new agroindustry. Given the social orientation of BAP, it promoted the new crop mainly among land-poor farmers. Technical assistance was obtained from an important Dutch firm. Soon, two other agroindustries started operations in Piura to buy asparagus from local small-scale producers.

competitive production of staple foods for the regional and national markets as a convenient platform for effective development (de Janvry, 1994a). As a matter of fact, this may be the best option to open access to economically feasible alternatives to peasant communities that lack the natural, capital, and infrastructure resources required to produce for the international markets or the agroindustries (see Box 8). '
Another dimension of this concept of broad competitive systems refers to the geographic territories in which it is possible to identify peasant farmer communities that can collaborate in common efforts to develop new market opportunities. Region-based projects can tap the social dynamics that exist between these communities and, at the same time, provide different options so that specific social groups (youth, women, landless peasants, local merchants) can participate in the new on-farm and off-farm activities that originate in a new economic initiative. Region-based projects, at the same time, provide an opportunity to integrate agricultural and rural development, as in the case of the successful "Cooperativa Cuatro Pinos" in Guatemala that not only has established a number of production-related services, but that also funds such projects as rural electrification.
Alain de Janvry (1994a) has argued that projects need to be designed to accommodate different and changing types of demands, due to the heterogeneity of peasant households and their income strategies; in particular, it is important the new generation projects avoid the tendency to narrowly define its area of interest to the issue of agricultural development, neglecting the potentials offered by the development of linkages with industries and services, including microenterprises.


Journal for Farming Systems Research-Extension








NEW DIRECTIONS


Box 8: Marketing of Field Beans in Chile
Several regional organizations of small-scale farmers, together with a private trading firm and the technical and financial support of the Agricultural Development Institute (INDAP), have implemented an innovative marketing system to improve the competitiveness of peasant farmers in the national and international grain legume markets (Tello and Gonzalez, 1993; Tello and Rojas, 1994). This mechanism has operated in two consecutive seasons (1992-93 and 1993-94) and has been approved for a third (1994-95).
In the first season, four organizations representing 1,880 small-scale farmers, in three of Chile's 13 regions, signed an agreement with a private trading firm and INDAP to improve the marketing of field beans and garbanzo beans. The agreement included guidelines for the following processes:
(1) Preparation of storage, facilities, and personnel, indicating the minimum facilities that each organization had to have in place in order to participate in this system. (2) Common quality and sanitary standards and product specifications and a common sack and logo.
(3) Buying criteria and procedures in which the trading firm faxed daily national and international price reports to each organization and each of them calculated the daily buying price by subtracting their costs and operational margins. In this way, the price farmers received was finely tuned to market tendencies. (4) Selling criteria and procedures; each day each organization informed the trading firm about the volume of bulk and processed beans in storage. Since the quality was homogeneous across regions and organizations, the trader could add all the individual volumes into a larger package. After that, the trader would contact the main six or seven bean exporters and have them bid for the product in storage. The offers could vary significantly among exporters, as could the form of payment and transportation conditions. The trader would choose the best offer and fax the information to the organizations, each of which would ship their lot to the agreed port and prepare a separate invoice to obtain their payment directly.
An Executive Committee (organizations trader and INDAP) would meet every 15 days to review the process and results. INDAP provided loans to the organizations (US$543,000 in the first season and US$833,000 in the second) to finance a revolving fund to buy the legumes from farmers and pay for their operational costs. The export firms found the system very convergent since it guaranteed that the quality and volumes ordered would be exactly those received, a condition which not always held true when they dealt with the traditional or informal market intermediaries.
During the second season (1993-94), four new organizations joined this program and the system was extended to a total of seven regions; the seven organizations represented 12,290 farmers, an increase of 690% with respect to the first season. A total of 34 local organizations participated with 32 buying points;with this extensive network, the system had an effect on average prices on a national scale, by introducing competition and minimum-prices in almost all the major bean producing regions. Also, the mechanism was expanded to include two new crops: lentils and chickpeas.
In the first season, the prices that farmers received were 18% and 40% higher than those paid by the traditional intermediaries for field beans and garbanzo, respectively. In the second year, the differentials were smaller since most intermediaries were forced to match the minimum price offered by the organizations, but there was a 23% jump in field bean prices the week that the program started operating during the second season. In addition to the price differentials obtained by farmers, the organizations made a total net profit of US$79,979 in the second season.
Finally, farmers were also able to obtain payment for the full 18% value-added tax (IVA), which they could then discount from their IVA accumulated when buying their agricultural inputs. This was possible since INDAP negotiated with the Internal Revenue Service two new decrees (N' 6 111 and 6 510) that ordered that buyers (i.e., the participating farmers' organizations as well as the traditional intermediaries) instead of sellers (i.e., farmers) retain the IVA tax and pay it to the Treasury. The two new decrees were a major blow to traditional intermediaries who after that could not avoid paying their taxes and adding the 18% legal value to the payments they made to the farmers. Thus in this system, all players won, with the exception of the traditional intermediaries.


Vol. 5, No. 1, 1995






BERDEGUE AND ESCOBAR


LEARNING TO BUILD STRATEGIC ALLIANCES WITH NEW PARTNERS
Most new generation systems-oriented research and development projects have learned that no single organization is capable of providing the required expertise to deal with all the relevant problems and dimensions involved in the task of achieving a competitive position in their markets. The self-sufficient project, or the self-sufficient organization, is a thing of the past.
It should be pointed out that the diminished ability of the public sector to act as direct provider of such services as credit or marketing infrastructure, implies that increasingly the private sector will need to be involved in farming systems research and development projects.
In Mexico there is vast experience in the development of this kind of partnerships. The Mexican Ministry of Agriculture and Hydraulic Resources (Suirez, undated) has registered 142 "associations in participation" (asociaciones en participaci6n, which are contracts between two or more parties that do not generate a permanent legal entity). These types of contracts have been used to start joint ventures between farmers and different types of businesses in agriculture, livestock, horticulture, and forestry, involving about 15 000 farmers and 40 000 to 50 000 hectares.
Technological and management innovation systems, marketing ofproducts and access to inputs and services, long term financing for investment, and strengthening of farmers organizations, are four key components that in one way or another, have been solved by all projects which have been successful in developing the long-term competitiveness of peasant agricultural systems (Berdegu6, 1994).
It is not conceivable that a single agricultural research or rural development organization, can develop the required expertise to understand and act effectively in these four basic instrumental components. Moreover, the public sector in most Latin American countries is clearly pulling out specifically of these areas that imply a direct economic intervention. The consequence is that agricultural research and rural development organizations must learn to build inter-institutional strategic partnerships, but also that public-private cooperation is a sine qua non condition for success.
The adjustment will be more difficult in the case of agricultural research institutions than in that of rural development organizations, because the former have had a historic difficulty in linking the issue of technological innovation with that of agricultural development. Moreover, there is a strong tendency in many research circles to see this line of thought as an unwanted and harmful intrusion in the essence of scientific undertakings. The problem will not be solved ifit is not understood that the question is not that ofscientists leaving research in order to undertake other tasks for which they are not prepared and trained, but one of orienting research in response to the demands of clients and of learning to work in close contact with other agencies and


Journal for Farming Systems Research-Extension






NEW DIRECTIONS


organizations that can provide the goods and services that are required for effective technological innovation.


FLEXIBLE PROJECTS FOR PERMANENT INNOVATION
It is necessary to revise our mostly linear methodological schemes in which a fixed and almost invariable objective is defined at the start of a project, to be achieved by a series of predesigned steps or stages.
The very nature of the innovation process in the new scenario, implies that it has become nearly impossible to establish optimal trajectories for projects. Hence, the new generation of systems-oriented projects tend to be demandled, or,to use de Janvry's (1994a) terminology, "project-making projects." Here, innovation processes, mechanisms for rapid adaptation to changing conditions; agile access to services; and strong, effective, functional organizations of producers; in the long run are as important products ofa project as the technological product itself.
To achieve this, second generation systems-oriented projects must develop an understanding of the relationships between macro and micro variables and must establish capabilities to research long term market and technological tendencies, to capture new development opportunities that arise from them and to react to unforeseen circumstances that may affect the results of the project.
Economic, social and technological developments in the new scenario of Latin American agriculture are much more dynamic than in the past. Institutions and projects must be equipped to work in this context of rapid changes.


DEVELOPING PERMANENT CAPACITIES IN THE LOCAL COMMUNITIES
Successful peasant development projects always involve organized groups of farmers, because individual and isolated small-scale farm systems cannot expect to become competitive in the new scenario.
Functional farmer organizations are required for peasant producers to achieve economies of scale, to have access to goods and services under adequate conditions, to negotiate with buyers of their products and improve their marketing systems, to implement high quality management services that are indispensable in this age of great competition, and, in the end, to be able to capture a greater share of the benefits of their productive activity.
A new "organizational technology" must be developed for this purpose, because the old schemes that were useful when the issue was that of representing peasant demands vis-a-vis the state, have been overcome by the new demand for organizations that are efficient and effective in providing competitive goods and services to their members.


Vol. 5, No. 1, 1995






BERDEGUt AND ESCOBAR


The development and strengthening of local organizations has often been seen as a distant task by most public agricultural research institutions. Nongovernmental organizations have developed an important experience in this respect. Fortunately, in many countries it is now a frequent occurrence to see innovative schemes in which governmental and non-governmental organizations work together in the design and implementation of agricultural research and rural development projects.
Several types of linkage mechanisms between NGOs and public agricultural research and rural development agencies have been documented by Bebbington et al. (1993). For example, NGOs in Chile are contracted by the public extension system to provide this service to close to 40% of all the attended peasant families. In Colombia, joint farming systems research programs are carried out between NGOs, the national agricultural research institute and an international agricultural research organization of the CGIAR system. In Ecuador, NGOs are implementing adaptive research programs to fill in the gap between the research conducted by the public sector institutions in the experiment stations and the small farmers' conditions. In Bolivia, local agricultural experimentation centers were developed by NGOs and later, agreements were established with the governmental research institutions to manage and operate them jointly.
This tendency for greater public-NGO cooperation is not without problems and limitations. However, the important fact is that common objectives have been identified in many cases and that successful schemes have been developed to design and implement joint and/or complementary research and extension programs. Since almost all NGOs base their work in some form of collaboration with local or regional farmers organizations, these schemes have in effect served the purpose of bringing public agricultural research institutions closer to dealing with organized groups of peasants.
There are specific issues concerning functional farmers organizations that still need to be better understood and that deserve greater attention from systems-oriented projects. It is a complex transition between more or less informal local groups of farmers and well structured farmers organizations capable of operating as efficient providers of goods and services. However, unless we can advance in this direction, small-scale farmer development will always be dependent on external funding, management, and technical support, and we know well that that is a sure prescription for frequent failures and frustrations.


REFERENCES
Ashby J.A., 1989. Colombia. Production and consumption aspects of technology testing
in Pescador. In: Working together. Gender analysis in agriculture. Volume 1: Case studies, H.S. Feldstein and S.V. Poats, ed., West Hartford, Connecticut, Kumarian
Press, p. 109-133.


Journal for Farming Systems Research-Extension






NEW DIRECTIONS


Baker D., 1993. Inability of farming systems research to deal with agricultural policy.
Journal for Farming Systems Research-Extension 4(1): 67-86.
Bebbington A., Thiele G., Davies P., PragerM. and Riveros H., 1993. Non-Governmental
Organizations and the State in Latin America, London, Routledge, 290 p.
Benito C.A., 1976. Peasant's response to modernization projects in minifundia economies. American Journal of Agricultural Economics 58(2): 143-151.
Berdegu6 J.A., 1994. Propuestas para la modernizaci6n de la pequefia agriculture.
Documento 16. Santiago, Chile, INDAP.
Berdegu6 J.A., 1993a. Desafios para la investigaci6n y extensi6n en sistemas agropecuarios: El aporte del Simposio IESA de Quito. Revista Investigaci6n/Desarrollo para
Amirica Latina 2: 47-53.
Berdegu6 J.A., 1993b. Challenges of farming systems research and extension. Journal of
Farming Systems Research-Extension 4(1): 1-9.
Berdegu6 J.A., 1992. Objetivos de la etapa de diseho en la metodologia de sistemas de
producci6n. V International Meeting of the International Network on Farming
Systems Research Methodology, Santiago, RIMISP.
Berdegu6 J.A., Sotomayor 0. and Zilleruelo C., 1990. Metodologia de tipificaci6n y
clasificaci6n de sistemas de produci6n campesinos de la provincia de Ruble, Chile. In: Tipificaci6n de Sistemas de Producci6n Agricola, G. Escobar and J.A. Berdegu6 ed.,
Santiago, RIMISP, p. 85-117.
Berdegu6 J.A. and Nazif I., ed., 1988. Sistemas de Producci6n Campesinos, Santiago, GIA. Byrnes K.J., 1993. Del sistema agropecuario al "campo de suerlos:" Los impedimentos
nficleos, operacionales y gen6ricos a la implementaci6n de la IESA en Amrica Latina.
First Latin American Symposium on Research and Extension of Agricultural Systems,
FUNDAGRO, Quito, Ecuador.
Byrnes K.J., 1990. A review ofA.I.D. experience with farming systems research and extension
projects, A.I.D. Evaluation Special Study No. 67. Washignton, D.C. Agency for
International Development.
Calder6n F., Chiriboga M. and Piieiro D., 1992. Modernizaci6n Democrdtica e Incluyente
de la Agricultura en Amirica Latina y el Caribe, Serie Documentos de Programa NO
28, San Jos6, Costa Rica, IICA, 96 p.
Campaia P., 1992. El contenido deginero en la investigaci6n de sistemas de producci6n,
RIMISP, Santiago. 56 p.
Castillo Cuazcle, F. Undated. La industria pifiera en la regi6n de Loma Bonita, Oaxaca.
In: Blanca Suarez. Documento base para la elaboraci6n del estudio sobre la transformaci6n productiva de la pequehia agricultura, Unpublished manuscript.
CEPAL [Comisi6n Econ6mica para Am6rica Latina], 1994a. Amirica Latina: Cuantificaci6n de nuevas categorias agroindustriales, Santiago, CEPAL, 24 p.
CEPAL [Comisi6n Econ6mica para Am6rica Latina], 1994b. Las relaciones contractual
y la transformaci6n productiva de la agricultura en Guatemala, Santiago, CEPAL, 70
p.
CEPAL [Comisi6n Econ6mica para Amdrica Latina], 1993. Analisisde cadenasagroindustriales en Ecuador y Perii, Santiago, CEPAL, 294 p.
CEPAL [Comisi6n Econ6mica para Am6rica Latina], 1990. Transformaci6n productiva
con equidad, Santiago, CEPAL, 185 p.
CEPAL [Comisi6n Econ6mica para America -Latina], 1989. Balance preliminar de la
economia de Ameica Latina y el Caribe, Santiago, CEPAL.


Vol. 5, No. 1, 1995






BERDEGUt AND ESCOBAR


de Camino R. and Mfiller S., 1993. Sostenibilidaddelaagriculturaylosrecursosnaturals.
Basespara establecer indicadores. Serie Documentos de Programa 38. San Jos6, Costa
Rica, IICA-GTZ, 133 p.
de Janvry A., 1994. Social and economic reforms: The challenge of equitable growth in
Latin American agriculture. In: Apertura econ6mica, modernizaci6nysostenibilidad de la agricultura. IV Congreso Latinoamericano y del Caribe de Economia Agricola, E.
Muchnik and A. Niho de Zepeda, ed. Santiago, ALACEA, p. 79-98.
de Janvry A., Sadoulet E. and Santos B., 1994. Project evaluation for sustainable rural
development: Plan Sierra in the Dominican Republic. VI International Meeting of
RIMISP, Santiago, RIMISP, 22 p.
de Janvy A., 1983. The Agrarian Question and Reformism in Latin America. Baltimore,
The John Hopkins University Press, 311 p.
Escobar G. and Berdegu6 J.A., 1990. Conceptos y metodologfa para la tipificaci6n de
sistemas de finca. In: Tipificaci6n de Sistemas de Producci6n Agricola, G. Escobar and
J.A. Berdegud, ed., Santiago, RIMISP, p. 13-43.
Espinosa P., Jitiva P. and Suirez G., 1990. Caracterizaci6n de los sistemas de producci6n
agricola de producores de mafz de la provincia de Bolivar en Ecuador. In: Tipificaci6n de Sistemas de Producci6n Agricola, G. Escobar and J.A. Berdegu6, ed., Santiago,
RIMISP, p. 157-166.
Estrada R.D., 1992. Anilisis de informaci6n secundaria en el disefio de sistemas de
producci6n, V International Meeting of the International Network on Farming Systems Research Methodology (RIMISP), Mexico, July 1992, Santiago, RIMISP,
(Manuscript).
Feldstein H.S., Flora C. and Poats S.V., 1991. La variable deginero en la investigaci6n
agricola, Ottawa, Canada, Centro Internacional de Investigaciones para el Desarrollo,
56 p.
Feldstein H.S., Poats S.V., Cloud K.C and Husinga Norem R, 1989. Conceptual
framework for gender analysis in farming systems research and extension. In: Working together. Gender analysis in agriculture. Volume 1: Case studies, H.S. Feldstein and S.V.
Poats, ed., West Hartford, Connecticut, Kumarian Press, p. 7-26.
Fialho J.T., Feitosa dos Santos L.M. and Trento E.J., 1992. Extensao Rural no Parand.
Um Modelo Ambiental, Curitiba, Brazil, EMATER-Parani.
FIDA [Fundaci6n Internacional para el Desarrollo Agricola], 1993. El Estado de la
Pobreza Rural en el Mundo. La Situaci6n en Amdrica Latina y el Caribe, Rome, FIDA,
99 p.
Figueroa E. and Muchnik E., 1993. Efectos sobre tl empleo de un ALC [Acuerdo de Libre
Comercio] Chil-EE. UU.: Elcaso de lossectoresagropecuarioy agroindustrial chilenos, IV
Latin America and Caribbean Congress of Agricultural Economists, April 1993, Vifia
del Mar, Chile, ALACEA, 15 p.
Flora C., 1994. Enlace generaci6n-transferencia de tecnologia. In: Transferencia de
tecnologia agropecuaria: De la generaci6n de recomendaciones a la adopci6n. Enfoques y casos, A. Monardes, G. Escobar and G. Gonzilez, ed., Santiago, Chile, RIMISP, p.
13-36.
Gladwin C.H., 1980. A theory of real-life choice: Applications to agricultural decisions.
In: Agricultural decision making: Anthropological contributions to rural development,
P. Barlett, ed. New York, Academic Press, p. 45-85.
Gladwin C.H. 1976. A view of the Plan Puebla: An application of hierarchical decision
models. American Journal of Agricultural Economics 58(5):881-887.


Journal for Farming Systems Research-Extension






NEW DIRECTIONS


Harrington L., Jones P. and Winograd M., 1994. Operacionalizaci6n del Concepto de
Sostenibilidad: Un Mitodo Basado en la Productividad Total, VI International Meeting
of RIMISP, Jaguarifina, Sao Paulo, Brazil, RIMISP, 27 p.
Hibon A., 1993. La investigaci6n ensistemasagropecuariosen la decada de losnovente: restos
para un enfoque econ6micamente viable, First Latin American Symposium on Research and Extension of Agricultural Systems, Quito, Ecuador, FUNDAGRO, March 1993,
(Manuscript).
Hurtado J., 1994. Alimentos Naturales de Irupana, una excepci6n que confirma la regla.
In: Agroindustria ypequefia agricultura, A. Schejtman and C. F. Toranzo, ed., FAO
and ILDIS, La Paz, Bolivia, p. 145-167.
INDAP [Instituto de Desarrollo Agropecuario], 1994. Programa degesti6n estratigicay
de modernizaci6n institucional. Santiago, Chile, INDAP, 13 p.
Kay C., 1994. Rural development and agrarian issues in contemporary Latin America,
Working Paper Series N_ 173, The Hague, Institute of Social Studies, 37 p.
Larraiaga 0., 1991. Ajuste macroecon6mico, agricultura y pobreza rural: Chile en los 80.
Series Investigaci6n, Programa de Post-grado de Economia ILADES-Georgetown
University. Santiago, IIADES,. 33 p.
L6pez Cordovez L., 1993. La apertura comercial y el desarrollo de producciones agricolas
de exportaci6n no traditional. Informe de Consultoria a la Unidad de Desarrollo
Agricola de la CEPAL, CEPAL, Santiago, 124 p.
L6pez J. and Mfndez J., 1994. El papel dela agricultura en una political integral de empleo,
CEPNA, Mexico, 78 p.
Low A.R., Waddington S.R., and Shumba E.M., 1991. On-farm Research in Southern
Africa: The Prospects for Achieving Greater Impact. In: Planned Change in Farming Systems: Progress in On-Farm Research, R. Tripp, ed., West Sussex, England, John Wiley
and Sons, p. 257-272.
Martinez E., Ortiz A. and Reyes L., 1990. Caracterizaci6n de los sistemas de producci6n
minifundistas de la parte alta de la cuenca del rio Achiguate, Guatemala. In: Tipificaci6n de Sistemas de Producci6n Agricola, G. Escobar and J.A. Berdegu6, ed., Santiago,
RIMISP, p. 221-231.
Matzuzaki P., 1994. Participaci6n campesina en la Sociedad La Campana S.R.L. In:
Agroindustria y Pequefia Agricultura, A. Schejtman and C. F. Toranzo, ed., La Paz,
Bolivia, FAO and ILDIS, p. 201-217.
Miranda E.E., 1990. Tipificaci6nde lospequefios agricultores: Ejemplo dela metodologia
aplicada a los productores de frijol de Itarare, Sao Paulo, Brasil. In: Tipificaci6n de Sistemasde Produccibn Agricola, G. Escobar and J.A. Berdegu6 ed., Santiago, RIMISP,
p. 119-138.
Namdar-Irani M. and Quezada X, 1994. Propuestas locales de desarrollo silvoagropecuario
para pequegos productores. Una experiencia metodol6gica. Tomo I. Santiago, Chile,
GIA-INDAP, 115 p.
Ramirez E. and Martinez H., 1994. Evaluaci6n de impacto ambiental en sistemas de
producci6n campesinos. Santiago, Chile, RIMISP, 45 p.
Reinhardt N., 1987. Modernizing peasant agriculture: Lessons from El Palmar, Colombia.
World Development 15(2):221-247.
Schejtman, A., 1994. Agroindustria y transformaci6n productiva de la pequefia agricultura. Revista de la CEPAL 53:147-157.
Sufrez B. Undated. Documento base para la elaboraci6n delestudiosobre la transformation
productiva de la pequena agriculture, Unpublished manuscript.


Vol. 5, No. 1, 1995






BERDEGUt AND ESCOBAR


Tello C. and Gonzilez M., 1993. Reporte final del programa de comercializaci6n de
leguminosas de grano y maiz, temporada 1992-93, Santiago, Chile, INDAP, 21 p.
Tello C. and Rojas C., 1994. R eportefinaldelprograma decomercializaci6n deleguminosas
degranoy mai; temporada 1993-94, Santiago, Chile, INDAP, 34 p.
Tripp R., ed., 1991. Planned change infarmingsystems: Progress in on-farm research. West
Sussex, England, John Wiley and Sons.


Journal for Farming Systems Research-Extension









Theory & Practice in FSRE:

Consideration of the Role of Modelling'

Barry Dent2



ABSTRACT
This paper has three major sections: the first attempts to analyze the process of farming systems research-extension (FSRE); the second represents an enquiry into the nature and need for models in FSRE; the third is a philosophical discussion about concepts and paradigms required for further modelling as an essential component for more
efficiency in FSRE.
In the first and second parts, emphasis is placed on FSRE as a holistic concept that incorporates some modelling component as a vital part of the process. The nature ofsuch models is discussed. The second part of the paper progresses from the point of view that any discipline or branch of study that does not develop its basic paradigms will fail to progress and will be replaced by more robust approaches. Research into FSRE itself has been limited and consideration is given to the role modelling could play in research activity into the ideas and concepts
of FS IE.


THE FSRE ACTIVITY SEQUENCE

Figure 1 attempts to capture the basic steps in the process of farming systems research-extension (FSRE). The most noticeable activities within FSRE are focused on the applied/adaptive research step. Simplistically, this may be viewed as the development of appropriate technology followed by demonstration of this in farmers fields. The research carried out is guided by the perceived needs of farmers in the district or defined agro-ecological zone. So, zone definition, description of the farm system(s), and establishment of databases must be the first step in the whole process. From this first step to the prioritization of research represents an enormous logical leap. In Figure 1, this leap is shown as assisted by some form of modelling. This is universally the case, even if for the most part the modelling is implicit and non-formal.

1 Paper presented at the Thirteenth Annual Farming Systems Research-Extension Symposium,
Montpellier, France, November 21-25, 1994.
2 Institute of Ecology & Resource Management, The University of Edinburgh, School of
Agriculture, West Mains Road, Edinburgh, EH9 3JG.


Vol. 5, No. 1, 199.5





DENT


Figure 1

FSR/E Sequence of actions


Research
-prioritisation


Journal for Farming Systems Research-Extension


Applied and
Adaptive research






THE ROLE OF MODELLING


Adoption of technology that will permit improved farm systems in terms of level and reliability of output is usually seen as the overriding aim of FSRE. The technology is generated in the series of vertical steps in Figure I together with the indicated feedbacks. An important feedback in this respect is implied within the single block of the figure linking applied research with the adaptive research procedure. Technology development is portrayed as reliant on field experimental work.
This dependence on field trials to develop 'technology packages' has got to be a matter of concern. Particularly in locations characterized by climatic variability, the cost/effectiveness of such work must be questioned. Observed and significant differences between treatments may be relevant only for the year and specific conditions of the trial. Other circumstances may negate or reverse treatment priorities. Repetition over years only expands the cost without necessarily improving the information provided and results in unacceptable delays. Other areas of biological science have developed adjacent methodologies (Jones & Kiniry, 1986) in response to this situation. Computer models of crop and livestock enterprises are widely available and well tested and should by now be replacing some routine field experimentation (Dent & Edwards-Jones, 1991). Here is a first, simple, and relatively well proven opportunity to apply modelling methods with potential gain to conventional FSRE.
The process of FSRE is then essentially driven by technology. Farmers are included in the research process because there is little doubt that, otherwise, factors can be overlooked which can hamper the subsequent adoption of successfully tested recommendations.
Hence, the concept has grown of FSRE being concerned to improve the efficiency of farming systems within the socio-economic constraints on development. This concept must surely be questioned; socio-economic factors should not be viewed as constraints but specifically as part of the system itself. The soil type may be a constraint, as may the climate, but the social organization of the farm family within its community is as much part of the system as the organization of crop production. Because FSRE recognizes the farmer as the main client of research, it must address high priority problems for farmers; these, however, are not always related to existing and currently operated technology or even to under-performing farm systems. (Byerlee & Tripp, 1988; Farrington, 1988). Hence, in rural areas, the farm as such, and its associated technology, should not be the sole (or even the main) focus for development. This is a crucial element because it begins to generate doubts about the kind of research as well as the research resources committed to FSRE in the name of development.
An alternative definition for FSRE and one that specifically emphasizes people and the social framework in which they live and work, as distinct from technology, is offered by Brossier & Chia (1994). They see FSRE as oriented toward farmers, of course, but argue "that researchers are not external to the


Vol. 5, No. 1, 199.5






DENT


systems under study." This Action-Research defines farmers and local operators (researchers, extensionists, local business) as part of the system. This is a view that generally appears to be favored by exponents of the soft systems approach (Checkland 1993; McAdam, et al., 1994); it specifically incorporates an analysis of processes and structures needed to deliver enhanced institutional frameworks and improved social, economic, and technical processes.
This paradigm is quite different from technology oriented FSRE. It is not necessarily (or mainly) on-farm experimental (trial) based and it is designed to release discovered "bottle-necks" that may be related to institutional incapacity, poor information flow, structural deficiency (e.g., inappropriately defined property rights), cultural restrictions, or technical limitations. Research/ investigation may then be prioritized as social, cultural, administrative, or technical problems within rural communities. From this realist perspective, FSRE is best seen as being activated when technology becomes an obvious restriction on the farming sub-system to the extent that it limits growth of the welfare of people in the defined community system.


MODEL TYPES IN FSRE
A crucial step that appears not to be formally addressed from the FSRE activity sequence is how (by what mechanism) perceived problems are identified, how they are prioritized, and what exactly is the 'research' process used to solve problems.
Figure 1 suggests that these processes are, in practice, achieved by a modelling activity. It is necessary now to clarify this by examining the model types that have been and may potentially be involved.

Implicit models
In FSRE practice such models are the norm and may perhaps better be described as 'mental' models. We all recognize such models that develop over time to become a mental framework that provides the structure for thinking about and defining a system. In an informal way the sub-systems, the driving variables, the outputs, and conditions (the 'environment') are assembled. Often such models are refined by discussions with colleagues and also with farmers and their families. Although informal, they can still be powerful constructs.
The problem with such models is that neither their structure nor the embedded interactions are stated or quantified. Neither can they be formally tested (validated). Furthermore, it is difficult to provide for sensitivity analysis, which is crucial for prioritization of research/investigation activity. Additionally, the system boundary definition is not specifically stated and this can cause fuzzy thinking in determining priorities of research. Most usually, the model relates to a perspective of a farm typical of the defined zone and the boundary is associated with the farming system.
Journal for Farming Systems Research-Extension






THE ROLE OF MODELLING 0

Clearly, FSRE practitioners use their experience against their mental model to gauge priorities for the research phase. Their judgements, however, are less easy to assess (and to share with others) than when the model is explicit and more formal.

Computer Models
Many reported examples have exploited mathematical programming concepts to create suitable models at all levels of application. In Action Research, for example, linear programming models have been used to explore the response of a representative farm to potential new technology or management.
This process has been used to assist in formulating research needs by discussing output from the model with the actors involved. Farmers are, of course, part of this dialogue, but in addition the outputs from such models have been used in discussions between farmers and their wider community (e.g., market organizations). In reported examples (Brossier, Vissac, and Le Moigne, 1990) this has been seen as a useful device to ensure that farm technology, infrastructure, and socio-cultural elements are considered in harmony. Some authors have specifically outlined the mathematical modelling methodology which provides for the links between farming systems and the broader communal economy (see Dent and McGregor, 1993; Doppler, 1994).
Those models that have been developed have tended to include structural variables such as farm size and enterprise mix, whereas the social elements of the systems (usually households) have been assumed to be uniform and to act as rational financial maximizers (Wossink et al., 1992). Although possible within such models, it is not usual to illustrate the impact of trade-offs between the many and different objectives within farm households. Other mathematical programming structures (such as goal programming) have this capability and examples of their value have been provided by Romero & Rehman (1989). Such models are conceptually quite simple but are still normative in nature. Thus an objective may, for example, be stated that the typical farm household for a defined zone would not wish to take more than X units of seasonal credit into their farming system and then only ifoverall 'profit' was likely to be greater than Y units.
But the issues relating to credit use are much more complex than this and few farmers are interested solely or even mainly on achieving high levels of 'profit.' As a result, attempts to create models of behavior of the typical farm family are now beginning to emerge (e.g., Edwards-Jones et al, 1994). In such models, and staying with the same example, the level of credit uptake has been linked to the age of the head of the household, whether this is a male or a female, and the level of education achieved by this person. Availability of credit may be determined by the equity status of the household and the past record of credit use.


Vol. 5, No. 1, 1995






DENT


Behavioral models that may associate decision making with key characteristics of the farm household and of the community in which the members live and work, as well as to the prevailing infrastructural and institutional conditions, are in the early stages of development. Mostly, they incorporate experiential information, and examples to date are established within a rulesbased framework.
The range of different models outlined in this section have different characteristics and it is important to select the model type to the circumstances of the FSRE program and the information available. Mathematical programming models largely require quantitative data input; behavioral models, on the other hand, are able to accommodate qualitative assessments and 'rules of thumb' as well as quantitative relationships. Behavioral model approaches may be said to be more closely akin to mental models but expressed within a more formal structure.

Soft Systems Models
Models developed under a soft-systems approach have characteristics that are similar to those described under Action-Research. Their main function is to create debate and discussion about an agreed problem area in development, by providing indicators that describe possible outcomes of alternative actions. Such models can be developed at a farming systems level but are much more likely at rural livelihood systems level. The soft systems methodology may be summarized as follows.
Soft Systems Steps:
1. Determine the problem situation.
2. Analyze the situation development of the 'rich picture.'
3. Create the conceptual model (ensure validity).
4. Generate feasible and desirable changes options for change by way of
model.
5. Actors discuss resource and infrastructure limitations.
6. Implement changes.
The process is not as focused as FSRE because field-type research is not a target. Work is centered on debate by key actors associated with the system under study and the conceptual model is drawn- up by all actors as a mechanism for helping the debate. The conceptual model may involve quantitative elements but discussions are often based on qualitative assessments.
The model involved in this process often takes the form of a description of the system-a picture description shared mainly by the actors involved. Conflicting objectives among groups of actors is likely to be endemic, whether at farm household level or at community level, and where progress is related to relaxation of institutional or infrastructural constraints, moving forward may clearly be painfully slow. On the other hand, the model may identify simple constraints that, once released, may move the system 'up a full step' in meeting objectives.


Journal for Farming Systems Research-Extension






THE ROLE OF MODELLING


Often answers will relate to lack of official commitment, institutional inadequacies, poor information flows, and poor participatory learning experiences. Experimentation and demonstration are not specifically brought into the cycle of events, but they are not excluded either. The process then is not related to the development of research priorities but to examination of the whole raft of conditions that restrict local people in attaining their (perhaps poorly inarticulated) targets.


A SYSTEMS APPROACH
To this point, discussion has been focused on the importance in development of defining appropriate systems and the role of modelling in the exploration of systems at all levels. Now it is useful to consider the basic concepts of the systems approach and relate them to FSRE.
The fundamental tenants of a systems approach may be set-out as follows:
1. Define the boundary of the system to be studied.
2. Define the relevant sub;systems and the hierarchy in which they are
placed.
3. Analyze the sub-systems and interactions between them by experimen
tation or observation.
4. Understand sub-systems that provide the building blocks for the total
system.
5. Systems synthesis-drawing the sub-systems together-usually by mod
elling.
There are two important issues here for FSRE. The first is concerned with definition of research priorities; the second with issues concerned with the transferability of knowledge.
The prospect of targeting research resources more effectively. Within a systems hierarchy of sub-systems and sub-sub-systems, it is important to define those that form the appropriate bottom level-the finest relevant amount of detail. These are sub-systems for which a clear understanding of structure and function is essential to an operational appreciation of the system defined at the top of the hierarchy. For a crop production system, sub-systems representing, say, soil-water dynamics may be required. On the other hand, for rural community systems, a sub-system dealing with the dynamics of household decision-making may be one of those at the base of the hierarchy-soil water subsystems may represent an unwarranted level of detail.
The issue at the heart of the problemis where research efforts should be directed. Research aimed at gaining a clearer understanding of sub-systems below the relevant bottom level of the hierarchy is likely to have poor pay-offs in terms of the goals for the main system. For example, in semi-arid areas in Southern Africa where concern is with improvement in the social welfare of communities, it may be said that sub-systems concerned with ruminant


Vol. 5, No. 1, 1995






DENT


metabolism are just too far down the hierarchy to warrant research attention. Research into sub-systems dealing with grazing property rights in communal lands, on the other hand, may be at the right level to be targeted. Such research may provide opportunities to improve the welfare of village communities. Unfortunately, both science-led and technology-driven research have usually been directed below the relevant bottom-line of sub-systems. Consequently, a lot is known about detailed science-based systems and relatively little about the wider social systems in which they are embedded. Research into farm livelihood systems and rural community systems is likely to bring increasingly rapid pay-offs if directed at socio-economic and socio-cultural sub-systems.
It is important to recognize here that the debate is not about relatively minor shifts in research resource allocation but rather major changes in research orientation with different skills and different disciplines becoming involved.
The prospect of transferring knowledge. Systems theory implies that subsystems are autonomous within a defined boundary. If this is true, such subsystems should be capable of being applied within a range of systems hierarchies. Thus it should be feasible to transfer knowledge within a subsystem (or at least the sub-system structure) from one location to another. Furthermore, research may be carried out on such systems in various locations by different teams and apply the findings in other locations. This is an approach well recognized in conventional science research but FSRE has been driven by case applications of the FSRE art. There is now a need, relevant to rural systems, not only to improve the efficiency of the development process but also to look for more rapid solutions to problems.
Roling (1994) has defined just such a paradigm (sub-system structure) that has application at a number of different levels. He distinguishes between the actual system on the one hand and the 'platform' for decision making about the system on the other. Others have expressed the same concept in terms of the actual (bio-physical) system and the information system used to manage it. Each, in this case, is more properly defined as a sub-system. Roling (1994) illustrates this by instancing at the farm level, the farm household as the platform and it interfaces with the farm production system. At a more communal level, a group of farm families with a common problem and with common aspirations represent the bio-physical domain, while the platform consists of the collective decision making representatives of the farm and other households involved. Figure 2 shows the systems hierarchy for application at farm level.
The platform that may operate at the communal level as well as at farm household level is a crucially important concept that has had even less research attention than issues related to research prioritization and transferable knowledge within sub-systems. The nature of the platform (its structure, components, and mechanisms) has hardly been analyzed. Figure 2 implies that in any given circumstance (that is for any class of farm household), social, cultural,


Journal for Farming Systems Research-Extension





Figure 2

SYSTEM HIERARCHY
0
0
Farm livelihood system
for defined type z






DENT


and infrastructural elements will determine at least the componentry of the platform. Only the most general view of the actual mechanisms of the decision process are available.
Figure 3 offers a farm-level view of this vital ingredient for FSRE. The platform here is hypothesized as a nested group of decision support systems (Dent, 1994). The farm family has a number of objectives (some specific, others general) falling under the headings of technical, social, and economic. Information concerning each of these is perceived by the farm decision making unit by way of various pathways. A second hypothesis is that providing such information indicates that objectives are being met, then the decisions made will generally coincide with those made during the last production cycle (last season).
This implies that the decision making unit not only has a view about its objectives but also monitors in some way (mainly informally) the extent to which these objectives are being fulfilled. When one or more objectives are not being met then corrective action will take place. In some circumstances, the platform involved is extremely simple, with maybe only one or two decision support systems functioning. For example, in a highly capitalized and intensive European farm system, the economic performance indicators are likely to override most others. More complex situations may be the norm in other places where cultural objectives and perhaps subsistence nutritional levels for the family may be paramount.
In passing, it should be noted that these kinds of processes can be relatively easily constructed into formal models using 'rule of thumb' and qualitative data within formats suggested by expert systems methodology.
It seems only a minor logical step to conclude that FSRE procedures should be directed only toward decision support systems that are 'active' for a group of farms. Indeed, it can be supposed that farm livelihood systems may be satisfactorily classified relative to the decision support systems that apply. Perhaps herein is a prospect of grouping farms within an agro-climatic zone according to useful operational characteristics. It can be argued that adaptive research, whether at farm technology level or at institutional level (i.e., infrastructure or people empowerment), will only be successful if it reinforces existing and operational decision support systems. Similar comments may be directed towards rural policy; to be effective it must bear directly on active decision support systems.
The concepts of platforms, the mechanisms of decision making, and the hypotheses involved are a far cry from conventional notions that farm families seek to attain the best financial performance or even specifically high levels of security and organize their resources accordingly. But the fact is that little is known about the decision-making process and this is a stark contrast with extensive and detailed knowledge of the biological components of rural systems.


Journal for Farming Systems Research-Extension





Figure 3


0
Indigenous Commeclal
Knowledge Media

0
MAKING UNIT 0fo

r Planning 1 z
1nm~ -1r . .1-inf.r-I I I I C






DENT


This debate has touched one paradigm that may be transferable and may indeed operate at a number of levels in organized society. This is but a single example and illustrates the kind of research and its orientation and the kind of approach to modelling that is now required. Other such paradigms easily come to mind: researching unsatisfactorily defined property rights; motivation and empowerment models for women; and creating infrastructural safety nets for rural communities in regions of extensive climate variability.


CONCLUSIONS
The discussions have highlighted a number of issues pertinent to rural development and have attempted to show the relevance and use of modelling.
1. It is questionable whether rural development should be addressed at the farm level.
2. A systems framework is appropriate for rural development research and action. Such work needs to be people oriented rather than technology driven.
3. Modelling is intrinsically part of the development process and more formal methodologies have advantages over conceptual models. Newer modelling techniques can utilize qualitative data and can mimic behavior patterns of individuals and communities.
4. More explicit use of modelling encourages the definition of concepts into general paradigms and theories. It may also encourage more attempts to put some order into sociocultural heterogeneity (classification). Both ofthese may assist in the transference of knowledge from one region to another.


REFERENCES
Anderson J.R. 1985. Assessing the impact of farming systems research: framework and
problems. Agricultural Administration 20:225-235.
Benoit M. 1994. Environmental issues: Use of Farming Systems Research/Extension to
resolve environmental and spatial problems. Dent JB & McGregor (eds). Farm &
Rural Systems Analysis: European Perspectives, CABI, Wallingford.
Brossier J. and Chia E. 1994. Participatory research: Water quality and changes in farming
systems, in, Dent JB & McGregor MJ (eds). Farm &Rural SystemsAnalysis: European
perspectives, CABI, Wallingford.
Brossier J., Vissac B. and Le Moigne J.L. 1990. Modelisation Systematique & Systeme
Agraire. Decisionet Organisation, INRA, Paris.
Byerlee D. and Tripp R. 1988. Strengthening linkages in agricultural research through a
farming systems perspective: The role of social scientists. Journal of Experimental
Agriculture 24:137-151.
Checkland P. 1993. Systems Thinking, Systems Practice. John Wiley & Sons, Chichester. Dent J.B. 1994. Enabling technologies for land use and resource management: The
human response, Proc. 5th International Congress for Computer Technology in
Agriculture, Stoneleigh, RASE, pp40-45.


Journal for Farming Systems Research-Extension







THE ROLE OF MODELLING


Dent J.B. and Edwards-Jones G. 1991. The context of modelling in future: the changing
role of R & D funding in agriculture. Aspects of Applied Biology 26:128-133.
Dent J.B. and McGregor M.J., 1993. A systems approach to conflict resolution in rural
resource management. Proc. International Fed. of Operations Research Societies
Conference, Lisbon.
Dent J.B. and Thornton P.K., 1988. The role of biological simulation models in Farming
Systems Research. Agricultural Administration & Extension 29:111-122.
Doppler W., 1994. The role of quantitative methods in integrating farming, village and
regional systems approaches, This Conference.
Edwards-Jones G., Dent J.B., Morgan 0., McGregor M.J. and Thornton P.K., 1994.
Incorporating farm household decision-making in whole farm models. International Benchmark Sites Network for Agrotechnology Transfer: A systems approach to Research
and Decision Making. Kleuwer, Amsterdam (in press) in Tsuji G. (ed).
Farrington J., 1988. Farmer participatory research: Editorial introduction. Expl.Agric
24:269-279.
Jones C.A. and Kiniry J.R. 1986. CERES- Maize:A simulation model ofmaizegrowth and
development. Texas A & M Univ. Press, College Station.
McAdam R., Van Asch R, Hedley B., Pitt E. and Carrol P., 1994. A case study in
development planning using a systems learning approach: generating a master plan for
the livestock sector in Nepal. Agricultural Systems. (in press)
Olsen M.E., Canan P. and Hennessy M., 1985. A value-based community assessment
process. Integrating quality of life and social impact studies. Sociological Methods &
Research 13:325-361.
Roling N., 1994. Interaction between extension services and farmer decision making: New
issues and sustainable farming, in, Dent J.B. & McGregorM.J. (eds), Rural&Farming
Systems Analysis: European perspectives. CABI, Wallingford
Romero, C. & Rehman, T. 1989. Mutliple Criteria Analysis for Agricultural Decisions.
Amsterdam, Elsevier, p257.
Wossink, G.A.A., Koeijer, T.J., Renkema, J.A. and de Koeijer, T.J. 1992. Environmentaleconomic policy assessment: a farm economic approach. Agricultural Systems39:421438.


Vol. 5, No. 1, 1995












New Resources for International

Agricultural Cooperation:

Village-Based Self-Help and Agricultural

Research in Japan'

Teruo Wada2, John S. Caldwells, and Shigeki Yokoyama4



ABSTRACT

Participatory elements in Japan's domestic agricultural development history are compared with its international development assistance.
Three movements which increased self-help in rural development and farmer participation in agricultural research from the 19th century are described, culminating in the systems research concept of s6g6 kenky' and its application in the 1952-62 eindshiken national program of onfarm trials. Japan's government-to-government agricultural development assistance has emphasized self-help at the national level, but farmer participation in implementation has been limited. An exception to this has been the cataloging of 449 indigenous technologies.
Japan's domestic experience and the registry of indigenous technologies could serve as models to increase farmer participation in agricultural development cooperation in Japan and other countries, through formalization of the assessment of indigenous technologies and knowledge systems as local resources for self-help; involvement of area specialists; application of participatory methods developed in, Japan; and establishment of a registry of farmer and extension
practitioners who could participate in development assistance.





1 Keynote paper presented at the Thirteenth Annual Association for Farming Systems ResearchExtension Symposium, Montpellier, France, November 21-25, 1994.
2 Professor, Department of Agricultural Economics, University of Tokyo, Tokyo, Japan.
3 Associate Professor, Department of Horticulture, Virginia Polytechnic Institute and State University, Blacksburg, Virginia, and former Abe Fellow and Visiting Researcher, 1992-1993, Department of Agricultural Economics, University of Tokyo, Tokyo, Japan.
4 Senior Researcher, Section of Comparative Farming Systems, Department of Farm Management,
National Agriculture Research Center, Tsukuba, Japan, and former Japan International Cooperation Agency (JICA) expert, UN EStAP CGPRT Center, Indonesia.


Vol. 5, No. 1, 1995






WADA ET AL.


INTRODUCTION
The assumptions and methods used in international agricultural assistance and cooperation have undergone an evolution over the past 50 years. This evolution has increasingly come to recognize the knowledge and capacity of farmers and rural inhabitants for self-improvement, when their initiatives are supported institutionally.
During the 1950s and 1960s, while there were major changes in perceptions of the rationality of traditional agriculture and farmers and rural inhabitants, a perceived need for an outside impetus for change was a constant assumption. In the 1950s, developing country agriculture was seen as being often tradition-bound and economically irrational. Change from traditionalism to rationalism and adoption of developed country agricultural technology through integrated rural development promoted by polyvalent village workers was advocated and implemented in many developing countries.
By the 1960s, however, the rationality of traditional agriculture and farmers had come to be recognized. The Green Revolution technology of highyielding seeds of rice and wheat accompanied by synthetic fertilizer was substituted for developed country technology, and farmers were expected to respond as economically rational producers to this superior technology introduced through massive extension and credit programs such as Masagana 99 in the Philippines (Caldwell, 1994b; Chambers, 1993; Ruttan, 1975; Ruttan, 1977; Schultz, 1964).
The 1980s and the 1990s have seen much greater recognition of farmers' and rural inhabitants' own capacities for change. From the beginning, farming systems research recognized the essential role of farmers as partners in sitespecific agricultural research. More recently, increased emphasis on participatory methods gives farmers the central role in analyzing their own farming systems and proposing and designing change (Chambers, 1993; Shaner et al., 1982).
Today, Japan is the largest donor of development assistance, known by the acronym ODA ("Official Development Assistance") in Japan (Ministry of Foreign Affairs, 1990). Concurrently, Japan is changing from a regional actor to a global actor in international agricultural assistance. Its efforts are increasingly moving both geographically and in types of farming systems beyond its traditional domain of technical assistance in wet rice production and associated farming systems in Southeast Asia. At this time of change, it is useful for Japan to look back on its own experience in light of the above international evolution in assumptions and methods of agricultural development assistance, to ask what aspects of its own domestic agricultural development history are most likely to be relevant to its current international efforts.
This paper seeks to address this question through an examination of several key elements of the history of rural development and agricultural research in Japan, as a basis for comparison with its approaches to international agricul-


Journal for Farming Systems Research-Extension






VILLAGE-BASED SELF-HELP


tural development assistance. The paper first provides an overview of the history of self-help in rural development in Japan, as background to presenting the history of the development of the agricultural station research system in Japan. The paper then shows how these two strands came together in the eindshiken program of 1952-1962. The paper next summarizes the approaches used in Japan's overseas development assistance, and compares these with the self-help tradition. The paper concludes with suggestions on how current overseas development assistance by both Japan and other countries could draw from the traditions of village-based self-help and agricultural research in Japan.


FARMER PARTICIPATION IN RURAL DEVELOPMENT AND AGRICULTURAL RESEARCH IN JAPAN


The Self-Help Tradition in Rural Development
Farmer participation in agricultural innovation and rural development has a long history in Japan. Three movements stand out as prominent examples of this history: the Senzokabu Kumiai cooperative in the first half of the 19th century, the ChdsonZe Movement from the late 19th century to the 1920s, and the Ndsangyoson Keizai Kdsei Keikaku during the Great Depression of the 1930s. In each of these movements, we can see a pattern in which farmer initiative and participation intersect with the central administration, reflecting the dual nature of the Japanese mura.
The mura or village community is the basic unit of both land use and local governance in Japan. Most mura originated with the initial settlement by groups of families in a previously unexploited valley or other area of cultivatable land. Many murain Central and Northern Honshu date back to the defeat of the Heike clan in a famous civil war Southern Japan that ended in 1195, but others were founded in later centuries, particularly in Northern Honshu. Japan also has a long history of central government, dating back to the seventh century, initially modelled after the Chinese Imperial system by leaders of clans who sought greater power as a result of social differentiation within Japan in the early centuries A.D.
While rainfed agriculture has played an important but less recognized role in Japanese agriculture, the role of irrigated rice farming has had a major impact on the relationship between the mura and the central government. Many mura have been able to develop networks of canals from small rivers, increasing their agricultural production through better irrigation and improvements in farming practices not dependent on the central government. This has provided the basis for the development of a tradition of local, selfgoverning autonomy in mura. In such mura, the central government is an outside force that draws on the mura's social organization to promote its objectives, and also extracts local wealth from the mura through taxation. But


Vol. 5, No. 1, 1995






WADA ET AL.


for the further development of agriculture, irrigation systems drawing from large rivers and large-scale land reclamation, serving many mura, become necessary. In these cases, mura become more dependent on the central government. Hence, mura have developed a dual nature being both an indigenous unit of local autonomy and a subordinate unit of centrally-derived administration (Goto and Inamura, 1993).
Senzokabu Kumiai. Ohara Yugaku ' (1797?-1858) was said to be the second son of a high official of one fiefdom in western Japan. At the age of 18 he separated from his family for reasons that are unclear, and as a result moved around and lived in a number of different parts ofWestern Japan over a 20-year period. This enabled him to acquire a wide practical knowledge of agriculture. Then, about 1835, he finally settled in the village of Nagabe in Shimosa (present Chiba Prefecture). There he organized one of the earliest agricultural cooperatives in Japan, called the Senzokabu Kumiai,6 to restore prosperity to this village which had fallen on hard times.
The SenzokabuKumiai was a kind of a joint trust fund based on mutual aid. Each member farm household contributed a portion of its land to the cooperative. The cooperative then let the pooled land out to tenants to farm, and used the tenancy rent to build a fund for cooperative member use. The cooperative used the fund to make loans to members in need. Through the trust fund, members who had gone bankrupt were able to reestablish themselves, and members who had had to pawn their farm land to lenders from other villagers were able to buy their land back.
The Senzokabu Kumiai carried out other activities as well. It purchased supplies and materials in bulk for the membership. It arranged for pooling and consolidation of fragmented farm land, even extending to moving the site of the farmhouse in the process. When fields had to be redemarcated for such consolidation, members donated their time voluntarily to the task.
Field consolidation was done with the objective of improving farming practices, and was part ofa series of innovations that Yugaku and the Senzokabu Kumiai introduced in the village. Field consolidation made it possible to drain low-lying, bog fields so that they could be used for double cropping ofdryland crops after rice. It also made it possible to introduce straight-line rice planting. Other practices that Yugaku introduced included the amendment of loam to improve soils, improved rice seedbeds, and better weeding. He also introduced the practice of keeping a calendar of farm activities. Some of the farm households in the village continued until the 1950s to keep this type of calendar that he introduced more than 100 years ago. This shows how strong
5 In this paper, we write Japanese proper names in the order they are written in Japanese: family
name first, followed by personal name (the opposite from English). Hence, Ohara is the family name and Yfigaku the personal name of Ohara Yfgaku (which would be written Yfigaku Ohara if in the English order). In addition, prior to the modern era that began with the Meiji Restoration in 1868, persons are often referred to by their personal name (Yfgaku, referring to Ohara Yfgaku), whereas from the modern era persons are always referred to by their family names (Maeda, referring to Maeda Masana).


Journal for Farming Systems Research-Extension






VILLAGE-BASED SELF-HELP


an influence he left on the village (Society for the Study Agricultural Development History, 1953; Kuwahara, 1974).
The dual nature of mura in Japan can be seen in the effects of Yugaku's work. Yugaku did not intend to lead farmers in a revolt against the central government, but the cooperative use of land that he promoted conflicted with the land tenure system of the Edo government. Under the feudal system, all land ownership rights were invested in the lord ofthe fiefdom, and farmers had only use rights given to them by the lord. The use rights permitted inheritance but not transfer of land, and farmers were not allowed to move from one piece of land to another. Thus, the pooling of land in the cooperative was contrary to the land control system of the feudal government. The Edo government and the various fiefdoms built river dikes and carried out land reclamation to open up new paddy areas, but they did not engage in field reorganization. Only towards the end of the Edo period did some wealthy farmers, who were often also local officials, carry out field reorganization as part of the introduction of new agricultural practices like straight-line planting, horse plowing, and field drainage. Some small farmers appear to have done field reorganization in secret, hiding this from governmental authorities, but Yugaku's organizing of small farmers into a cooperative was a radical step. For this reason, the Edo government broke up the senzokabu kumiai, and Yugaku himself was disgraced and forced to commit ritual suicide.
Yugaku's followers nevertheless continued his work, and both Yugaku himself and his followers wrote a number of books describing his approach, but his approach was not adopted beyond the eastern part of Chiba, nor did Yugaku's ideas become well known until the 1910s. In 1911, the economist and member of the House of Peers of the pre-war Diet Tamura Inajiro assembled a 566 page Complete Works of Yugaku, and the Ch6sonZe movement described in the next section adopted Yugaku's philosophy as one of the guiding principles of the movement. It is fair to say that the ideas and concrete methods that Yugaku first put into action are the foundations of the modern cooperative movement in Japan today (Kimura, 1981).
Ch6son Ze Movement. Ch6son Ze 7 was a policy of stimulating rural industries, focusing on although not limited to agriculture, that was carried out in villages throughout Japan from the end of the 19th century until the 1920s. The idea of Chdson Ze was first advanced by Maeda Masana in 1893. Maeda was a high official in the Ministry of Agriculture and Commerce. He proposed ChdsonZe as a movement to be undertaken by agricultural societies called Nakai. Ndkai were local organizations that worked for the improvement of agriculture.
6 Senzo are ancestors, and kabu means stock (in the financial sense). Kumai is a cooperative or union,
hence Senzokabu Kumai literally means "Ancestors' Stock Union."
7 Chdson is a word combining ch8 and son, township and village. Ch8 is the Chinese-style reading for
township, read machine indigenous Japanese reading, or kunyomi, when used as a single word. Son is the Chinese-style reading for villages, read as mura in kunyomi when used as a single word. Ze is usually used in combination with other characters, such as in zesei, to correct a situation. Hence, ch6son ze is to "rectify" townships and villages: to put their economic situation on a more stable
and self-sustaining basis.
Vol. 5, No. 1, 1995






WADA ET AL.


Initially, Ch~sonZe began as a locally organized movement as envisioned by Maeda, but from 1904 this began to change. With the outbreak of the RussoJapan War, the government began to take over Ch4son Ze and carry it out in a standardized manner as a national policy. In this policy, township and village planning was officially encouraged under the names of Ch6 Ze (Township Rectification) and Son Ze (Village Rectification). Each township and village first carried out an analysis of its existing situation through a survey and then submitted a plan for stimulating the rural economy. Townships and villages were the basic units of government in Japan, and their plans became the basis for planning at the higher levels of the county (Gun Ze), the prefecture (Ken Ze), and the nation (Koku Ze).
This hierarchy ofplanning starting from townships and villages reflected the belief of Maeda that successful planning had to be built up from the most basic unit of government. As each township and village made a plan centered around the promotion of its most typical economic activities and enterprises, the economy of the region and the nation as a whole would be stimulated.
The Chdson Ze planning surveys included an inventory of land holdings, farm production, non-agricultural industries, and sources of township or village public revenue. From tie results of the survey, Ch~son Ze plans promoted both technical and socio-economic change. On the technical side, they advanced such changes as selection of improved varieties from farmer seed stock using the salt water method, field reconsolidation, and introduction of improved agricultural practices. On the socio-economic side, they included measures to protect tenants, organization of cooperatives for joint purchase of supplies and materials, repurchase of land that had fallen into the hands of nonvillagers, farm household financial management to encourage economization of expenses and increased saving, and education of children (Takeda, 1960; Tanaka, 1977).
While Ch6son Ze planning thus came to be promoted in a top-down way, it contributed to bottom-up self-help in two key ways. First, it established a tradition of self-analysis for planning, and gave this official approval. Second, it recognized local planning and initiative as the basis for developing national planning policy. Although local government was forced to serve as a means of implementing national policy, in essence it became the source of that national policy at the same time.
Nsangyoson Keizai Kdsei Keikaku (Agricultural, Forest, and Fisheries Village Economic Renewal Planning Program). This rural development planning program began in 1932, in an attempt to address the economic difficulties of villages during the Showa Depression (corresponding to the Great Depression in the United States). This program proposed to select 1,000 townships and villages a year and have each prepare an economic renewal plan. In return, these villages would receive a small amount of assistance from the national government to implement their plans. By 1940, 82 percent of the townships and villages in Japan were in this program.


Journal for Farming Systems Research-Extension






VILLAGE-BASED SELF-HELP


The Village Economic Renewal Program envisioned using the spirit of mutual aid that characterized village culture to reorganize village and township economies in a more organized, planned manner. Concrete measures included development of production plans, organization of producers' cooperatives for the sale of agricultural products, the purchase of supplies and materials, and provision of credit to farmers.
Development of the economic renewal plans was left to local initiative, but few regional characteristics were reflected in the actual plans that emerged. Most followed a similar "cookbook" pattern that in effect envisioned a return to a subsistence economy based on self-help. Key elements of the model plan prepared by the Imperial Farmers' Society (the national-level organization to which the local Nkai belonged) were: (1) increasing the percentage of village self-sufficiency in food, feed, and fertilizer production; (2) reducing costs of production, and in particular minimizing cash outlays; (3) formation of marketing and joint purchasing cooperatives; (4) increasing production; and
(5) achieving economies in domestic expenses (Tanaka, 1977).
This program can be seen as an attempt by the government to use the rural traditions of self-help and self government for its own political reasons. The government had no resources for improving rural infrastructure or otherwise providing financial support to villages, so to deflect potential criticism of lack of a rural policy, it chose to fall back on exhortations to villages to be more selfsufficient. Nevertheless, while the government's motives may have been political, and many of the plans tended to be "cookbook," this program had the effect of continuing to encourage and recognize value in villagers' selfanalysis and planning. In this respect, although more by default than by design, paradoxically the government's lack of financial resources to carry out national programs conceived by central authorities forced it to recognize and promote rural self-reliance and self-help.

Development of the Station Agricultural Research System
Japan has had a nationwide agricultural research system for approximately 100 years. Japan also has a long history of farmer participation in agricultural research and development. This participation goes back to the beginnings of modern agriculture in Japan, in the early Meiji period (1868-1912). This reflects the fact that the development of modern agricultural science in Japan resulted from a process of the synthesis of indigenous agricultural science and agricultural science brought in from the West.
Initially, between 1871 and 1886, the Meiji government sent high government officials to the West to learn about agricultural practices and its systems of agricultural research and education. When the first agricultural college was opened, the Meiji government also hired instructors from England, the United States, Holland, and Germany to carry out research and education. At the same time, veteran farmers (rn) and innovating farmers (tokun6)


Vol. 5, No. 1, 1995






WADA ET AL.


continued research begun in the latter part of the Edo Era on indigenous agricultural methods (Saito, 1970).
At first, there was no contact between the two types of research, Westernderived and indigenous. The government rejected tokun6 technology as lacking a scientific basis, and for their part tokun6 relied only on experience and ignored modern analytical science. However, as the government began to realize that it would be extremely difficult to transplant directly Western agriculture and agricultural science under the completely different natural and socio-economic conditions of Japan, it came to recognize through a process of trial-and-error that a modern agricultural science for Japan would have to be built upon the incorporation of knowledge from modern Western agricultural science into indigenous agricultural technology and science. At the same time, the tokund themselves also began to incorporate some of the results of research by agricultural scientists into their own thinking and practices. They treated the results of agricultural research like technology from another region of Japan that they combined with their own practices, within the framework of indigenous knowledge (Saito, 1970). It is important to keep this process of incorporation in mind to understand the background behind the establishment of the national agricultural experiment station system.
The government first began to recognize the importance of tokun6and r~n6 in the 1880s. After the largely unsuccessful attempt to import Western agriculture into Japan in the 1870s, the Meiji government requested each prefecture (equivalent to a state or province in North America, a dipartement in France, or a cercie in West Africa) to identify one or two leading rdnd. These farmers gathered information on practices and varieties developed or identified by innovating farmers, or tokun6, and disseminated them to other farmers. In 1885, the work of the rdn6was formalized in the Itinerant Instructor System (Hayami et al., 1979; Ogura, 1968). This parallels one of the roles for farmerparticipatory research proposed by Chambers (1993).
Prior to the establishment of the Itinerant Instructor System, the first forerunners of the national agricultural experiment stations were established. In 1876, the Sapporo Agricultural College was established; this would later become the College of Agriculture of Hokkaido University. In 1878, the Komaba Agricultural College was established; this became the College of Agriculture of Tokyo University. These were the main pathways through which Western agricultural science entered, with all courses initially taught by foreign instructors. After graduation, students of these two schools went abroad for further study before returning as instructors in their schools of origin. These instructors were able to provide much more realistic instruction than the foreign instructors had been able to do. They also began to carry out research to determine the scientific basis of indigenous agricultural practices whose real-world validity had been proven empirically. In this process, they were aided by the fact that some rdn6 were appointed instructors at the Komaba Agricultural College. At Sapporo Agricultural College, in Hokkaido


Journal for Farming Systems Research-Extension






VILLAGE-BASED SELF-HELP


where Western agricultural techniques were more directly applicable, returned instructors began to conduct experiments on problems associated with the agriculture of newly-opened lands (Hayami et al., 1979; Ogura, 1968; Saito, 1970).
The impetus for the establishment of the national agricultural experiment station system was the economic depression in the rural areas and the debate about its causes and proposed solutions during the 1880s. One school of thought took the position that the way to rebuild depressed rural economies was through reduction of the high rate of tenant rents, and that this in turn required reduction of the land tax. Government agricultural scientists took the position, however, that the main cause of high tenant rents was competition for land by tenants, and that lowering the land tax would not change this. They argued that a more fundamental solution could only be obtained by increasing productivity through the development and extension of new, more performant technology. A call arose for the establishment of a national agricultural research station as an institution to carry out such development and extension of improved technology (Saito, 1970). This technology-centered solution proposed by government agricultural scientists has parallels with the conclusions advanced by Schultz (1964) that traditional agriculture could only be improved through new technology.
The initial conception of a national agricultural experiment station came from Sawano Atsushi. Sawano became an Itinerant Instructor for the Agricultural Bureau of the government after his graduation from the Komaba Agricultural College. As he went around giving lectures in different parts of the country, he became increasingly aware of the need for an experiment station to determine scientifically if the content of what he presented was indeed appropriate to each location. In 1886, he first advanced a proposal to establish an experiment station. This was based on the need common to all the Itinerant Instructors for better information to extend to farmers. He believed that such an experiment station could respond to this need by carrying out applied research closely linked to real-world production (Saito, 1970). In essence, Sawano's proposal sought to give the Itinerant Instructors a broader "menu" of technologies based on what has been called "upstream" farming systems research (Gilbert et al., 1980).
Sawano's proposal was not immediately implemented, but the first steps towards an agricultural experiment station system soon appeared. Between 1886 and 1890, "key grain and vegetable test production" trials were begun at 10 sites on rented farm land in the greaterTokyo area. Trials on rice, wheat, barley, and vegetables were given to r6n6 to carry out. The real-world experience of r6n6 was seen as essential to achieve the multiple objectives of these trials. Their main objective was to develop technology that would meet both technical and economic requirements of farmers' production. This objective closely parallels the "can adopt" and "willing to adopt" objectives for on-farm trials of Norman (1983). At the same time, they also had an


Vol. 5, No. 1, 1995






WADA ET AL.


educational and demonstration objective, to show farmers new techniques (Saito, 1970).
These early trials were simply empirical tests of new technology, not carried out according to a systematic methodology for comparing it to existing technology. Gradually the need for more systematic comparisons came to be recognized. In response to this recognition, in 1890 the "Provisional Experiment Station of the Agriculture Bureau" was established in Nishigahara, just north of Tokyo. This was followed by the establishment in 1893 of six branch stations in the prefectures ofMiyagi, Ishikawa, Osaka, Hiroshima, Tokushima, and Kumamoto. With this a national agricultural experiment station system came into being (Saito, 1970).
Each branch experiment station carried out some research on commercial crops of particular importance to its region, but the overall the emphasis of the experiment station system was on rice research. This basic emphasis on rice has remained unchanged to the present. Initially, applied research and demonstrations made up the bulk of the work of the stations, and basic research did not play a prominent role. Applied research and demonstrations are more properly the province of prefectural experiment stations and research carried out at the local level, but these institutions did not come into being for some time. The national experiment stations also lacked the facilities needed to do basic research (Saito, 1970).
In addition to their research role, the national experiment stations carried out a number of other roles, essentially extension in nature. At the request of local government officials, they would send staff members to give lectures on agriculture and serve as judges of agricultural products, improved agricultural tools and implements, and improved agricultural practices at agricultural fairs and contests. They collected local varieties for research and breeding, and on request they carried out free promotion and distribution of seed of varieties they selected or bred at the stations. For example, from 1890 to 1894, they distributed seed of six varieties of wheat and barley to some 90,000 farmers. They provided analytical and evaluative testing services for soils, fertilizers, feeds, and agricultural products. They also responded to requests from farmers for advice and for observational visits to tests that farmers themselves initiated (Saito, 1970).
The extension role of the national agricultural experiment stations was due to several factors. The current system of governmental extension offices was established only after World War II. Before World War II, extension was carried out primarily by the farmers' associations (Ndkai) and production cooperatives, with technical support from the national experiment stations. Even after the establishment of the governmental extension office system, farmers' cooperatives (now called N6ky6) have continued to play an important role in extending new technology, especially for specialty crops such as locally labelled fruits and vegetables. The main task of the government extension system has shifted to lifting up small-scale and marginal farmers. In this sense,


Journal for Farming Systems Research-Extension






VILLAGE-BASED SELF-HELP


the two forms of extension (farmer-based and governmental) complement each other. We also see again the role that farmers, through their associations, have continued to play in promoting agricultural technology change.
The development of the prefectural experiment station system began with the promulgation of an ordnance in 1894 for their establishment. In 1899, the national government passed a law providing for partial financial support for these stations from the national budget. As a result, by 1903 every prefecture had its own agricultural experiment station. With the establishment of the prefectural experiment station system, the national experiment stations abandoned three of their branch stations (Miyagi, Hiroshima, and Tokushima) and began to reorganize themselves to focus more on basic research (Saito, 1970).
The tradition of drawing on tokun6 knowledge continued at the prefectural experiment stations. Prefectural experiment station technicians took over the role that the rdn6 had played in the Itinerant Instructor System. These technicians were trained at new agricultural high schools and at training centers at the national and prefectural experiment stations. Their training thus gave them knowledge of the results and methods of formal, Western-derived agricultural science. At the same time, however, since the mission of the prefectural experiment stations was to carry out applied agricultural research on real-world problems, they interacted with tokun6 and drew from tokund knowledge (Hayami et al., 1979; Ogura, 1968). One can say that it was these technicians who were the real synthesizers of Western-derived science and indigenous knowledge within the agricultural experiment station system. Even up to the 1960s, informal interaction between tokun6innovating farmers and prefectural experiment station researchers played an important role in generating new technology (comments by researchers in a discussion following a seminar on FSRE given by Caldwell at the National Agricultural Research Center, Tsukuba, Japan, July 7, 1992).

Convergence of Self-Help and Agricultural Research: The Eindshiken Program
The ein6shiken program was a national program of on-farm trials carried out from 1952 to 1962. The word ein6shiken could be translated as "agricultural management trial." The first word, ein6, literally, "managing agriculture," combines concepts of planning and evaluating one's overall farming with physically carrying out all the concrete steps of producing each crop and type of animal in one's farming. A shiken is simply a "test," so eindshiken means a test of how to improve one's overall farm management through specific production activities.
Ein6shiken were on -farm trials carried out with farmer collaborators, usually for a period of three to four years with the collaborator. Comparisons were made across farms, between farmer-collaborators and noncollaborators. The trials did not attempt to change the entire farming system of collaborators at once, however, but rather began with specific technologies. Thus, ein6shiken


Vol. 5, No. 1, 1995






WADA ET AL.


can be seen as on-farm trials with double objectives: to find better means of managing agriculture, and to test specific technologies in the context of the whole farm. They can be seen as systems tests of specific technologies.
The theoretical basis of eindshiken can be found in a theory of systems research that had been developed in the 10 years preceding the establishment of this national program. A type of systems research called sdg6 kenky4, or comprehensive research, was proposed with the creation of a new agricultural station in the Tohoku region of northern Honshu, established for the specific purpose of initiating this new type of research.
There are several parallels between agriculture in the late 1940s in Tohoku, and agriculture in the parts of the developing world where FSRE originated, West Africa, Central America, and Southeast Asia. Like those regions, Tohoku also has many upland areas suited only for rainfed agriculture. Farmers in the Tohoku region had similarly evolved complex rotation systems with a range of rainfed crops, including foxtail and barnyard grass millets, wheat, buckwheat (soba), and soybean. Agriculture was predominately subsistence, with shifting cultivation and land preparation by hand co-existing with cultivation in permanent fields using animal traction (Sasaki, 1988).
The theory of s4g6 kenkyf adapted an ecological model of feedback and equilibrium to a concept of a farming system seen as consisting of three main elements, people, crops, and animals (Nishigori, 1952; Iwasaki, 1952; translation into English of key sections and more detailed discussion of their implications are provided in Caldwell, 1994a). This closely paralleled similar ideas thirty years later, when the theoretical underpinnings of FSRE appeared: the ecologically-based systems analysis of Hart (1979); the concepts of equilibrium and feedback in agricultural systems of Ruthenberg (1980); and the household, crop, and animal subsystems of the structural models of McDowell and Hildrebrand (1980). S6g6 kenkyA also proposed methods similar to FSRE. It sought "the harmonious fitting together of components," assuming that the new "harmonious equilibrium, or combining, of complex things cannot be achieved at once, but is the result of many efforts of trial and error, achieved only through real tests of management." Sdg6 kenky4 combined the "static observation" of social science survey research and the creation of a "dynamic perturbation" in the system by technical research. The objective of this combination of methods was to subject technology to a "socioeconomic filter" (Nishigori, 1952; Iwasaki, 1952; Suzuki et al., 1958).
The eindshiken program began in 1952, when Japan was not yet fully out of an era of subsistence agriculture, spanned the period of post-war economic recovery through the 1950s, and continued until 1962. The year 1960 can be seen as a watershed year, when a new administration came into power after a political crisis and proposed a program of rapid economic transformation, capsuled in the promise to "double income." As part of this program, in the following year, 1961, a new Basic Agriculture Law was promulgated. This Law established the objective of achievement of income parity between agriculture


Journal for Farming Systems Research-Extension






VILLAGE-BASED SELF-HELP


and industry. With the movement of rural labor into the rapidly-expanding industrial sector, the creators of the Law envisioned a consolidation of farm land into fewer, larger farm units; mechanization of the remaining farm units; and increasing specialization of agriculture, particularly into commodities such as animal and fruit tree production for the urban market with its increasing incomes and diversifying food preferences. In other words, after 1960, agriculture in Japan began to be transformed to an industrial model. The eindshiken program was thus the last stage in culmination of efforts to improve agriculture under conditions of diversity and subsistence, prior to entering an industrial transformation.
Over the 10 years from 1952 to 1962, three series of eindshiken were conducted at a total of 389 sites. Figure 1 shows the distribution of the types of einrshiken carried out over this period. Rainfed agriculture was the most common single type, comprising 24 percent of the total trials. Trials focusing on rainfed agriculture, animal husbandry, and rice production spanned both the first and second series of trials (1952-1955 and 1956-1961), continuing over nine of the 10 years of the program. Trials on soil improvement, multiple cropping after rice, and fall yield decline syndrome (aki-ochi) were carried out only during the first series of tests (1952-1955). The third series of trials was fundamentally different from the first two, and reflected the new Agricultural Law. Those trials introduced mechanized communal farming during the last two years of the program (1961-1962) (Kodama, 1965; Mori, 1956; Suzuki, 1986).
Three concrete examples from the first series of trials will illustrate how these eindsbiken trials had many parallels with on-farm trials in developing countries today.
In the Kanto region, one set of eindshiken trials on rainfed crops involved improved relay cropping through the substitution of wheat for barley as the first crop in the rotation. This permitted a 15-day earlier planting for relay upland rice, on May 10 instead of May 25. Barley could be harvested three weeks earlier than wheat, on June 2 instead June 23. However, this earlier harvest created a labor conflict with the transplanting of wet rice on paddy fields. Due to the labor conflict, farmers would have to employ seasonal labor. The improvement of relay cropping had brought out another problem, which required another solution: an improved rice seedbed permitting wet rice to be transplanted in mid-May instead of early June (Nakamura, 1956).
Another set of trials in the same series focused on the need for improved farm implements. Row seeding improved the growth of wheat and reduced lodging, but it was tiresome to prepare ridges when farmers only had the hoe for land preparation and seeded by hand. New hand implements were developed and tested for making ridges and for seeding in rows. Initially, only men were involved in the trials, and women were left with the simple task of spreading fertilizer. When researchers realized that women had not learned how to use the new implements, they involved women directly in the trials.


Vol. 5, No. 1, 1995








Figure 1. Types of Einoshiken on-farm trials

(389 sites, 1952-1962)


Rainfed agriculture Animal husbandry Soil improvement Multiple cropping Communal farming Fall lodging Rice production


68 17.5% M/z




12 3. 1%


45 11.6%


25 6.4%


81 20. 8%






VILLAGE-BASED SELF-HELP


When both men and women worked together, yield increases were greater (Nakamura, 1956).
Integration of crop and animal production is a focus of much on-farm research in West Africa. Trials with a similar objective were also important in the ein6shiken program. An example was mulberry - dairy integration. Mulberry has long been an important crop in Japan because of its use in silkworm production. With increased demand for milk after World War II, dairy farming had also appeared.
One of the main constraints of mulberry producers who had begun dairy farming was the provision of feed for cows. Farmers planted feed crops between mulberry rows only in the winter, avoiding summer feed crop interculture because of negative effects on mulberry leaf quality. Farmers depended largely on purchased feed for their dairy cows.
The trials explored a number of ways to improve land use for better integration of mulberry and dairy cow production. The most important was summer interculture using a wider spacing between the mulberry rows. In the farmer system of 120-150 cm spacing between rows, the mulberry tree canopies would fill the space between rows in the summer, and interculture was not possible. With new spacings of 300 to 420 cm, a summer soiling crop could be planted without decreasing mulberryyield (Kant6 T6san Agricultural Experiment Station, 1959). The improved method resulteded what today could be called "mulberry agroforestry."


FARMER PARTICIPATION IN JAPAN'S
OVERSEAS DEVELOPMENT ASSISTANCE


Government-to-government Agricultural Development Assistance
Japan's government-to-government agricultural development assistance consists of two main components: programs of the Japan International Cooperation Agency (JICA) and government-supported volunteer programs. In addition, a non-profit corporation created by two government ministries interacts with JICA and both government and non-government volunteer activities to catalog indigenous technology in developing countries.
JICA Technical Cooperation. Japan's government-to-government agricultural development assistance is carried out primarily by JICA. Research support for the solution of technical production problems in tropical agriculture identified through JICA field research or by host country research and extension is provided by the Japan International Research Center for the Agricultural Sciences (until 1993 called the Tropical Agriculture Research Center, or TARC). JICA provides three types of technical cooperation: project-type technical cooperation, development studies, and development cooperation. The main goal of JICA's technical cooperation is the development of human resources in the countries where it provides assistance. Vol. 5, No. 1, 1995






WADA ET AL.


In project-type technical cooperation, JICA provides a coordinated response to needs identified by host countries through three types of cooperation: dispatch of technical experts, training of host country research and extension personnel, and provision of needed equipment. Project-type technical cooperation aims at transferring Japanese knowledge and technology. Preliminary and long-term studies begin at least two years before the inception of project assistance, to assure that the project will address real needs. Human resource building is accomplished by having on-site JICA experts work with host country counterparts in their country, and by bringing host country counterparts to Japan for technical training.
In 1992, JICA had 73 project-type technical cooperation programs in 34 countries. This represented a three-fold increase in the number of programs and a doubling in the number of countries from the 1975 figure of 25 projects in 16 countries. Table 1 shows the breakdown of the 73 programs by type of project (agriculture, livestock, forestry, and fisheries) and region (East, Southeast, and South Asia; West Asia and North Africa; Sub-Saharan Africa; Latin America; Oceania). While agriculture comprises 53 percent of the programs, other areas, particularly forestry, are increasing in importance. East, Southeast, and South Asia has declined to approximately half (56 percent) of the total number of programs, with Latin America now comprising nearly a quarter (24 percent) of the programs.
JICA conducts six types of development studies, of which long-term master plans and feasibility studies are the most important. These studies are increasingly moving from a production infrastructure to a resource management focus. In agriculture, irrigation and drainage development studies have predominated, but rural and environmental development studies are now becoming more important. Similarly, in forestry and fisheries, studies designed to improve resource management for sustainable use are increasing in importance. These studies involve sending a team of experts from Japan to the host country to carry out detailed field surveys. Host country counterparts participate in the field work and thereby strengthen their own survey and analysis techniques. After the first field survey, the data obtained are studied in detail in Japan, and an interim report prepared as the basis for a second field survey. The survey work may also be supported by on-site technology verification trials. At the end, a final report is prepared in the language of the host country, and a seminar on the results presented.
In 1992, 45 development studies were underway in 26 countries. Fourfifths (80 percent) focused on agriculture, with the balance divided between forestry (11 percent) and fisheries (9 percent). East, Southeast, and South Asian countries made up 46 percent of the countries with these studies, with the remainder divided among West Asia/North Africa (19 percent), SubSaharan Africa (15 percent), and Latin America (19 percent).
Development cooperation is a program that links the private sector and agricultural development. In this program, JICA provides loans to Japanese


Journal for Farming Systems Research-Extension






VILLAGE-BASED SELF-HELP


companies for experimental projects and facilitates technology improvement, feasibility studies, verification studies, and technical guidance. Three-fourths (76 percent) of the 25 development cooperation programs in 1992 were for experimental projects. In experimental projects, JICA loans support the collection of necessary data on natural and social resources and technical research to develop appropriate technology, to reduce the risk for the private sector to provide new technology on a commercial basis. Experimental projects involve fruit, horticulture and other special crops; beef cattle; and afforestation (JICA, 1992).
As the above description shows, the emphasis in JICA's approach to date has been to strengthen host country capabilities by providing technical knowledge and infrastructure. At the national level, JICA has emphasized selfhelp. However, at the local level of implementation, while there are some outstanding examples of farmer-to-farmer cooperation (Iijima, 1986), these have reflected more initiative taken by individual members in specific projects, than a conscious programmatic effort by JICA to make farmer participation a key element of its overall program.
Despite the general lack of programmatic effort to increase farmer participation, some JICA specialists have carried the farmer-based tradition into their international cooperation work, resulting in a field orientation in their work. "Field" here is used in the anthropological sense of the area one is working in, or in the sense of the "real world," and not in the sense of a crop field. This field orientation is expressed by three sayings frequently used by JICA specialists coming from this tradition:
" "[In development], you have no right to speak without field experience first."
" "Regardless of discipline, if you seek to be an agricultural technology transfer specialist, there is no substitute for constantly going to the field and observing."
* "There is no such thing as 'better' or 'worse' technology, there is only
'appropriate' and 'inappropriate' technology."

Government-supported Volunteer-based Cooperation. Two volunteer programs, the Japan Overseas Cooperation Volunteers (JOCV) and the Senior Volunteer Programme (SVP), complement the larger-scale JICA programs. Both programs have placed emphasis on technical skills and experience in their recruitment of volunteers and in the manner in which volunteers work.
JOCV began in 1965. Its volunteers are young persons (age 20-39), and in this regard are similar to most U.S. Peace Corps volunteers. There is a key difference in skills between the two programs, however. The U.S. Peace Corps has traditionally used many generalists and provided them with some skills training before sending them to work in a community development mode. JOCV, however, has sought out persons with specialized technical skills and experience. Moreover, these volunteers have been drawn not only from college B.S. programs, such as in the agricultural production disciplines, but


Vol. 5, No. 1, 1995






WADA ET AL.


have also included young persons who have graduated from technical high schools and then gone back into real-world work using their skills. Skills brought to the developing counties by such volunteers have included traditional apprenticeship crafts, such as pottery (JOCV, 1993).
The Senior Volunteer Programme is a recent (1990) creation. The SVP accepts persons in the age range 40-69 as volunteers, and even more than JOCV it has sought out persons with real-world experience. In essence, SVP has provided an opportunity for mid-career or retired, skilled technicians and craftsmen, such as lathe operators and even bicycle repairmen, to share their skills in developing countries.
Both of these programs are building on the tradition of local technology development that has been a major factor in the transformation of agriculture in Japan. While these volunteers are not often directly linked to JICA projects, they do provide an important complementary mechanism to involve farmers in technology adaptation and development.
Cataloging of Indigenous Technology. As described in Section II.B, Japan's agricultural research system has a long tradition of drawing out indigenous technology and systematizing it for wider dissemination and application. This tradition has been applied in JICA's program as well, through a registry of indigenous technology developed by the Association for International Cooperation of Agriculture and Forestry (AICAF). AICAF was established in 1978 as a public service corporation under the joint supervision of the Ministry of Agriculture, Forestry, and Fisheries (MAFF) and the Ministry of Foreign Affairs. Its mandate is to collect, compile, and supply information as well as to conduct surveys on agriculture and rural development in developing countries aimed at promoting technical cooperation (AICAF, 1993). The registry was begun in 1981, and 449 technologies have now been cataloged. These technologies are listed in the publication, "Useful Farming Practices," first published in 1982 and updated annually. Eight English versions have also been issued, beginning in 1985, with the latest English version published in 1994.
The concept behind the registry is similar to a gene bank. The practical use of a particular wild species may not be evident at the time of its collection, but it is still saved because once lost through development, there is no way to regenerate it. Similarly, a wider use of an indigenous technology found in a given location may not be evident when it is encountered by an outsider, but once it is lost due to development, it cannot be regenerated if it has not been documented.
There is also a fundamental difference between the information stored in seed in a gene bank, and information about indigenous knowledge systems. If the seed of that wild variety is saved in the gene bank, even if not all the value of the information stored in its genetic code is apparent or understood at the time of its storage, this information can be held for the future, when its value may become apparent, or progress in science may make it possible to understand better all of its genetic code. However, indigenous knowledge of


Journal for Farming Systems Research-Extension






VILLAGE-BASED SELF-HELP


social arrangements (such as organization of cooperatives like the senzokabu kumiai) is part of the life of its users. Although some structures or physical objects may remain (such as the field demarcations in Nagabe resulting from the social arrangements of the senzokabu kumiai of a century and a half ago), the social arrangements themselves do not have a physical identity apart from the people who practice them (the use of the calendars introduced by Yugaku could be reconstructed today because farmers continued to use them into the 1950's). Thus, documentation of social arrangements is not as simple even as storing tools and implements or recording the physical movements of how to make or use such objects. It requires representation of the knowledge of human interactions through coding and translation into more abstract, universal terms that others removed in time and place can understand. At the same time, it is critical that this coding be not simply an apparent translation into the etic categories of researchers who use modernization criteria alone as their coding base, but that an effort be made to capture and integrate into the translation the emic criteria of the users of those social arrangements as well.
To document such indigenous technology, AICAF provides JICA specialists, JOCV volunteers, and others involved in technology generation and transfer with technology cataloging cards. When a specialist or a volunteer encounters a technology that they think is worth recording, they note its location of use, characteristics and manner of application, effectiveness, and background on one of these cards. They then send the card back to AICAF together with a photograph or drawing of the technology. AICAF staff write this up and publish it in the registry. Recently, AICAF has begun adding two additional types of technologies to the register: farmer technology in Japan that might be applicable in developing countries; and improved technology in developing countries that results from the activities of Japanese development specialists finding and improving upon indigenous technology in developing countries.
Table 2 shows that 69 percent of the 449 technologies were found in East, Southeast, and South Asia. Latin America and Sub-Saharan Africa were sources of 13 percent and 10 percent, respectively, of the technologies. This bias in regional distribution is consistent with the fact that more of Japan's agricultural development efforts have been carried out in East, Southeast, and South Asia (Table 1). More technologies were found for horticulture (29 percent) than for non-horticultural agriculture (25 percent). The larger share of horticulture may be partly due to its diversity (in addition to the diverse nature ofvegetable production, tree crops are included under "horticulture"), and to the diversity in methods of cultivation among the many types of horticultural crops. The importance of horticulture may also reflect the fact that less research and technology transfer has been done for horticulture compared to non-horticultural agriculture. The latter includes wet rice agriculture, the basis of farming systems in much of East and Southeast Asia, where most of the Japanese researchers and development practitioners have


Vol. 5, No. 1, 1995







WADA ET AL.


Table 1. Geographical and Programmatic Distribution of JICA Project-type Technical
Cooperation Programs

Region Type of Program
Agriculture Livestock Forestry Fisheries Total
E/SE/S 21 4 11 5 41
Asia
W Asia/ 1 0 0 1 2
N Africa
Sub-Saharan 5 1 2 0 8
Africa
Latin 11 2 1 4 18
America
Oceania 1 0 1 2 4
Total 39 7 15 12 73




worked. "Green Revolution" technology may have displaced more indigenous technology for wet rice agriculture, whereas there may have been less displacement for horticulture, leaving a larger pool of indigenous technology that the persons contributing to the registry have observed and drawn from. The fact that wet rice technologies make up only 14 percent of the cataloged technologies (Table 2) is consistent with this interpretation (JNCIC, 1994).
Table 3 shows that while specialists have played an important role in identifying these technologies, the role of volunteers has also been important. Specialists from JICA and TARC (since 1993, the Japan International Research Center for the Agricultural Sciences) have identified 41 percent of the technologies in the registry. At the same time, JOCV volunteers have identified 20 percent, and volunteers from Japan's largest NGO, OISCA (Organization for Industrial, Spiritual, and Cultural Advancement' have identified 10 percent of the technologies (JNCIC, 1994).
Despite the potential value of the registry, like the tradition of field orientation of JICA specialists, this effort has lacked programmatic support.

8 The word "spiritual" in the name of OISCA is a literal translation of the Japanese word seishinteki.
However, OISCA should not be therefore misinterpreted as being an organization that includes a religious agenda. The Japanese word seishintekiand the Englishword "spiritual" have fundamentally different cultural meanings. The word "spiritual" in English reflects the traditionally monotheistic basis of Western European and North American society, and refers to religious values. In Christianity, the essence of the human spirit derives from the soul, which is believed to be given by God. In contrast, in Japan the essence of the human spirit is not derived from religion, but has its own non-religious value. Hence, seishinteki does not have the religious connotation of the English word "spiritual". Seishinteki might more accurately, although more awkwardly, be


Journal for Farming Systems Research-Extension










Table 2. Indigenous Technologies Cataloged by AICAF by Region and Type of Technology, 1982-94


Type of Technology

Region Agriculturey Horticulture Livestock Forestry Fisheries Engineering' Others Total

E/SE/S Asia 81 (60) 92 34 9 7 58 31 312

W Asia/N Africa 2 (2) 4 2 0 1 5 0 14

Sub-Saharan Africa 11 (0) 14 1 1 4 10 6 47

L. America 12 (0) 17 12 4 2 5 6 58

Oceania 5 (0) 2 0 0 0 5 2 14

Not known 1 (0) 0 1 0 0 2 0 4

Total 112 (62) 129 50 14 14 85 45 449

Association for International Cooperation in Agriculture and Forestry. x Agricultural Engineering, machines, tools.
Y Number in parenthesis is that for wet rice.


*14











Table 3. Indigenous Technologies Cataloged by AICAFZ by Region and Collector, 1982-94


Affiliation of Collector

Region JICA Expert TARCy JOCV OISCAW Others Total

E/SE/S Asia 102 26 52 45 87 312

W Asia/N 12 4 26 0 19 61
Africa

L. America 30 4 7 0 17 58

Oceania 5 0 6 1 2 14

Not known 0 1 0 1 2 4

Total 149 35 91 47 127 449


Z Association for International Cooperation in Agriculture and Forestry.
Y Former Tropical Agricultural Research Center, Ministry of Agriculture, Forestry,
International Research Center for the Agricultural Sciences. X Japan Overseas Cooperation Volunteers.
w Organization for Industrial, Spiritual and Cultural Advancement.
V MAFF Staff other than TARC, other NGO than OISCA, and others.


and Fisheries (MAFF), now called Japan






VILLAGE-BASED SELF-HELP


Documentation of indigenous technology is not a part of the official job description of JICA specialists, but simply something they may do if they choose to. As a result, coverage of potentially documentable technologies is spotty, and those technologies that are documented are not always accurately and fully described.

Non-governmental Cooperation
The total number of Japanese non- governmental organizations is estimated to be over 300 (Ministry of Foreign Affairs, 1990). Among these, 186 are listed in the Directory ofNon-governmental Organizations inJapan (JNCIC, 1994). According to this Directory, 74 organizations are engaged in agricultural and/ or rural development. Most of these were established from the late 1970's to help Indochinese refugees and to provide famine relief in Africa (Table 4). Eighty percent of their activities are based on Asia. Africa is next in numbers after Asia, although its share is less than half of that of Asia. Japan's NGO activities are least in Latin America and Oceania where the need for assistance is increasing (Table 4). The overwhelming presence in Asia is based on historical, geographical, and cultural ties. The weak presence ofNGOs in Latin America despite the fact that Japan has strong ties with some countries of the region through its immigrants is largely due to concerns about political instability. There is a need to undertake more human resource development of assistance personnel, to enable the NGOs to respond to increasing numbers of requests from different countries and regions.
The financial status of Japanese NGOs has improved substantially since the government started a subsidy program for NGOs in 1989. Donations by large private companies aiming at improving the companies' public image have also increased. It appears that the government could not ignore NGOs as their social and political presence became stronger both domestically and internationally. A more pragmatic reason why the government strengthened its commitment to the NGOs may be to utilize their know-how for effective implementation of official aid.
This trend of increased government subsidization of NGO finances has two potential implications, one negative and one positive. The negative implication is the danger that the more the NGOs rely on the government financially, the less they will be able to maintain their political independence and private voluntary character. The positive implication is the possibility that the NGOs experiences at the grass-root level can contribute to making ODA more people-oriented and more effective. There would seem to be a natural complementarity between the financial capacity of ODA and human capital in the NGOs. However, while it is clear that the Japanese government has more
translated as "of the human spirit," while shnky6teki (religious) would more accurately reflect the meaning of "spiritual" as used in most English contexts. OISCA might be therefore more accurately be written in English as the Organization for the Advancement of Industry, the Human
Spirit, and Culture.


Vol. 5, No. 1, 1995







WADA ET AL.


Table 4. Number of Japan's Agriculturally-Related NGOs Active in 1993 by Year
Established and Target Region.


Year Total Target Regions
Established

Asia Africa Latin Oceania
America

Before 1950 1 1 1 0 0

1950-59 0 0 0 0 0

1960-69 3 3 3 1 1

1970-79 12 12 3 1 1

1980-89 47 35 17 8 4

1990- 11 8 4 1 0

Total 74 59 28 12 8

Percentage 100 80 38 16 11

Source: Compiled from the Directory of Non-governmental Organizations in Japan,
Japanese NGO Center for International Cooperation, 1994.



Table 5. Number and Distribution of Japan's Agricultural Related NGOs Carrying Out
Different Types of Activities



Type of Activity Number Distribution (%)

Financial aid 56 (76)

Provision of materials 38 (51)

Sending personnel 49 (66)

Acceptance of trainees 17 (23)

Development education 12 (16)

Information service 28 (38)

Total 74 (100)


Source: Compiled from the Directory of Non-governmental Organizations in Japan,
Japanese NGO Center for International Cooperation, 1994.


Journal for Farming Systems Research-Extension






VILLAGE-BASED SELF-HELP


than adequate funds available for development aid, how rich the NGOs are in both quality and quantity of human resources is questionable. Table 5 shows that even in the NGOs, human resource-related activities are relatively weak compared to financial and material assistance activities. Among their human resource-related activities, sending personnel is the most important (carried out by 66 percent of the NGOs), while only 23 percent of the NGOs engage in training young people from developing countries. To fill this large gap between financial capacity for development assistance and actual activities in human resource-building, an effort to promote and appeal to the younger generation in Japan is increasingly needed. Moreover, aid activities as a whole need the support of all sectors of society if it is to expand in the future. In this respect, development education in Japan could play an important role in creating more public awareness of conditions in developing countries and the necessity of aid. However, only 12 NGOs or 16 percent are involved in this activity.
Some NGOs have also inherited the traditions of self-help and farmer participation in development. The Organization for Industrial, Spiritual and Cultural Advancement International (OISCA), Japan's biggest NGO established in 1961, advocates hitozukurior human resource development, and can be a show case (OISCA, 1994; Kobayashi, 1992; information on OISCA's activities hereafter is taken from these two references except as indicated otherwise in explanatory notes). The first OISCA technical assistance mission arrived in India in 1966. To gain the trust of the villagers, they visited many places searching for tokund or advanced farmers respected by their neighbors. They worked to acquire those farmers' skills and then modified and improved on them little by little. Eventually they succeeded in making significant progress, obtaining a high-yield rice harvest on land which the villagers thought was barren and useless. Seeing the large potential of the land they cultivated and the techniques they practiced, the villagers realized their own latent human potential. Once OISCA gained full trust of the villagers, technology transfer became smooth.
From that time until now, OISCA has continued to apply this same philosophy. During the past nearly 30 years, it has sent a total of around 1,700 persons (technical trainers) to 13 countries and regions in Asia and Pacific.
Human resource development both in the developing countries and in the acceptance of trainees from Japan as well as to Japan is an even more important activity for OISCA. OISCA has established and operates more than 40 training centers and farms in 12 countries and regions in Asia and the Pacific, where nearly 6,000 people have been trained so far. A unique feature of these training programs is that young Japanese volunteers who want to work in development assistance are trained at the same time together with local people. By 1994 around 3,500 persons, including 400 long-term persons with the balance short-term persons, have been sent from Japan to those training centers. OISCA also accepts 200 to 300 trainees yearly from Asia and the Pacific and


Vol. 5, No. 1, 1995






WADA ET AL.


trains them at four training centers in Japan. The cumulative number of trainees to date is 4,600, of whom nearly 60 percent are in agriculture. OISCA does not necessarily attach importance to technology transfer itself, although it supports training in practical skills. Even if a person from a developing country is well trained and acquires certain skills, he or she may face difficulty when coming back to his or her home village where farming conditions arevery different from the place of training in Japan. The ability to adapt techniques to make them suitable for local conditions is more important. In addition to practical skills, OISCA teaches trainees the fundamentals of basic subjects such as mathematics, chemistry, and biology. Through this training OISCA hopes the trainees will be respected by local people and become leaders of the community (talk given by Shibako Misao, Director of OISCA's Indonesia training center, 1990).
The idea that technology itself, though important, is merely one part of the development process, and that farmers' practices are also resources with large potential, is being increasingly recognized among development assistance practitioners based on their experience. Another important effect of aid activities is the change in the orientation of OISCA volunteers themselves. Simply stated, the orientation of volunteers changes from helping to learning. Learning indigenous technology and ways of life is significant not only for development itself but also for improvement of the quality of Japanese rural (and perhaps also urban) life. In the course of rapid economic growth, Japan is losing many of the values of rural society. Returned volunteers try to feed back what they have learned in the developing countries to revitalize fading Japanese rural society. These volunteers feel that their role now is to translate in a simple way the energy and message they receive from local people in the country they have worked to common Japanese people (Saito et al., 1991).


CONCLUSIONS:
SELF-HELP AS A MODEL FOR FARMER PARTICIPATION IN AGRICULTURAL DEVELOPMENT COOPERATION
As the above comparisons have shown, Japan has traditions of self-help in agricultural development and openness of the formal agricultural research system to farmer knowledge and technology. Moreover, these traditions have converged to form programs within Japan that incorporated many of the characteristics of FSRE some 20 years before the formalization of FSRE in developing countries.
Japan's international development cooperation has benefited from these traditions at an informal level, because it has drawn its specialists from an agricultural research community with those traditions. However, at the formal, programmatic level, its international development cooperation has not incorporated in a significant way the key lesson Japan's domestic experience


Journal for Farming Systems Research-Extension






VILLAGE-BASED SELF-HELP


provides: that indigenous farmer knowledge is an essential resource for effective agricultural change. It is this gap between the informal influence of Japan's domestic traditions and the more top-down, infrastructure-technical bias of its programmatic approach that represents both a need and an opportunity.
A programmatic shift in JICA's programs could enable Japan's tradition of self-help to serve as a model for farmer participation in its agricultural development cooperation. The registry of indigenous technology could be the model for such a programmatic shift. The shift would be achieved by building four new components into JICA project-type cooperation: formalization of the assessment of indigenous technology and knowledge systems as local resources for self-help, involvement of Japanese area specialists in the social sciences, application of participatory methods developed within Japan, and establishment of a registry of JICA Practitioners and their inclusion in JICA projects.
A formal assessment of indigenous technology and local knowledge systems could be the made the first activity in the preliminary and long-term studies that precede design and implementation of JICA project-type technical cooperation. This assessment methodologically could draw from Japan's own rich tradition in participatory methods, including the KJ method, the TN method, and various muraokoshi (rural revitalization) approaches (Monma et al., 1992; Sato, 1991; Caldwell et al., 1993; Caldwell, 1994a; Hiyama et al., 1995). These efforts could build on the experience of a pilot project that has just begun in Southeast Sulawesi, Indonesia, applying an approach similar to the TN method.
The problem of language differences and the inclusion of emic categories of local knowledge systems could be addressed in two ways. First, if a JOCV volunteer was or had recently been working in the area, that person could be made a full team member in the assessment phase. Second, such teams could include an area specialist from one of the social sciences throughout their lifetime. For example, a number of young anthropologists in Japan have now acquired language skills and knowledge of culture, local agricultural practices, and the rationale behind those practices in West Africa. Area specialists with language skills and insights into indigenous agricultural practices, knowledge, and logic similar to those that anthropologists acquire can be found among the other social sciences as well, particularly in agricultural economics focusing on farmer economic behavior and village-level or sub-national regional-level economics. Both the JOCV volunteer and the anthropologist or area specialist would work with JICA technical specialists to insure real farmer input in the design of the JICA project. The anthropologist or area specialist would not have to continue as a long-term specialist, but would provide checking and feedback on the cultural congruence of the technology being developed through periodic short-term return visits to the project area during its implementation.


Vol. 5, No. 1, 1995






WADA ET AL.


In inviting anthropologists into JICA work, it would be important to respect the particular area (village or county equivalent) in which each anthropologist does his or her long-term field work, and not ask for the anthropologist to contribute in that area. The development of the long-term relationship between an anthropologist and his or her community of longterm study is much more of an art, and the relationship must be slowly nurtured without other outside influences for the anthropologist to be able to acquire valid knowledge and insight into the culture of study.
At the same time, however, collaboration in a JICA project in an area similar to an anthropologist's area of long-term study could be of value to anthropologists for their own work. It would represent more than their simply serving as resource persons. This collaboration would provide them with a companion study area in which change is actively promoted, to compare with their longstudy area. This could open up new avenues for comparative research in cultural change in anthropology, as well as help break down the tension that often exists between anthropologists and development specialists. Anthropologists would not feel that they only are brought in after it is too late to undo inappropriate development efforts, and development specialists would not feel that anthropologists simply criticize after the fact, without providing insights beforehand that could help avoid mistaken efforts.
A catalog of participatory methods could also be developed by AICAF in collaboration with JICA and MAFF. A small joint task force could be established, for example based in a relevant research institute of MAFF and the international training center of JICA located at Tsukuba, for this purpose. The task force could develop training programs both for JICA specialist teams, and for use in internships and training carried out in Japan for developing country specialists as part of JICA project-type technical assistance programs.
In carrying out JICA projects, farmers and agricultural research and extension specialists with on-farm trial experience in Japan could be included on the specialist team. Including one such person on each team where on-farm research would be identified in the project design as a necessary component would allow JICA to draw directly from Japan's domestic experience in onfarm, farmer-based research. Depending on who was available and best-suited to meet a given project need, these persons might either be members of the Senior Volunteer Programme, or be brought in a new category of JICA Practitioners.
To identify potential JICA Practitioners, AICAF could add a registry of technology development practitioners to its existing system of registry of indigenous technologies. The same task force that would develop training programs in participatory methods would work closely with the National and Prefectural Agricultural Research Stations and Extension Offices to identify such individuals.
The establishment of a JICA Practitioner system could have two additional benefits. First, it could help insure that Japan's agricultural research system


Journal for Farming Systems Research-Extension






VILLAGE-BASED SELF-HELP


retain its farmer-based traditions. As in other developed countries, the younger generation of researchers is increasingly specialized in mbre narrowlydefined scientific disciplines, and often lack field orientation. Through working together in JICA projects, JICA Practitioners would help transmit the field-based, farmer-oriented tradition of agricultural research to younger specialists who may not have had farmer-based research experience. These specialists would return from their JICA assignment not only with knowledge of their country of assignment, but also with a new perspective on the relationship between agricultural research and farmers within Japan itself.
The establishment of the JICA Practitioner system could also have important social benefits for agriculture and rural communities in Japan, as has been shown already in the experience of the Karaimo K6ryfi program in Kyushu (Yamada, 1990). Recognizing and drawing upon the experience of farmers and agricultural research and extension personnel in this way would send a strong signal to young people in rural areas that agriculture is an internationally-valued occupation. This could help reduce the negative image of agriculture and consequent lack of pride that are one of the major causes of the lack of kdkeisha (younger farmers who take over farming from their farmer parents). Increasing k6keishais critical today with the aging of the agricultural work force and rural communities in Japan.
The experience returning JICA Practitioners would bring back to their communities would also provide a new window to the outside world, giving those communities a real understanding of at least one society in a country in the developing world. This in turn could help counteract the Western developed country bias that permeates much of the popular media in Japan. Such an understanding could also help break down the psychology of insularity in Japan that has been a persisting residual effect from the Edo Era period of national seclusion, and build greater support for international cooperation.
Other developed countries could likewise apply these same ideas. The Indigenous Technology Register might be expanded to include indigenous technology in other developed and newly-industrializing countries in Europe, North America, and Asia (Taiwan, South Korea, and China). For example, horse-drawn implements still in barns and even sometimes even still used on farms in Appalachia and the Southeast United States (Caldwell et al., 1985) should be cataloged before these technologies are lost. A JICA Practitioners program could be used as a model in other similar programs, such as USAID or CIRAD collaboration in field-based research. In the United States, USAID could draw from the small-farm technicians and para-professionals of TVA, Balanced Farming, Farm-and-Home Development, 1890 extension, and other similar programs. These programs have emphasized farmer participation and represent predecessors of FSRE in the United States (Caldwell, 1985; Hagan, 1984; Johnson, 1981).


Vol. 5, No. 1, 1995







WADA ET AL.


ACKNOWLEDGEMENTS

An earlier draft of this paper was first presented as a keynote address on November 21, 1994, at the 13th AFSRE International Symposium on Systems-Oriented Research in Agriculture and Rural Development," held November 20-26, 1994, Montpelier, France.
The research reported here was supported by a grant from the Abe Fellowship Program, funded by the Center for Global Partnership (CGP), Tokyo, and administered by the Social Science Research Council (SSRC), New York. The authors wish to thank both the CGP and the SSRC for their support to make this work possible. All information and opinions presented here are exclusively the responsibility of the authors and do not represent the opinions of either the CGP or the SSRC.


REFERENCES

Association for International Cooperation in Agriculture and Forestry (AICAF) (eds).
1993. Useful farming practices. AICAF, Tokyo (in Japanese; an earlier edition is
also available in English).
Caldwell, J.S. 1994a. From ein6shiken [on-farm trials] to farmer-led muraokoshi
[village revitalization]: farming systems research in Japan. Pp. 7-33 (Japanese) and 70-93 (English), in [S. Sato (ed.)], From ein6shiken [on-farm trials] to farmer-led muraokoshi [village revitalization]. Tohoku National Experiment Station Agricultural Management and Rural Life Research Bulletin No. 13. Tohoku National Experiment Station, Morioka, Japan (English version reproduces the working paper presented in the poster session of the 12th AFSRE symposium, East Lansing, Michigan, September 13-18,1992; Japanese version modified from Caldwell, Sato,
Yagi, and Wada, 1993).
Caldwell, J.S. 1994b. Farming systems. In C.J. Arntzen (ed.), Encyclopedia ofAgricultural Science. Academic Press, San Diego, California.
Caldwell, J.S., M.H. Rojas, and A. Neilan. 1985. The difficulties in superimposing a
farming systems research and extension project on the existing cooperative extension structure in Southwest Virginia. Pp. 13-30. In C.B. Flora (ed.). Proceedings from Kansas State University's 1984 Farming Systems Research Symposium: Implementation and Monitoring. Manhattan, Kansas.
Caldwell, J.S., S. Sato, H. Yagi, and T. Wada. 1993. From ein6shiken [on-farm trials]
to farmer-led muraokoshi [village uplifting]. N6gy6 oyobi Engei [Agriculture and
Horticulture] 68(3):335-342 and 68(4):447-454 (in Japanese).
Chambers, R. 1993. Methods for analysis by farmers: the professional challenge.
Journal for Farming Systems Research-Extension 4(1):87-102.
Gilbert, E.H., D.W. Norman, and F.E. Winch. 1980. Farming systems research: a
critical appraisal. MSU Rural Development Paper No. 6, Michigan State University, East Lansing, Michigan.
Goto, J, and N. Imamura. 1993. Japanese agriculture: Characteristics, institutions,
and policies. In L. Tweeten et al., Japanese Agriculture: Tradition and Progress in
Conflict. Westview Press, Boulder, Colorado.


Journal for Farming Systems Research-Extension







VILLAGE-BASED SELF-HELP


Hagan, A. 1984. Balancedfarming in Missouri: Afarming systems approach to assisting
farm families. University of Missouri, Agricultural Economics Department Paper
No. 1984-31, Columbia, Missouri.
Hart, R.D. 1979. An ecological systems conceptual framework for agricultural
research and extension. Iowa State University CATIE-IICA Seminar on Agricultural Production Systems Research, Turrialba, Costa Rica. In Farming Systems Support Project. 1983. Readings for Farming Systems Research and Extension
Methods. University of Florida, Gainesville, Florida.
Hayami, Y., V.W. Ruttan, and H.M. Southworth (eds.). 1979. Agriculturalgrowth in
Japan, Taiwan, Korea, and the Philippines. University of Hawaii, Honolulu,
Hawaii.
Hiyama, S., S. Sato, and J.S. Caldwell. 1995. Methods and principles of farmer-led
muraokoshi (rural revitalization): results and potential applicability. Paper presented at the 13th AFSRE International Symposium on Systems-Oriented Research in Agriculture and Rural Development, November 20-26, 1994, Montpelier, France. Iijima, T. 1986. Results and appraisal of agricultural technical cooperation between
Japan and Nepal: A good example of an inter-governmental economic aid agreement. Journal of the Institutefor Asian Studies 13:55-81 (in Japanese).
Iwasaki, K. 1952. On the significance, task, and research methods for management
research. In [H. Nishigori (ed.)], Ein6 research in Tohoku. Tohoku National
Agricultural Experiment Station Report 2 (in Japanese).
Japan International Cooperation Agency (JICA). 1992. Annual Report. JICA, Tokyo. Japan Overseas Cooperation Volunteers (JOCV) (eds.) 1993. Seinen kaigai ky6ryoku
jigy6 y6ran [Essentials of JOCV activities]. JICA, Tokyo (in Japanese, English
version in preparation).
Japanese NGO Center for International Cooperation (JNCIC). 1994. Directory of
Non-governmental Organizations in Japan. JNCIC, Tokyo (in Japanese).
Johnson, G.L. 1982. Small farms in a changing world. In W. J. Sheppard (ed.).
Proceedings of Kansas State University's 1981 Farming Systems Research Symposium: Small Farms in a Changing World: Prospects for the Eighties. Paper No. 2, Kansas
State University, Manhattan, Kansas.
Kant6 T6san Agricultural Experiment Station. 1959. High level use ofmulberrygardens
to promote mulberry-dairy farming. Gumma, Japan (slides with accompanying text,
in Japanese).
Kimura, M. (ed.). 1981. Ohara Yugaku and his periphery. Yagi Shoten, Tokyo (in
Japanese).
Kobayashi, Yuichi, 1992. Zaidan H6jin OISCA Sangy6 Kaihatsu Ky6ryoku Dan [The
OISCA Foundation for Industrial Development Cooperation]. NGO KydryokuJ6hJ
[NGO Cooperation Information], No. 18:51-53 (in Japanese).
Kodama, T. 1965. Bibliography of literature and documents on eindshiken projects
(draft publication). National Agricultural Experiment Station, Ko-no-Su, Saitama,
Japan (in Japanese).
Kuwahara, M. (ed.). 1974. The theoretical basis of agricultural cooperative movements.
Ie no Hikari Ky6kai, Tokyo (in Japanese).
McDowell, R.E. and P.E. Hildrebrand. 1980. Integrated crop and animalproduction:
making the most of resources available to small farms in developing countries.
Working papers. The Rockefeller Foundation, New York.


Vol. 5, No. 1, 1995







WADA ET AL.


Ministry of Foreign Affairs (MFA). 1990. Annual report of Japan's Official Development Assistance 1990. MFA, Tokyo (in Japanese).
Monma, T., S. Yasunaka, and S. Asai. 1992. Structure and function of strategic rural
revitalization support system (TN Methods). Tohoku Rural Planning Report
(T.R.P.) 9:1-25 (in Japanese).
Mori, H. 1956. Ways to advance rainfed agricultural systems. Chiky6i Zensho, Tokyo
(in Japanese).
Nakamura, K. 1956. Thegoalsofimproved rainfed agriculture. N6sangyoson Bunkaky6kai, Tokyo (slides with accompanying text, in Japanese).
Nishigori, H. 1952. History of the Sanbongihara farming station. In [H. Nishigori
(ed.)], Ein6 research in Tohoku. Tohoku National Agricultural Experiment Station
Report 2 (in Japanese).
Norman, D.W. 1983. The farming systems approach to research. In C. B. Flora (ed.)
and W. J. Sheppard (compiler), Proceedings of Kansas State University's 1982 Farming Systems Research Symposium: Farming Systems in the Field. Paper No. 5,
Kansas State University, Manhattan, Kansas.
Ogura, T. (ed.). 1968. Agricultural development in Japan. Fuji Publishing Co.,
Tokyo.
OISCA. 1994. OISCA katsud6 an'nai (summary of OISCA's activities). OISCA,
Tokyo (in Japanese).
Ruttan, V.W. 1975. Integrated rural development programs: A skeptical perspective.
International Development Review 1975/4:9-16.
Ruttan, V.W. 1977. The green revolution: Seven generalizations. International
Development Review 4:9-16.
Ruthenberg, H. 1980. Farming systems in the tropics, 3rd ed. Clarendon Press, Oxford,
Great Britain.
Sasaki, T. 1988. The sources of rainfed farming methods in Kenpoku [the northern part
of Iwate prefecture]: Slash-and-burn (araki) and botta ball planting. Ethnography
in Iwate 9:22-43 (in Japanese).
Saito, N., H. Kanda, T. Hayashi, J. Sugino, and K. Nakagami. 1991. The role and
necessity of NGOs in agricultural development in developing countries. NGO
Kylry6kuJ6h6 16:2-18. (in Japanese).
Saito, Y. 1970. Study of the development of modern agricultural science in Japan.
National Research Institute of Agricultural Economics, Tokyo (in Japanese).
Sato, S. 1991. A multifaceted approach to the restructuring of regional agriculture.
Paper presented at the summer seminar on agricultural management in the Tohoku
Region, July 31-August 1, 1991, Fukushima, Japan (in Japanese).
Shaner, W.W., P.F. Philipp, and W.R. Schmehl. 1982. Farming systems research and
development: Guidelines for developing countries. Westview Press, Boulder, Colorado.
Schultz, T.W. 1964. Transforming traditional agriculture. Yale University, New
Haven and London.
Society for the Study of Agricultural Development History (ed.). 1953. The history of
agricultural development in Japan, vol. I. Chf6 K6ron Sha, Tokyo (in Japanese). Suzuki, F., T. Tanaka, and I. Akimoto. 1958. Studies on the processes and effects of
introducing the improved techniques offarming to farms. Tohoku National Agricultural Experiment Station Report 13:160-181 (in Japanese).


Journal for Farming Systems Research-Extension







VILLAGE-BASED SELF-HELP


Suzuki, F. 1986. On the farming systems approach. International Cooperation of
Agriculture and Forestry 8(3):2-9 (in Japanese).
Takeda, T. 1960. Historical resources on Ch6son Ze and village structure in the Meiji
Era. S6ken Gepp6 [Monthly Bulletin of the National Research Institute of Agricultural Economics] 47-56 (in Japanese).
Tanaka, M. 1977. The lineage of thoughts on the promotion of regional agriculture.
Ngy6 Keizai Kenky4 [Journal of Rural Economics] 49(2):51-57 (in Japanese). Yamada, M. 1990. Villages under an open system: the Maeda report as viewed from
muraokoshi in Southern Kyushu. In M. Nakamura (ed.), Change and Structure in
Regional Economies. Taga Publishing, [Tokyo].


Vol. 5, No. 1, 1995












Toward the Concerted Management
of Agro-Ecosystems'

A. Huijsman2



THE NEED FOR LOCAL-LEVEL CONCERTED ACTION AND EMPOWERMENT
In many areas of the world, agro-ecosystems are under strong pressure; the quality of natural resources is rapidly declining due to overexploitation. The rate of deterioration ofarable lands increases alarmingly. These processes point to the idea that in many agricultural production areas time is running out: immediate action is required. Given these circumstances, the inclination is to legitimize quick fixed, top-down operations. The compelling argument is that the longer environmental degradation persists, the higher its remedial costs, both in social and economic terms.
However, in many of those countries facing environmental degradation, we are confronted with the following situation. On the one hand, we see governments trying to change people's behavior, reflected in policy objectives and intentions, but lacking means, capacity, and administrative structures to effectively design and implement strategies to do so. On the other hand, we see individual land users facing a high level of legal, organizational, and economic uncertainty, which paralyzes their decision-making and action. There is a growing recognition that such situations may lead to inertia: environmental degradation will be considered as everybody's problem but nobody's responsibility.
The central tenet of the papers' contribution to this session is that sustainable land use critically depends on the way management of natural resources is organized, shaped, and institutionalized at the local level. The premise here is that rural people have both a right and an obligation to participate actively in their own development, and in the control and management of their environment. Entrusting responsibilities and accountability to local-level institutions-to the direct users-is considered a prerequisite for successful environmental management and natural resources use.
The papers3 deal with institutional and management issues--the focus is not on the "what" but rather on the "how" of natural resource management, and

1 Session paper presented at the Thirteenth Annual Association for Farming Systems ResearchExtension Symposium, Montpellier, France, November 21-25, 1994.
2 Director Agriculture and Enterprise Development, Royal Tropical Institute (KIT), Mauritskade 63, 1092 AD Amsterdam, The Netherlands.


Vol. 5, No. 1, 1995






HUIJSMAN


process rather than impact. The papers present both conceptual and methodological discussions and specific case studies. In total, nine papers were selected by the Symposium organizers for this session. Most of the case studies deal with group action of farmers at the community level. Although reference is made to the importance of linking local-level action to higher systems levels, except for Mali, no concrete examples are given. This will limit the field of actors that will be referred to in the discussion.
Agro-ecosystem management deals with the organizing and decisionmaking related to the multiple and competing uses of natural resources. These uses include crop and livestock production, forestry, gathering of wild products, nature and gene reserves. The need foragro-ecosystem management arises from the fact that decisions related to the utilization of natural resources, whether it concerns individual fields or freely accessible common resources (pastures, woodlands), can not be solely left to individuals.
Attention to natural resources management brings to the foreground a number of factors that until recently have received little attention from systems researchers and development practitioners. As highlighted by the various papers, key features of dealing with agro-ecosystem management include interdependencies, conflicting situations, power relations, and decision-making mechanisms.

Inter-Dependency
Increasing resource scarcities articulate interdependencies between natural resources users in both space and time. The importance of managing the spatial dimension of natural resources is felt more and more as space becomes more restricted. The impact of decisions made by one individual or groups of individuals on others ('externalities') is becoming more pronounced. The time dimension addresses the interdependency of resource users over time at a given locality. It draws attention to the intertemporal effects of resource utilization, the fact that the livelihood of future generations depends on the rate and nature of resource exploitation of present generations. Conventional FSR approaches hardly touch upon these issues.

Conflict Resolution
The increasing scarcity of natural resources (cultivable land, pastures, forestry products) is giving rise to spatial conflicts between user groups within and between rural communities, as well as between rural communities, urban centers and the State. Addressing these conflicts requires taking into account 3 Given the geographical and -consequently- the institutional setting of the papers, four are
based on experiences from Sub-Saharan Africa, three on experiences in the United States, Australia and -to a lesser extent- the Netherlands, and one is based on India. The discussion will focus on those situations where environmental degradation has become a threat to the livelihood ofpeople and concerted action with respect to natural resource managementis most
urgently needed.


Journal for Farming Systems Research-Extension






CONCERTED MANAGEMENT


the composition of local communities and the different groups of outside users: permanent and temporary residents; landowners, tenants and landless people; men, women and children. It is essential to acknowledge that each category has its own background, perceptions, and interests. This makes joint decision making related to natural resources use an inherently political and complicated process; unique, optimal solutions do not exist. Processes of conflict resolution, negotiated settlement, and agreement are of crucial importance (Van Beek, 1991). Yet few experiences exist as how to manage these processes.

Power Structures
Dominant groups may attempt to control decision-making processes or existing community management structures may frustrate the participation of the village community as a whole. Often specific attempts have to be made to stimulate participation of different groups, as has been done for example in Mali (Diarra et al., 1994). In many rural communities, the role of women and young men as users of natural resources and their possible contributions to improved agro-ecosystem management is not sufficiently taken into account. But, as Rbling argues, increased interdependencies in natural resources use implies that also socially less powerful actors have a potential capacity to affect the outcomes of decision making processes. Knowledge about power structures is important to assess the impact of intervention programs and outcomes of change processes.

Empowering Local Institutions
Social and political acceptance among rural land users of the idea that environmental management requires concerted decision making is needed for defining rules and regulations and joint action to implement them. In different papers it is argued that strong local institutions are not only necessary to see to it that individuals abide by the rules that have been agreed, restrict the chances of free riding, and solve conflicts when they arise. They also have an important role to play in linking the community to higher agro-ecosystems levels (e.g., watersheds) as well as in providing claim making power to higher institutional levels for technical assistance and financial support (Laban, 1994).


SOFT SYSTEMS, COLLABORATIVE LEARNING, AND ACTION
To address these issues, various papers indicate the need to improve upon existing system concepts, research methodologies, and intervention approaches. Joldersma calls for an action-oriented research methodology for socioeconomic and institutional experimentation (e.g. Joldersma et al., 1994). Ruling argues for the combination of hard system thinking and soft systems approach-


Vol. 5, No. 1, 1995






HUIJSMAN


es, with the former focusing on the biophysical aspects of agro-ecosystems and the later dealing with social actor networks, interdependencies, conflicts and power structures. Campbell (1994) and William (1994) suggest new models for interactive learning, inquiry and action. It is argued that present systems approaches are still strongly dominated by the technical and economic perspectives, and lack an effective complementary operational social and institutional perspective (e.g., Rling, 1994).
Bringing together these concepts and methods, the following components appear to be central to a people-centered approach to agro-ecosystem management:
" an actor-oriented approach
" supported by participatory action research
" based on systems thinking

Actor-Oriented Approach
The essence of actor-oriented approaches is to make explicit the different life-worlds, interests and strategies of the various actors in a given situation, including the network of institutional actors (Campbell, 1994a). As a diagnostic methodology it is aimed at understanding the dynamics between actors and the conditions determining their decisions. It provides insights into social processes governing natural resources use. Social actor analysis, however, is not aimed for action to improve situations. Ruling suggests making the approach operational by introducing the notion of "human platforms" as a way to bring actors together facing interlinked problems. Creating human platforms focuses on a gradual process of establishing mechanisms for mediation and conflict resolution between interest groups, accommodation between multiple perspectives, and on reaching joint agency, i.e., the ability to act as a group and to decide upon joint action to improve a problem situation. Such an approach clearly deviates from conventional FSRE programs that consider farmers as clients and end-users of technologies rather than as interacting stakeholders competing for resources.

Participatory Action Research
The principles of participatory research are well established, although often much less applied in practice. The increasing complexity and location specificity of the issues involved, however, underscores the need to further develop robust participatory methodologies focusing on action. In line with the actor orientation, there is increased attention for empowering participation stimulating actors-to participate in decision making. As argued by Campbell, the basic idea behind empowering or decision making participation is that it extends ownership of both process and results to natural resources users and thus promotes sustainability. Other papers deal extensively with new approaches to interactive learning and action research. These may help groups to


Journal for Farming Systems Research-Extension






CONCERTED MANAGEMENT


explore common interests, and may lead to consensual solutions in conflicting situations. William provides an overview of participatory approaches that are available for addressing complex problems through collaborative learning and joint action.

Systems Thinking
Systems thinking provides the basis for making visible and discussing interdependencies in natural resource use. Various papers mention system tools and techniques that have been developed to assist groups in revealing patterns and relationships and exploring insight into complex problems: the GRAAP methodology applied in West Africa, participatory rural appraisals using relational diagrams and land use mapping by villagers, and rapid appraisal of agricultural knowledge system. William and Campbell mention some new techniques carrying such esoteric names as Farmer-Scientists Focus Sessions, learning or behavioral inventories, and mind maps and matrices. This seems to be a fruitful area for exchange of experiences between the North and the South, although one obviously should keep in mind the social and cultural contexts within which tools have been developed and applied. Various papers indicate that hard-systems oriented research has a powerful contribution to make in supporting group learning by assessing together with those concerned the use of natural resources and assessing systems dynamics to depict future trends and effects of different scenarios.


PUTTING INTO PRACTICE
The underlying principles in applying actor-oriented approaches supported by participatory action research include building on existing knowledge, local wisdom and experiences; going ahead only with the support of the population; and implementing only programs that fit the rhythms of the community and are institutionally viable. Such a process approach implies dealing with problems step by step, breaking down large, complex problems into smaller units amenable to solution. These approaches go typically through the following steps (William, 1994):
* Situational/problem analysis of 'what is' aimed at sharing views, perceptions, and interests and determining common ground;
* Brainstorming on 'what ought to be' searching for common interests in improving natural resources management, and developing goals that are acceptable to everyone;
e Strategic thinking on 'how to get there' exploring different courses of action and examining the consequences of each alternative.
The Mali paper by Diarra provides an example of such process. It describes how participatory land use planning may be applied as an iterative process, integrating new elements in a plan through learning by doing, combining new


Vol. 5, No. 1, 1995






HUIJSMAN


ideas with local practical experience. It is stressed that land-use plans are the end result of a process. If land use plans are prepared by outsiders to be discussed with the village population, the process might well end right from the start. Land-use planning is a continuous, flexible process; it must gradually evolve and be adjusted over time to meet changing circumstances.
There is general agreement that outside agents play an important role in facilitating the learning process as well as linking different village (interest) groups and the local community, and the community to outside sources of information and assistance. However, putting facilitation into practice is difficult. It requires skilled listening and asking questions, analytical capacities, a sensitivity to group processes and a capacity of synthesis, handling conflicts and fostering synergy, while knowing when to lead, when to wait, when to challenge, and when to withdraw (Campbell, 1994). This requires an understanding of communication processes within communities. Until now, these aspects receive little attention in training programs.
It is much less clear from the papers how and at what level of system aggregation learning processes and agency development could start and should be supported. This obviously depends on the institutional setting and the type of issues at stake. For instance, in Mali three different approaches are used to address natural resource management issues ranging in focus from land improvement, management of natural resources at village level, to embedding natural resources management in a comprehensive approach to community development. In assessing the effectiveness of approaches, it seems indicated to carefully consider the balance between the current trend to more complex, multicomponent interdisciplinary approaches and the need to keep approaches comprehensible and applicable for practitioners to make headway. The underlying dilemma is that if the approach is too specific, it loses the holistic context that is necessary to develop comprehensive strategies for natural resource management. On the other hand, if too broad, the approach becomes synonymous with the concept of development itself and may lack an operational and pragmatic orientation of what can and should be done in agroecosystem management.


THE NEED FOR HARD SYSTEMS RESEARCH AND
TECHNOLOGY DEVELOPMENT
It is clear that research into institutional and organizational aspects is vital to sustainable natural resources use. However, as various papers quite correctly indicate, bringing these to the foreground does not take away from the need to develop technologies and hard system research. The long-term challenge here is to reverse environmental degradation by shifting from curative to preventive measures. In many agricultural production areas, this implies somehow curtailing the 'extensification process,' as the expansion of cultivat-


Journal for Farming Systems Research-Extension






CONCERTED MANAGEMENT


ed areas is increasingly encountering ecological, technical, economic, and social problems. Hence, there is an urgent need to find ways to intensify agricultural production in an ecologically and economically sustainable way. Measures to rehabilitate the environment should be combined with sustainable productivity improvement.
In the search for better technologies, attention should focus on the whole of range of options, from low to high external input techniques, from agroforestry techniques to biotechnology. Existing biases toward high use of external inputs and short-run productivity improvement must make way for strategies based on long-run nutrient cycling processes, maximum use of biophysical potential, and integrated land use management systems. What counts is creative thinking because more sustainable farming systems are inherently more knowledge-intensive in nature posing a challenge to the different actors involved: natural resource users, commodity and locationspecific systems researchers, extension agents. It must be better realized that different situations require different types of inquiry, based on different assumptions and requiring different type of skills. Working towards sustainable agriculture demands not only new theoretical and practical methods of its professionals, but a new sense of what it means to be a professional in agriculture (Campbell, 1994). Simulation studies such as presented by Weber aimed at modelling man-resource interactions under different context conditions and management strategies of resources users may help to understand system dynamics and testing the implications of different scenarios.

REQUIRED CHANGES IN THE MACRO ENVIRONMENT
Strengthening joint decision-making and action at the local level supported by better technologies for sustainable agricultural productivity improvements is a necessary but not sufficient condition. The present focus on participatory approaches, local initiative, and decentralized decision making should not be taken to the other extreme, in which all responsibility is shifted to local organizations. In many areas in the world, efforts to stimulate better use of natural resources through participatory action will be in vain if conditions in the legal and economic environment will not change (Laban, 1994).
First, various papers mention that the lack of a coherent framework of adequate rules and regulations related to natural use at the national and regional level may seriously hamper local level initiative in resolving disputes and taking decisive action (Joldersma et al., 1994; Laban, 1994). Successful efforts to introduce more sustainable land use practices-through participatory approaches-on a larger scale (Indonesia, Australia, Netherlands) are all nested in a well defined regulatory context (Rbling). Among other factors, in many countries existing land legislation and customary law need to be reviewed, simplified, and conflicts between these systems need to be resolved.


Vol. 5, No. 1, 1995






HUIJSMAN


Such formal legislation could find in many cases inspiration in local customary right systems. At the same time, ways and means have to be found to adapt customary systems to the changed environmental, socioeconomic, and political conditions of today (Laban, 1994). Active involvement of government agencies in experiments with natural resources management, such as in Mali, may facilitate this process.
Market failure or poor functioning markets is a second major cause of environmental degradation. Open access resources allow land users to externalize environmental costs, shifting these to the community and future generations, especially when alternatives for a more sustainable use are not economically attractive. The growing natural resource scarcity should be reflected in increasing resource prices. This would stimulate efforts to reduce the growth of demand and improve the efficiency of land utilization. Market distortions due to government policies and monopoly situations render input and output prices such that farmers' choices are far from optimal from an environmental point of view.
The third, and probably most severe constraint, is the limited investment capacity of rural households in marginal areas. The conflict between income and environment is a stark reality for the fast growing number of resource-poor rural land users searching for a livelihood. They have limited access to financial opportunities and other means of production, and poor alternative employment opportunities. Their room to manoeuvre is extremely limited. Farmers often have no other option but to overexploit their land, or to migrate to more productive areas, reinforcing the process of environmental degradation in those areas.
In such situations, specific incentives and special funding programs will be necessary to bring about a change in behavior, stimulating people to participate in environmental (investment) activities. This applies specifically to the often expensive, onetime investments in environmental rehabilitation measures. Recurrent maintenance costs should in principle be considered part of the costs of exploiting natural resources; ways need to be found to include these costs as production costs. In assessing the benefits of financing incentives, funding agencies should take into account the opportunity costs of nonintervention such as increased deforestation and desertification (climate), migration to urban centres and international migration as well as social unrest and warlike conflicts. The likely response of different types of farm-households, and the differences between agricultural zones, have to be carefully assessed if incentive programs are to succeed. The combination of systems thinking, participatory action research and a social-actor approach may contribute considerably to the identification of promising incentives, combining the micro and macro perspective.


Journal for Farming Systems Research-Extension






CONCERTED MANAGEMENT


ISSUES
The strategy advocated in the various papers requires resisting the urgent need to combat environmental degradation directly through top-down coercive intervention programs, and instead to act first by investing time in people, as well as their organizations. Strategies and programs are needed that gradually provide the rural population with the means, decision-making power, and authority that will enable them to take the management of natural resource use into their own hands, and to do this with confidence.
As demonstrated by the papers, there is a growing body of experience on working with rural communities on environmental management. It is also clear that much work remains to be done in the methodological field of combining hard- and soft-system research and developing robust participatory action approaches. A number of outstanding issues include:

'Facilitation expertise'
Group learning approaches may initially create more problems than they solve. In predominantly agricultural societies, dealing with natural resources management touches upon the very fabric of social life. Latent conflicts may come to the surface and new conflicts may arise. Comprehensive solutions may not be found immediately. Outside agents facilitating such social processes must have the appropriate skills and professional background. Amateurish tinkering with social engineering processes may have lasting negative effects on rural communities. The question is where will these facilitators come from and who will train them. Nongovernmental development organizations are seen as important contributors. However, in the short term their role should not be exaggerated, as these organizations are themselves often still in a process of defining their roles and building up their own capabilities.

Institutional cultures
Supporting institutions often have institutional cultures, which do not favor participatory and interdisciplinary approaches. For example, top-down, supply-driven T&V extension systems are still dominant in many countries. An important question is how to align these systems to the sort of participatory learning and action approaches advocated in this session. Although a platform approach may offer learning possibilities to overcome these constraints, one has to be cautious that platforms do not become restricted to supporting institutions only, excluding the institutional participation of natural resource users as appears to be the case in Mali. Concertation and collaboration may become an end in itself or an excuse for not starting concrete actions.


Vol. 5, No. 1, 1995






HUIJSMAN


Vertical integration
Local-level environmental management institutions must be embedded in a multi-tier management system, in which national, regional, and local government, as well as nongovernmental organizations, play their respective roles (Laban, 1994). Finding the right balance between central government regulation and local autonomy is probably one of the most difficult challenges in establishing environmental management systems. Critical questions include what type of organizations are needed at different levels and, more importantly, what type of mandates these organizations will have in terms of decision making, implementation, and monitoring. A related question is how to bring together the approach of participatory agency development as advocated by Ruling with the current efforts taking place in many countries of decentralizing political authority to provincial and local governments.

External triggers
Similar with FSR programs in the past, there is a risk that, without changes in the macro-environment, participatory approaches aimed at natural resources management do not get past the diagnostic stage. It is relatively easy to initially mobilize interest, but concrete options for improvement in terms of knowledge, profitable technologies and resources must be available to keep the momentum going and to institutionalize the learning-by-doing process. Moreover, looking at examples from Australia, Europe, and Northern America, it is my firm belief that specific incentive measures and funding mechanisms will be required as an external trigger to motivate land users to work together and invest in environmental measures. Based on micro-level information and opinions of natural users, propositions should be made to policy makers about the most promising incentives and other measures.

Scaling up
If the sort of people-centered approach advocated is to be successful in combatting environmental degradation, due attention must be given to the process of scaling-up and institutionalization of activities from the outset taking into account the resource implications of broader application in terms of funding, human resources, and institutional support. A very critical question is whether the rate of environmental degradation does not outrun the speed with which and the scale at which participatory approaches may have an impact on natural resources management.

Time horizon
Participatory learning and action processes are incremental and necessitates dealing with problems step by step. Such processes take time to mature; for example land consolidation processes in the Netherlands often took over 25


Journal for Farming Systems Research-Extension







CONCERTED MANAGEMENT


years to be completed. Time horizons of support programs should be in line with these realities.

International context
Finally, a debate about the state of the environment necessarily includes the developmental process at large, and consequently, not only policy making of local and national governments but also the relations between the South and the industrialized North. Environmental problems are inherently global and political, demanding redistribution of resources (Redclift, 1984; Blaikie, 1985).


SYMPOSIUM PAPERS

Bousquet, F., Antona, M. et J. Weber. 1994. Simulations multi-agents et gestion des
ressources renouvelables.
Campbell, A. 1994a. Participatory Inquiry. Beyond Research and Extension in the
Sustainability Era.
Campbell, A. 1994. Landcare in Australia. Spawning new models of inquiry and
learning for sustainability.
Diarra, S., Hilhorst, T. et N. Coulibaly. 1994. Vers une gestion durable des ressources
naturelles: Quelques experiences avec des programmes tests Gestion du Terroir
Villageois au Mali-Sud.
Joldersma, R. et.al. 1994. L'int~gration de la recherche-syst~me et la gestion des
ressources naturelles: un marriage de bon sens.
Laban, P. 1994. Accountability, an indispensable condition for sustainable natural
resource management.
Ruling, N. 1994. Creating Human Platforms to Manage Natural Resources: First
Results of a Research Program.
Singh, K.P. 1994. Farmers Involvement in Improving Farming Systems in an Area of
Resource-poor, Rainfed Agriculture in Tribal Region of Bihar, India.
William, R.D. et.al. 1994. Improving Oregon's Natural Resources: Collaborative
learning, systems approaches, and participatory action research.


OTHER REFERENCES

Blaikie, P.M. 1985. The political economy of soil erosion in developing countries.
London: Longman.
Redclift, M. 1984. Development and the Environmental Crises: Red or green alternatives? London: Methuen.
Van Beek, P. 1991. Using a workshop to create a rich picture: defusing the ponded
pastures conflict in central Queensland. Paper for the Workshop on Managing Complex Issues in Uncertain Environments: Systems Methodologies in Agriculture. Brisbane: Queensland Dpt of Primary Industries, 26-29 August.


Vol. 5, No. 1, 1995












Management of Natural Resources:

Systems Approaches to Striga Control in
Sub-Saharan Africa

Helmut Albert' and Artur Runge-Metzger3



ABSTRACT

Striga hermonthica is a parasitic weed that is still spreading throughout sub-Saharan Africa. To date, many control methods have been developed that differ with respect to their demand for land, labor, capital, and management capabilities. Some have been applied successfully in parts of Africa. However, in most situations, limited adoption of Striga control methods have not prevented further
dispersion of the so-called 'witch weed.'
In order to develop a promising extension strategy, scientific and indigenous knowledge must be integrated. First, as government budgets are restricted, extension activities must target areas where Striga control will provide the best returns. Therefore, a monitoring concept has to be developed. Second, Striga control methods must be screened for technical and economic feasibility. Third, control methods must be evaluated from the farm household perspective. Fourth, farmers' perceptions about the biology of Striga must be considered.
The occurrence of Striga has to be regarded as a symptom of deteriorating soil fertility. The continuing spread of Striga appears to show that past investments made by individual households into the maintenance of soil fertility were highly inadequate. Clearly, individual decisions lead to undesirable outcomes in the long term. From the societal point of view, individual decisions lead to external effects that have a negative impact on the income possibilities of future generations. It is therefore necessary to discuss whether policy interventions are justified and which type of intervention is operational and feasible
under the conditions of sub-Saharan Africa.
The results of a recent study in northern Ghana are presented. These show how participatory interviews, formal household surveys, regional surveys, and secondary data sources may be integrated to improve
natural resource management.
I Paper presenteed at the Thirteenth Annual Association for Farming Systems ResearchExtension Symposium, Montpellier, France, November 21-25, 1994.
2 Nyankpala Agricultural Experiment Station, POB 483, Tamale, Ghana.
3 Institute of Agricultural Economics, Georg-August-University, Platz der Gbttinger Sieben 5,
37073 G6ttingcn, Germany.
Vol. 5, No. 1, 1995






ALBERT AND RUNGE-METZGER


INTRODUCTION
Sustainable management of natural resources has become the primary objective of rural development. In the past, different schools of thought covering various scientific disciplines have contributed toward improvement of the understanding of various aspects of natural resource management, notably institutional economics, resource economics, human ecology, and farming systems research. But how can these different aspects be integrated in a practical methodological approach to achieve sustainable development? A systems approach is discussed, taking as example Striga hermonthica.
Striga hermonthica is a parasitic weed that is spreading throughout subSaharan Africa. In the past, several control methods were developed. These differ in their requirements for land, labor, capital, and management capabilities (Robson and Broad 1989; Gworgwor 1993; Sauerborn 1991). Some of the methods have been applied successfully in parts of Africa. However, in most areas, farmers have not adopted recommended Striga control methods to levels required to prevent further dispersion of the so-called 'witch weed.'
This paper aims to identify the reasons underlying the technology transfer gap, and to determine appropriate interventions to alleviate the identified constraints using a systems approach.


HYPOTHESES AND METHODOLOGY
Constraints may occur at different levels of the system, notably:
* Government policy may favor agricultural practices that result in the propagation of Striga;
* Research does not address farmers' constraints adequately, leading to unacceptable recommendations;
* A weak performance by extension services restricts dissemination of effective Striga control measures to farmers;
* Farmers reject the control measures because: they are unfamiliar, they do not fit into their farm household system, they are not regarded to be effective, Striga is not perceived to be a major problem. In addition, agricultural practices facilitate the propagation of Striga.
Systems analysis therefore appears to be an appropriate methodological approach to identifying the constraints. The analysis comprises a set of five well defined steps to identify:
(a) actors in the system and their goals;
(b) resources (including natural, financial, perceptual, and knowledge) available to the different actors;
(c) behavior and activities of the different actors and their impacts on the performance of other actors and on the status of resources;
(d) constraints within the system; and
(e) potential future interventions.


Journal for Farming Systems Research-Extension






STRIGA CONTROL


Steps (a) and (b) are largely descriptive. At each step different empirical methods are used. Step (c) is the most cumbersome because, ideally, functional relationships have to be established between the actors, activities, and performances (i.e., state and flow variables dependent upon time). It also includes assessment of known interventions, such as new agricultural technologies. Appropriate, mainly standard methods must be employed to assess possible impacts of interventions depending on the scope of intervention. Basically, this must be done in a participatory manner, with detailed explanations of interventions to farmers being backed by extensive discussion.
This approach helps identify specific constraints of adoption. Moreover, the results of farm interviews also provide the basis for more rigorous approaches to economic analysis. The impacts of minor changes are modelled using partial budgeting. If inputs and outputs are affected over several years, investment analysis has to be employed. More drastic changes in the agricultural system could be simulated at household level on the basis of comparative static farm household models. When new crops are introduced or substantial yield increments are anticipated, aggregated effects beyond the farm household level must also be considered. Hence, market analysis is necessary. Synthesis of the results of the preceding analytical steps allows explicit identification of constraints in step (d). Step (e) concerns possible future interventions.
The general aim of this study was not to be innovative in terms of creating a new sophisticated econometric model, but to demonstrate how various standard methodological modules can be integrated in a systems approach to participatory problem solving using the case of Striga control in a West African setting.


DATA BASIS
The study was based on the results of different empirical field studies. Detailed secondary information was available from earlier surveys:
* Village level surveys conducted between 1984 and 1992 (Baur 1992; Runge-Metzger 1993a);
* Striga questionnaire for extension agents (Sprich 1993);
* Striga surveys (Sauerborn 1991b; Vogel 1993; pers. comm. Sprich, Schellinger, Universitdt Hohenheim, 1992; Honisch 1989);
e Official sources (Food and Agriculture Organization 1992; PPMED 1993; Statistical Services 1989).
In addition, two empirical surveys were conducted in 1993-an in-depth questionnaire was circulated in August/September 1993 to 69 farmers in three different areas of northern Ghana (Runge-Metzger, 1993b), and a short questionnaire targeting 4218 farm plots was completed in October/November 1993.


Vol. 5, No. 1, 1995






ALBERT AND RUNGE-METZGER


ACTORS AND RESOURCES

Table 1 shows the different actors of the system-government, research, extension services, and the farmer-who indirectly and directly influence the population dynamics of Striga.

Table 1. Actors of the system, their goals and resources. Actors Goals Resources
Government Income (reelected), Budget, manpower, know-how
equity, economic growth,
balanced budget
Scientists Income, respect Budget, manpower, know-how
NARS'
Cotton research
Regional network
MoFAb
Extension services Income Budget, manpower, know-how
MoFA
Private (cotton)
NGOs'
Farmers Income, secured satis- Natural resources (land,
Resource rich faction of basic needs soil fertility, rainfall, knowResource poor (shelter, food prefer- how, capital)
ences)
Striga hermonthica Propagation Weak cereal crops
a. NAIRS, national agricultural rcsearch systems.
b. MoFA, Ministry of Food and Agriculture, Ghana.
c. NGO, nongovernment, nonprofit organization.

For government, scientists, and extension services, predominantly financial resources are converted into staffing and logistics. But also very important is the quality aspect of the staffing, its know-how or human capital. The government of Ghana relies for its activities on a limited budget that has become even more constrained. This inevitably affects availability of research and extension resources. For scientists, most funds are channelled through the government. A limited number of researchers are employed privately by cotton companies. A special budget for Striga research in northern Ghana-funded by a foreign donor-strengthens the available manpower.
Extension is conducted by government, private (cotton), and nongovernment, nonprofit organizations (NGOs). Government service has been reorganized considerably in recent years, notably with introduction of the World Bank favored T&V system. The number of extension workers has been drastically reduced. Almost all former technical assistants have been made redundant, reducing substantially the ratio of extension workers to farmers. In addition, government extension workers do not always have the appropriate


Journal for Farming Systems Research-Extension






STRIGA CONTROL


know-how for dealing with Striga. Nongovernment organizations-predominantly church agricultural projects-can be found in several communities scattered throughout northern Ghana.
Besides labor and capital, which can be converted to inputs and hired labor, farmers exploit natural resources including soil fertility, available water, and solar radiation. Earlier studies of farming systems revealed that farmers in the region differed greatly in terms of access to these resources, major factors being population density and market access (Baur 1992; Runge-Metzger 1993a). As differences in relative resource availability determine to some extent applicability and profitability of innovations, farm households are stratified according to the population density and access to capital (Runge-Metzger 1993a; Smith, Weber 1993). The capital assets of farm households are predominantly held in livestock and this plays an essential role in risk management.
Information about farmers' indigenous knowledge and perceptions regarding the biology of Striga was gathered during the in-depth field survey of August/September 1993. In general, farmers regard Striga as a severe problem that has increased over past years. However, their perceptions about biological aspects do not necessarily match scientific knowledge. For instance, many farmers are unaware of the exact mechanisms whereby Striga seeds are further propagated, while a considerable number wrongly associate the symptoms of other diseases (e.g., wilting of yam and groundnuts) with Striga. Farmers' perceptions of the impact of various farming practices on Striga were also examined (Figure 1). In particular, the interrelationship between Striga and soil fertility is well-known.
Striga relies on the presence of cereals, which provide carbohydrates and mineral nutrients for its growth. Cereal seeds provide the main vehicle for the spread of Striga seeds (Berner et al. 1993).
An important feature of water, one of the major natural resources on which all crop activities and Striga depend, is that its availability is highly variable from year to year and across locations within the same year.


PERFORMANCES OF ACTIVITIES AND IMPACTS ON THE OCCURRENCE OF STRIGA
The goals and resources of the actors of the system were presented in a topdown manner; activities and associated performances will be presented bottom-up.
Parasitism of cereal plants is an activity essential to the multiplication of Striga hermonthica. The performance of Striga is highly variable as it depends on the state of various fluctuating natural conditions. Another important factor is the nutritional status of the host plant (Patterson 1987; Gworgwor 1993; Vogt 1993). When the cereal plants are nourished well, the damaging effects of Striga are reduced, and conditions become unfavorable for seed


Vol. 5, No. 1, 1995







ALBERT AND RUNGE-METZGER


Figure 1. Farmers' perceptions of the impact of agronomic practices
on the incidence of Striga, Ghana (1993).





*~i1 ]


00 0


, i.mil


[LI 1111111 I I











>' 0
E-~ '~
z
a
~
~








00 '~ 00 00



>. 0 I- ~

n
o ~


~ a! ~
C
C C C U
C


Journal for Farming Systems Research-Extension