A Framework for Incorporating Indigenous Knowledge Systems into Agricultural Research, Extension and NGOs for Sustainable Agricultural Development

1. Introduction

1.1 What is indigenous knowledge?

Indigenous knowledge is local knowledge that is unique to a given culture or society (Warren, 1987). Indigenous knowledge is the systematic body of knowledge acquired by local people through the accumulation of experiences, informal experiments, and intimate understanding of the environment in a given culture (Rajasekaran, 1993). According to Haverkort (1991), indigenous knowledge is the actual knowledge of a given population that reflects the experiences based on traditions and includes more recent experiences with modern technologies. Local people, including farmers, landless laborers, women, rural artisans, and cattle rearers, are the custodians of indigenous knowledge systems. Moreover, these people are well informed about their own situations, their resources, what works and doesn't work, and how one change impacts other parts of their system (Butler and Waud, 1990).

1.2 Value of indigenous knowledge

Indigenous knowledge is dynamic, changing through indigenous mechanisms of creativity and innovativeness as well as through contact with other local and international knowledge systems (Warren, 1991). These knowledge systems may appear simple to outsiders but they represent mechanisms to ensure minimal livelihoods for local people. Indigenous knowledge systems often are elaborate, and they are adapted to local cultural and environmental conditions (Warren, 1987). Indigenous knowledge systems are tuned to the needs of local people and the quality and quantity of available resources (Pretty and Sandbrook, 1991). They pertain to various cultural norms, social roles, or physical conditions. Their efficiency lies in the capacity to adapt to changing circumstances. According to Norgaard (1984, p. 7):

Traditional knowledge has been viewed as part of a romantic past, as the major obstacle to development, as a necessary starting point, and as a critical component of a cultural alternative to modernization. Only very rarely, however, is traditional knowledge treated as knowledge per se in the mainstream of the agricultural and development and environmental management literature, as knowledge that contributes to our understanding of agricultural production and the maintenance and use of environmental systems.

1.3 Diversity of indigenous knowledge

Indigenous knowledge systems are:

• adaptive skills of local people usually derived from many years of experience, that have often been communicated through "oral traditions" and learned through family members over generations (Thrupp, 1989).

• time-tested agricultural and natural resource management practices, which pave the way for sustainable agriculture (Venkatratnam, 1990).

• strategies and techniques developed by local people to cope with the changes in the socio-cultural and environmental conditions.

• practices that are accumulated by farmers due to constant experimentation and innovation.

• trial-and-error problem-solving approaches by groups of people with an objective to meet the challenges they face in their local environments (Roling and Engel, 1988).

• decision-making skills of local people that draw upon the resources they have at hand.

1.4 Building on indigenous knowledge

The above statements clearly illustrate that indigenous knowledge systems are invaluable, diversified, and comprehensive, although this is not always the perception among outsiders (Thurston, 1992). In fact, they are often overlooked by western scientific research and development because of their oral tradition (Warren, 1990). Hence, by facilitating these systems, outsiders can understand better the basis for decision-making within a given society. Further more, by comparing and contrasting indigenous knowledge systems with the scientific technologies of International Agricultural Research and Development Centers (IARDCs) and regional research stations, it is possible to see where technologies can be utilized to improve upon local systems (Warren, 1987).

Policy actions, especially in the 1990s, should give attention to actively preserving this diversity of knowledge. This can be done by documenting, incorporating, and disseminating indigenous knowledge, and by creating awareness and supporting projects among local populations. Hence, indigenous knowledge is a key to successful participation of local people in agricultural and rural development programs.

Section two discusses constraints in conventional transfer of technology paradigm. Section three deals with methods of facilitating the utilization of indigenous knowledge systems by agricultural researchers and extensionists. Section four and five explains methods to incorporate indigenous knowledge systems into agricultural research and extension programs. Strategies to strengthen indigenous organizations by utilizing non-governmental organizations are provided in section six. Section seven discusses the steps involved in technology dissemination.

2. Constraints in Conventional Transfer of Technology Paradigm

Agricultural research for the most part has been highly reductionist, parochial, and discipline-oriented (Richards, 1989). Normal science generates packages, whereas resource-poor families engage in farming as a continuous performance. Research station technologies have focused primarily on attaining high yield of target crops. The introduction of high energy technologies through the application of chemical fertilizers, agrochemicals, machinery, and modern methods of irrigation in developing countries was a departure from traditional agriculture and has led to pollution and land degradation (Ezaza, 1989).

In addition, lack of relevance to small farm conditions was found to be one of several constraints in the station research technologies. Kerr and Sanghi (1992) provided a specific example from Andhra Pradesh, India, to support the above statement. The conventional graded bunding system is not an appropriate soil and moisture conservation technology under small-scale dryland farming conditions due to the following reasons:

1. Continuous bunds leave corners in some fields thus creating the risk of losing the piece of land to the neighboring farmer;

2. Contour farming causes inconvenience in field operations (particularly where multi-row implements are used) and reduces the efficiency of operations (where the desi plough is used) due to repeated cultivation in the same direction;

3. Systems based on a central water course provide benefit to some farmers at the cost of others with regard to disposal of excess runoff; and

4. The overall system emphasizes only long-term gains, hence creating an impression that short-term gains are not possible through such measures (Kerr and Sanghi, 1992, p.2).

2.2 Neglecting local classification systems

Agricultural researchers and extensionists usually are not aware of local classification systems of farmers regarding soils, crops, livestock, and other natural resources. A case study conducted by the International Crops Research Institute for Semi-Arid Tropics (ICRISAT) in Shirapur, a South Indian village, showed that the indigenous soil categories of farmers were more accurate than the formal system in stratifying the soils into groups for analysis and provided improved bases for indexing variations in land quality (Dvorak, 1988). In addition, indigenous soil types are considered better for long term sustainability of the soil structure and soil fertility (Warren, 1992d). Because soil analysts in soil testing laboratories (STLs) are not familiar with the indigenous classification, their fertilizer recommendations may not fit in with the local soil categories.

A case study conducted by Rajasekaran indicates that the indigenous classification of rice varieties in Chengalpattu District, Tamil Nadu State, is based on criteria such as water source, cropping season, crop duration, and grain quality (Rajasekaran and Warren, 1993). The village extension workers disseminate information on the seed varieties recommended by the researchers to the farmers. These extension decisions are reflected in the types of seeds made available through the seed multiplication units. Although several varieties suitable to semi-arid zones of Tamil Nadu are adapted to severe drought conditions, most of the varieties being encouraged through the agricultural extension system are suitable only in resource-rich environments such as those with an assured supply of irrigation. The indigenous, locally adapted varieties of rice are no longer as easily available (Rajasekaran, Warren and Babu, 1991).

2.3 Underperceiving farmers' experiments

During the process of technology development, farmers' informal experimentation are not considered as a source of innovation. In spite of increased coordination between research and extension through periodical extension-scientific workers' conferences, it is found that farmers' innovations are not considered while conducting on-farm research trials (Rajasekaran and Martin, 1990). On-farm trials conducted by researchers and extensionists mostly concentrate on crop varietal comparison, fertilizer response, and testing of different packages of practices for cereals and millets. In contrast, farmers experiment on alternative coping strategies to avoid extreme conditions such as droughts and floods, diversified food production techniques such as intercropping and border cropping in order to broaden their food and fodder requirements, and adjusting their sowing and harvesting periods to meet the local market demand.

2.4 `Technical messages' syndrome

Farmers are mainly seen as the recipients of technical messages but not the originators of either technical knowledge or improved practice (Moris, 1991). The technical messages concentrate mostly on seed-to-seed packages of practices for different crops grown in the region. Resource conservation strategies such as watershed management, agroforestry, and soil conservation rarely form part of the technical messages. Technological recommendations based on the findings of research stations, though initially followed by contact farmers were not well received by other groups of farmers (non-contact farmers). Farmers who were active during the initial stages of implementation of the Training and Visit (T&V) extension system became bored of the stale technical messages. In general, the nature of the technical messages can be grouped into three categories based on the nature of constraints in them (Rajasekaran and Martin, 1990):

(1) Repeated nature of technical messages: Most of the technical messages were developed entirely based on research conducted at regional research stations in India. These messages concentrate mainly on seed-to-seed package of practices. Dissemination of these crop production technical recommendations was a matter of gaining new knowledge and skills in the beginning of the T&V implementation. Once the message has been repeated season after season, farmers not only became bored but also tended to play an inactive role in the entire system.

(2) Technical messages that do not reflect local crop production conditions: Some technical messages do not reflect local crop production conditions. For instance, line planting has been recommended as one of the technical messages for rice production under wet land conditions in Tamil Nadu state, India. Though planting of rice seedlings in lines certainly increases per unit production of rice when compared to random planting, the cost of labor incurred towards line planting is significantly higher than that of the latter method.

(3) Blanket technical messages: Some technical messages tend to be blanket recommendations which are evolved from the research and cannot be adapted to heterogeneous farming conditions. For instance, the regional research stations recommend only blanket recommendations for fertilizers such as urea, di-ammonium phosphate and muriate of potash whereas nitrogen, phosphorous, and potassium content of soil varies widely from village to village, in many cases from plot to plot.

2.5 Attitudes of outsiders

Attitudes generated by the top-down transfer of technology (TOT) paradigm have precluded learning indigenous knowledge of farmers. Reasons for non-adoption of innovations resulting from the conventional TOT paradigm have been attributed to outsider's stereotypes of small-scale farmers (e.g., ignorance, laziness, conservatism) or an inadequate delivery system (e.g., poor extension service, lack of credit facilities) but seldom to the characteristics of the innovations themselves (Waters-Bayer, 1987). Chambers (1990, p. 3) stated that:

As we enter the 1990s, the dominant paradigm of development expressed by normal professionals and implemented through normal bureaucracy is still top-down and center-outwards. Power is concentrated in hands of the old men in high offices and central places. Knowledge is generated in universities, laboratories, and research stations, and then transferred packaged for adoption. The approach is centralized, standardized, and simple. Reductionist research, high input packages, and top-down extension had their successes: in the uniform and controlled conditions of industrial agriculture. But the sustainability of that increase is open to question, and TOT does not work well with the more complex, diverse and risk-prone rain-fed agriculture of much of the poorer South.

Certain inherent limitations in indigenous knowledge systems have strengthened the attitudes of outsiders that indigenous knowledge systems are `primitive', `unproductive' and `irrelevant': (1) indigenous knowledge systems are of oral in nature; (2) indigenous knowledge systems are not formally recorded and documented; (3) Each individual possesses only a part of the community's indigenous knowledge systems; (4) indigenous knowledge systems may be implicit within local people's practices, actions, and reactions, rather than a conscious resource (Reijntjes et al., 1992); and (5) Finally, farmers' rarely recall information on quantitative data pertaining to their indigenous knowledge systems (Rajasekaran, 1993).

2.6 Feedback from farmers: a missing link

After technology dissemination, feedback from farmers regarding the characteristics of the introduced technologies are rarely recorded. Development of technologies in research stations has become a continuous process without judging what is happening in the field. Following are the factors contributing to this problem (Rajasekaran, 1993):

1. Agricultural researchers do not investigate the impact of the technologies they develop. They feel their responsibility ends once the technologies are released to the extension system;

2. Agricultural extension personnel perceive that dissemination of technologies to farmers is their only responsibility. Once the technologies are disseminated to the farmers, they are completely satisfied with their jobs; and

3. Even some enthusiastic extension workers who have tried to bring feedback from the farmers are not encouraged either by extension administrators or researchers.

2.7 Consequences of disregarding indigenous knowledge systems

Undermining farmers' confidence in their traditional knowledge can lead them to become increasingly dependent on outside expertise (Richards, 1985; Warren, 1990). Small-scale farmers are often portrayed as backward, obstinately conservative, resistant to change, lacking innovative ability, and even lazy (IFAP, 1990, p. 24). The International Federation of Agricultural Producers (IFAP) enumerated certain reasons for such a perception:

1. Lack of understanding of traditional agriculture which further leads to a communication gap between promoters and practitioners giving rise to myths;

2. The accomplishments of farmers often are not recognized, because they are not recorded in writing or made known; and

3. Poor involvement of farmers and their organizations in integrating, consolidating, and disseminating what is already known.

Loss and non-utilization of indigenous knowledge [which] results in the inefficient allocation of resources and manpower to inappropriate planning strategies which have done little to alleviate rural poverty. With little contact with rural people, planning experts and state functionaries have attempted to implement programs which do not meet the goals of rural people, or affect the structures and processes that perpetuate rural poverty. Human and natural resources in rural areas have remained inefficiently used or not used at all. There is little congruence between planning objectives and realities facing the rural people. Planners think they know what is good for these `poor', `backward', `ignorant', and `primitive' people.

2.8 Need for a conceptual framework

Despite continuous importance given to linkages between research-extension-farmer while developing, disseminating, and utilizing sustainable agricultural technologies, several socio-political and institutional factors act as constraints for such an effective linkage (Oritz et al., 1991). After a decade of rhetoric about feedback of farmers' problems to extension workers and scientists, a large gap remains between the ideal and reality (Haugerud and Collinson, 1991). Kaimowitz (1992: 105) provided illustrations to support the above statement:

Researchers perceived extension agents and institutions to be ineffective and unclear about their mandate, making researchers reluctant to work with extension. When researchers did work with extension agents, they tended to look down on them and view them as little more than available menial labor, an attitude strongly resented by the extension workers.

Keeping these potential constraints in conventional transfer of technology, a framework for incorporating indigenous knowledge systems into agricultural research and extension has been developed with the following salient features:

1. strengthening the capacities of regional research and extension organizations;

2. building upon local people's knowledge that are acquired through various processes such as farmer-to-farmer communication, and farmer experimentation;

3. identifying the need for extension scientist/ social scientist in an interdisciplinary regional research team;

4. formation of a sustainable technology development consortium to bring farmers, researchers, NGOs, and extension workers together well ahead of the process of technology development;

5. generating technological options rather than fixed technical packages (Chambers et al., 1989);

6. working with the existing organization and management of research and public sector extension;

7. bringing research-extension-farmer together at all stages is practically difficult considering the existing bureaucracies and spatial as well as academic distances among the personnel belonging to these organizations. Hence, utilizing the academic knowledge gained by some extension personnel (subject matter specialists) during the process of validating farmer experiments;

8. outlining areas that research and extension organizations need to concentrate on during the process of working with farmers.

9. understanding that it is impractical to depend entirely on research stations for innovations considering the inadequate human resource capacity of the regional research system.

Chambers and Jiggins (1987, p.5) supported the need for such a framework:

1. The transfer of technology (TOT) model fits badly with the needs and priorities of resource-poor farmers.

2. Agricultural extension programs are still biased towards techniques and strategies which are capital-intensive.

3. Resource-poor farmers (RPF) are scattered and are not able to make their needs and priorities readily known and felt.

4. The TOT model cannot easily handle the complex interactions of RPF farming; links between crops, especially with intercropping and multiple tiers; agro-forestry and livestock-crop-tree complementaries; and the progressive adjustments required in the field in the face of seasonal and inter-annual fluctuations.

3. Facilitating the Use of Indigenous knowledge Systems

3.1 Establishing indigenous knowledge resource centers

Establishing a national indigenous knowledge resource center forms the starting point for the entire framework of incorporating indigenous knowledge systems into agricultural research and extension (Warren, 1992b). The resource persons in the national indigenous knowledge systems resource center will provide training on the methodologies for recording indigenous knowledge systems. The concept of establishing national resource centers was developed by Professor Michael Warren, Director of the Center for Indigenous knowledge for Agriculture and Rural Development (CIKARD). He has pioneered the establishment of 11 national indigenous knowledge resource centers so far in Nigeria, Mexico, Philippines, Indonesia, Ghana, Kenya, Sri Lanka, the Netherlands, Brazil, Burkina Faso, and Germany. The functions of national indigenous knowledge systems resource centers include (Warren, 1992b):

1. Provide a national data management function where published and unpublished information on indigenous knowledge are systematically documented for use by development practitioners;

2. Design training materials on the methodologies for recording indigenous knowledge systems for use in national training institutes and universities;

3. Establish a link between the rural people of a country who are the originators of indigenous knowledge and the development community;

4. Facilitate the active participation of rural people in the conservation, utilization, and dissemination of their specialized knowledge through in situ knowledge banks, involvement in research and development activities, farmer-to-farmer training, and farmer consultancies; and

5. Act as a two-way conduit between the indigenous knowledge-based informal research and development systems and formal research.

3.2 Bringing a desirable change in the attitudes of outsiders

Training programs on indigenous knowledge systems are inevitable for bringing a desirable change in the attitudes of researchers and extensionists. The need for conducting training programs for extension workers on the role of indigenous knowledge in agricultural development has been expressed by Waters-Bayer and Farrington (1990, p.12):

1. If the extension personnel including village extension workers and agricultural extension officers are provided training on scientific technological innovations, but have not learned to regard farmers as their colleagues, their potential to support farmers' local research efforts will be comparatively lower;

2. Training programs on the role of indigenous knowledge in agricultural development help to remove the impression among the extension workers that research scientists are the only generators of technological innovations and their (extension workers) job is to merely transmit those innovations;

3. Information provided in these training programs regarding local farmer organizations and their functions can stimulate ideas among extension workers for a number of viable action-programs; and

4. Extension workers can help local farmers' organizations establish and strengthen links with agencies such as government services, private organizations, commercial firms, and other farmer organizations for information and other inputs.

Training on indigenous knowledge systems should be conducted in two stages. Initially, the resource persons of the national indigenous knowledge systems resource center will organize training of trainers workshops. Extension trainers of regional extension training institutes and extension education institutes of agricultural universities from various regions of the country form the target audiences for these workshops. As a second stage, regional extension trainers are expected to provide similar training programs for district-level subject matter specialists. In parallel, extension educators of extension education institutes of the agricultural universities should conduct training programs for research scientists on the methodologies for recording indigenous knowledge systems.

3.4 Supplementing training programs

A training manual for introducing the methodologies for identifying and recording indigenous knowledge systems is essential for supplementing the training programs. The training manual should be based on the peasant forms of communication which are related to rural, everyday life, which has its own seasonal and life rhythms (Salas and Tillman, 1989; Frio, 1991). Monthly zonal workshops and bi-weekly training programs can be used as forums for conducting the training programs under the extension setting. Separate in-service training programs should be organized for research scientists either at the state-level or regional level. Kerr and Sanghi (1992) emphasized that training on indigenous knowledge systems has to be preceded by a change in attitudes and behavior towards the farmers. The process of attitude change has to start from the top, from teachers in universities to policy makers/ implementers in government.

The need for a training manual to present the methodologies to record indigenous knowledge systems is emphasized by Warren and Rajasekaran (1991, p.1):

Though the value of Indigenous knowledge systems in facilitating development and extension is gradually being recognized by national and international development agencies, the concepts, principles, and methodologies for recording and utilizing these systems are not yet familiar to many professionals working in agricultural and rural development. Many extension and training programs are still focused exclusively on scientific and technological developments generated through formal on-station research. This manual is designed to help agricultural extension and training programs to experience ways in which Indigenous knowledge systems can facilitate understanding and communications between farmers and extension workers leading to participatory approaches to agricultural development.

4. Technology Development by Incorporating Indigenous knowledge Systems into Agricultural Research

4.1 Need for an inter-disciplinary approach

The regional research stations are responsible for developing agricultural technologies related to disciplines such as plant breeding, agronomy, entomology, soil science, and plant pathology. There are approximately twenty agricultural scientists working in Krishi Vidyan Kendra (KVK), the regional research station for the Union Territory of Pondicherry. There is no social scientist or extension scientist working in this station. The proposed framework recommends that an extension scientist should be recruited to work with KVK. He/she is expected to play a key role in linking the research mandates with those based on farmers' perceptions. In the farmer-back-to-farmer model, Rhoades and Booth (1982) provided a specific case to show how incorporating social scientists in an interdisciplinary research team would bring farmers' perceived needs and problems into the research agenda.

4.2 Identifying problems

Problem identification forms the first step during the process of developing sustainable agricultural technologies. Problems are biological as well as socio-cultural limiting factors or inefficiencies in the use of resources that restrict the productivity or sustainability of a farming system (Tripp and Woolley, 1989). Problems should be identified jointly by biological scientists, and social scientists in consultation with farmers. During this stage, farmers' perceptions regarding needs and priorities should be taken into account. Farmers should be viewed as co-researchers, developers, and extensionists who can provide crucial inputs to determine what problems to address and how to proceed (Chambers et al., 1989).

Working with various groups of local people separately is important while identifying problems since each group of local people perceive the same problem differently. For instance, women laborers in Tamil Nadu, India, perceived that transplanting rice using 2-3 seedlings is a time-consuming as well as laborious process, hence they prefer to use more than 5 seedlings for transplanting. On the other hand, farmers perceived that transplanting rice using more than 5 seedlings reduces rice yield significantly.

The social scientist in coordination with disciplinary scientists, should define the identified problems in clear terms. The definition of problems requires a good understanding of the farming system, an appreciation of farmers' resources, perceptions, and priorities, and a continual dialogue between farmers and researchers (Tripp and Woolley, 1989). Rhoades and Booth (1982) have also provided examples for problem definition.

4.3 Recording relevant indigenous knowledge systems

Indigenous knowledge is the systematic body of knowledge acquired by local people through the accumulation of experiences, informal experiments, and intimate understanding of the environment in a given culture. Indigenous knowledge systems are dynamic, changing through indigenous mechanisms of creativity and innovativeness as well as through contact with other local and international knowledge systems (Warren, 1990). In the process of technology development, knowledge of indigenous livelihoods is an indispensable resource (Haverkort and Zeeuw, 1992). Indigenous knowledge may not be as abstract as scientific knowledge. It is often concrete and always dynamic. It relies strongly on intuition, directly perceivable evidence, and an accumulation of historical experiences (Farrington and Martin, 1987). Indigenous knowledge reflects the dignity of the local community and puts its members on an equal footing with the outsiders involved in the process of technology development (Haverkort and Zeeuw, 1989). Indigenous knowledge systems also provide mechanisms for facilitating understanding and communications between outsiders (extensionists, researchers) and insiders (farmers). Improved understanding and communications enhance participatory approaches to problem identification (Warren, 1992b).

Once problems are identified, the next step during the process of developing sustainable agricultural technologies is to record the indigenous knowledge systems of farmers which contribute to the solution of the problem. In other words, how do farmers try to overcome or adapt the problems using their own knowledge? For instance, informal exchange of rice seeds from farmer-to-farmer is used as a strategy by farmers to solve the growing demands of quality rice seeds in Tamil Nadu, India. The social scientist in the regional research station in coordination with respective disciplinary scientists should record indigenous knowledge systems.

Methods to record indigenous knowledge systems are imperative in order to create a rethinking in the attitudes of researchers and extensionists. Though an awareness of indigenous knowledge systems is rapidly increasing, lack of methods creates a vacuum in this area. Well-defined methods would help outsiders to make use of the available resources--time, money, and human---effeciently. Use of clear-cut methods would also reduce the bias on the part of the outsiders during the process of documenting indigenous knowledge systems. Moreover, this would also take into consideration the sociocultural variabilities that are prevalent in the rural areas.

Two of the selected methods to record indigenous knowledge systems are discussed in the following:

Participant observations: Participant observations are effective in identifying and recording various indigenous technical practices of farmers. Jorgensen (1989, p. 82) provided the salient features of participant observation as a method to document the insiders' world:

Participant observations begin the moment the participant observer makes contact with a potential field setting. Aside from collecting information, the basic goal of these largely unfocused initial observations is to become increasingly familiar with the insiders' world so as to refine and focus subsequent observation and data collection. It is extremely important that you record these observations as immediately as possible and with the greatest possible detail because never again will you experience the setting as so utterly unfamiliar.

The following step-wise procedures provide an idea of how to conduct participant observations to document indigenous technical practices (ITPs) of farmers. The procedures recommended by Jorgensen (1989) and Colfer et al. (1988) were modified with an objective to document indigenous technical practices:

1. The researcher/extension worker should walk through farm holdings of the study villages and select those holdings where farmers adopt indigenous technical practice/s;

2. After entering the field (i.e., farm holdings), the researcher/extensionist should look for the agro-ecological features of the farm holdings. Certain specific questions should be kept in mind during observation. A few examples are: What are the crops grown in various agro-ecological environments of the study villages? Is it a monocropped or intercropped area? What are the sources of irrigation? What is the size of the farm holding? What are the primary soil types of the farm holding?

3. As a second step of the participant observation stage, the researcher/extensionist should look at the role of the farmers. How are the farmers classified? What kind of division of labor exists? What are the roles of men and women laborers? What are the tools and implements used by them? This step would help the researcher/ extensionist to get acquainted with the participant farmers;

4. After becoming familiar with the agro-ecological and human settings of the selected farm holdings, the researcher/extensionist should begin observing the matters of specific interest, i.e., ITPs. This process of observing the matters of specific interest has been referred by Jorgensen (1989) as `more focused observations';

5. The following procedures should be adopted while observing and documenting ITPs: (1) Observing ITPs: ITPs adopted by farmers and farm laborers in their respective farms can be observed; (2) Documenting ITPs: The observed ITPs can be documented using a camera; (3) Analyzing ITPs: The salient features of ITPs can be recorded in a pocket notebook by carefully observing, and listening to the conversations between laborers and farmers.; and (4) Titling ITPs: Later on, an appropriate title for each of the ITPs recorded can be identified through informal discussion either with the participant farmers or with the laborers who are encountered in the farm holdings;

6. By adopting the above procedure systematically and patiently, ITPs related to the following areas of food production and resource conservation can be documented: (1) cropping systems, (2) seeds and sowing, (3) seed processing, (4) soil health care management, (5) planting techniques, (6) crop nutrient management systems, (7) weed management techniques, (8) plant protection strategies, and (9) post-harvesting procedures.

2. Unstructured Interaction

The purpose of unstructured interactions is to elucidate relevant information pertaining to ITPs that are documented during the participant observation stage: (1) farmers' beliefs, values, and customs related to the ITPs, and (2) the process of decision-making while selecting the ITPs. This interaction would provide an in-depth understanding of the `emic' perspectives of local farmers. The `emic' perspective involves putting oneself as much as possible into the farmers' shoes to understand how they view their practices in both technical and socio-cultural terms (Rhoades and Booth, 1982).

The success of unstructured interactions lies in the careful involvement of key informants. Key informants are those local people who are willing to talk or be interviewed intensively regarding the matter of specific interest (Jorgensen, 1989). Selection of key informants may be done by a few preliminary discussions with the following people: (1) the local extension agent, (2) local school headmasters, (3) credit cooperative society officials, (4) village milk cooperative society members, (5) farmers, and (6) men and women laborers. The following criteria can be used to select the key informants: (1) good knowledge about the historical background of food production and resource conservation of study villages; (2) a minimum of ten years of farming experience; and (3) not being involved in other stages of the study.

Table 1a, b provides an overview of selected methods for recording indigenous knowledge systems with specific examples.

4.4 Forming a sustainable technology development consortium

The purpose of a sustainable technology development consortium is to bring farmers, researchers, extensionists, and NGO representatives together in order to classify the identified problems and indigenous knowledge systems and set agendas based on them (Figure 2). Kaimowitz (1992) emphasized the need for such a common forum to link researchers and extensionists. He also enumerated certain constraints in the existing research-extension linkages:

1. Researchers may have their research plans already established and may not really be open to inputs from extension;

2. Personnel representing the extension agencies may be regional or national officials who have little direct knowledge of conditions in the country side;

3. The objectives and agenda may not be clear and their mandate may be too broad to be feasibly addressed in the time allocated;

4. Even when extension workers' perceptions are accurate, researchers may perceive them to be uninformed or subjective; and

5. Researchers may be reluctant to accept them because of extension's lower status and researchers' low esteem for the extension agent's abilities.

Forming a technology development consortium is an attempt to overcome these potential constraints with its salient features such as: (1) giving equal footing to problems and indigenous knowledge systems as recorded by researchers and extensionists; (2) bringing divisional-level subject matter specialists (SMSs) who are familiar with both local conditions and extension headquarters; and (3) respecting extension workers' intimate contact with farmers.

In the consortium, research should be represented by all scientists of the regional research station, extension should be represented by regional-level extension administrators and SMSs and NGOs by their representatives. Amanor and Farrington (1992) stated that complementary linkages between NGOs, research, and extension encourage interaction among many sources of technical innovation to arrive at dynamic technological options. The framework assumes that it takes no more than two days to classify problems and to decide on the agenda for each organization. The specific objectives of the consortium are to:

1. discuss all problems and indigenous knowledge systems as perceived by local people;

2. prioritize problems and indigenous knowledge systems with active participation from farmers; and

3. decide who should work on what problem area.

   Problems and indigenous knowledge systems that need research station facilities and advanced academic training should go to researchers. The subject matter specialists of the extension organization should take care of problems and indigenous knowledge systems that do not require any on-station support. NGOs can concentrate on problems and indigenous knowledge systems related to strengthening and empowering indigenous organizations or local networks.

4.5 Developing a research agenda

Understanding farmers' problems and indigenous knowledge systems allows a framework for posing technical, scientific questions in research and also provides the basis for evolving technologies that are not imposed as alien `packages' that contradict existing practices (Scoones, 1989). The capacity of farmers using their indigenous knowledge systems to classify and evaluate the research technologies can complement station-based development of agricultural technologies (Raman and Balaguru, 1990). Once problems are identified and classified, the next step is to set research priorities. Conventionally, research priorities are determined by policy makers and researchers with little or no farmer participation (Doorman, 1991). This framework suggests a change in the normal tendency. Once certain problems are described in farmers' terms, they have to be 'translated' to match the definitions and concepts used by researchers. This indicates that sufficient knowledge of both the farmers' problems and related indigenous knowledge systems is necessary to translate the defined problems into research priorities. In setting research priorities, the essential starting point is for social scientists and disciplinary scientists to give close attention to the farmers' own detailed knowledge of existing practices (Haugerud and Collinson, 1991). Each area of research should indicate farmers' problems, relevant indigenous knowledge systems and proposed solutions by researchers.

4.6 Conducting participatory on-station research

It is not sufficient if farmers are involved only during problem identification and recording of indigenous knowledge systems. Participatory research is a two-way flow that both takes scientists to farmers' fields and brings farmers to research stations (Haugerud and Collinson, 1991). Hence, involving "research minded" farmers while conducting on-station research is essential and at the same time challenging. Since farmers and scientists each know and understand many things, but have little overlap between their domains of knowledge, farmer-scientist interaction should help both groups learn. Since both professions are constantly experimenting, more interaction should improve each other's experiments (Bentley and Andrews, 1991; Bentley and Melara, 1991).

Scientists have a wealth of knowledge concerning biological factors related to food production whereas small-scale farmers have a wealth of knowledge concerning the management of ecological, technological, and organizational factors related to food production under specific local conditions (Fernandez and Salvatierra, 1989). For instance, incorporating farmers into on-station germplasm screening can produce useful information at little cost (Haugerud and Collinson, 1991). Before conducting on-station research on cultivar selection, plant breeders should bring village-level seed producers (farmers) to the research station and listen to their criteria of varietal selection. For instance, one village-level seed producer indicated that coarse grain rice varieties never lodge during the earhead stage under irrigated conditions

Conducting on-station research can be divided into two sub- components: (1) Developing new research station technologies based on indigenous knowledge systems:

Prain (1992) found that farmers evaluated cultivars using a wide variety of criteria that can be of immense interest and value to crop breeders. In Zambia, the farmers evaluation of a high-yielding hybrid maize variety and description of the positive and negative characteristics of locally-adapted open-pollinated varieties led to a more effective national maize breeding program (Warren, 1989). Hence, during the process of technology development, scientists at the research station should conduct research by building on the acquired indigenous knowledge systems.

Developing new varieties of food crops by restoring the traits of local landraces is one of the examples of this process. For instance, a local variety of chili crop used by farmers in Kizhur, Pondicherry, India, is well adapted to agro-ecological and environmental conditions. Moreover, it is resistant to certain pests and diseases of chili. While introducing a variety from outside in order to obtain higher yields, farmers experienced new pest and disease problems. In this case, research station scientists can solve the farmers' problems by developing a new genotype by integrating the traits of local varieties (adapted to environmental stress and resistant to the fruit borer) and varieties from the research station. The new cultivars thus developed can be evaluated for their local adaptability using the procedures discussed in the latter stages.

(2) Integrating indigenous knowledge systems and existing research station technologies:

In some cases, research can be conducted by matching the indigenous knowledge systems and existing research station technologies. For instance, casuarina farmers in Pillayarkuppam, Pondicherry, India, conducted informal experiments by growing legumes such as blackgram or cowpea as intercrops in casuarina (a multipurpose tree) fields. But most of them faced problems such as the shattering of legume pods and spreading of legumes between casuarina trees. The research station scientists can conduct on-station research experiments with an objective to evaluate the performance of various legume varieties in casuarina fields and select certain legume varieties which are suitable for intercropping in casuarina fields.

The successful combinations of casuarina and legume varieties can be taken to farmer-oriented on-farm research for its validation under farmers' field conditions. Other examples where an integrated technology can be developed by blending Indigenous knowledge systems and existing research station technologies are developing IPM strategies by blending indigenous crop pest management systems and selected chemical pest control methods, and conducting integrated crop nutrient research to formulate crop nutrient schedules by mixing cattle, sheep manures and chemical fertilizers.

4.7 Conducting on-farm farmer-oriented research (OFFOR)

Participatory on-station research formed the base-line for conducting on-farm farmer oriented research (OFFOR) (Figure 2). The purpose of OFFOR is to validate the findings of the participatory on-station research. The primary role of the researchers is to match technological options that are developed from on-station research to selected farming conditions, and to provide leadership in designing the research (Baker, 1990). The disciplinary scientists should conduct the OFFOR in coordination with extension scientists.

The following are the salient features of OFFOR proposed in the model: (1) OFFOR keeps the indigenous knowledge system of farmers as a base; (2) It facilitates a rigorous farmer participation well ahead of the on-farm research process in order to generate a basket of technological options; (3) It can be taken to wider areas among a wider spectrum of farmers covering different castes and gender with minimum cost (Chand and Gurung, 1991); (4) It has enabled researchers to get direct and firsthand feedback that helps researchers to improve or modify technologies.

Research scientists must present the integrated technological options developed during participatory on-station research (POSR) stage for consideration of selected farmers. The selected farmers are encouraged to identify technological options that would fit into their individual problems and resource constraints. For example, farmers with soil alkalinity problems might select a soil reclamation trial. Marginal farmers who rear cattle as their off-farm occupation are very knowledgeable of fodder trees that can be grown on field bunds and hence might select a trial on fodder tree evaluation. Marginal farmers who own bullocks to plow other farmers' fields for labor possess a bountiful knowledge on indigenous soil classification and the fertility status of that location or village and hence may prefer to choose a fertilizer trial.

Instead of selecting experimental plots, the OFFOR utilizes the entire farm for OFFOR research. By selecting experimental plots, we are narrowing the focus to a particular crop (mainly cereals and millets) in the farm, while neglecting the value of associated crops, trees, and livestock. For instance, farmers in the study villages grow legumes such as black gram and green gram in rice bunds. In some other cases, marginal farmers harvest crabs from the burrows of field bunds. Hence, selecting the entire farm for OFFOR is important. Such an effort would facilitate not only an in-depth understanding of the interactions among crops-trees-livestock but also their role in sustainable food production and resource conservation.

On-farm research should incorporate farmers' own methods of informal experimentation, their standards of judgement, and their suggestions concerning experimental design (Haugerud and Collinson, 1991). Baker (1990) provided certain essential guidelines that need to be given consideration while conducting OFFOR: (1) Management, not just implementation, should be left to farmers whenever possible; (2) Farmer assessment is an important component of overall analysis. Measurements need to be made in order to quantitatively analyze outcomes and to diagnose reasons for observed responses; (3) Farmer control is particularly important for site selection, plot sizes, seed rate, planting patterns, and timing of agronomic operations; and (4) Exploration and demonstration activities are required to stimulate awareness and interest in technological options.

Socio-cultural factors, e.g., local labor constraints should be taken into account while conducting OFFOR. In research stations, constraints in labor are not always recognized whereas complex labor problems prevail in on-farm conditions. Laborers are employed by research stations on a permanent basis, and they are willing to perform any amount of laborious work since they are highly paid when compared to their counterparts in the country-side. The extent of labor constraints varies from region to region and many times from village to village. For instance, "planting 2-3 seedlings in rice" is an economically viable rice technology. Since planting 2-3 seedlings in rice is a time consuming process, it is difficult for farmers to convince women laborers to adopt this practice. Solutions for these kinds of problems can be identified only at the local-level. Local organizations and informal networks must be geared up so that negotiations can be drawn between farmers and women laborers. Such negotiations might end up with an intermediary technology which is congenial to both the parties.

Data pertaining to the following should be recorded from OFFOR farms:

1. Crops grown including homestead gardens;

2. Crops grown in marginal areas;

3. Direct and indirect costs involved;

4. Indigenous technical practices of farmers and their impact on productivity and sustainability of agricultural system;

5. Resource allocation due to the interaction of indigenous knowledge and research station technologies;

6. Interaction among crops, trees, livestock, and fish; and

7. Short-term benefits accrued and long-term benefits expected.

Evaluating the technological options is an essential component while conducting OFFOR. The extension scientist should evaluate the performance of technological options with respect to:

1. Compatibility with agro-ecological conditions

2. Compatibility with socio-cultural environments

3. Usage of labor

4. Usage of cash

5. Profitability

6. Need for institutional support

7. Contribution to reducing risk

Feedback from on-farm research to research station is one of the weakest linkages in on-farm research programs (Merrill-Sands and McAllister, 1988). Conducting OFFOR might contribute significantly to overcome this constraint.

Table 3a, b, c, d provides suggested methods to incorporate indigenous knowledge systems into agricultural research and extension organizations with specific cases.

4.8 Evaluating technological options

Finally, extension scientists with input from farmers should evaluate the technologies that have been tested during the OFFOR in terms of their contribution to: (a) productivity of crops and associated livestocks, (b) sustainability of the agricultural system, (c) complexity (e.g., ease of experimentation), and (d) labor intensity. They are expected to arrive at any one of the following decisions:

1. Drop the technological option that has been tested

2. Technological options need long-term research

3. Technological option is ready for further dissemination.

The technological options that are proved to be viable after the on-farm research should be disseminated to farmers using procedures outlined under the section, "process of disseminating sustainable agricultural technology to farmers by collaborating with research, extension and NGOs."

5. Technology Validation by Incorporating Indigenous knowledge Systems into Agricultural Extension Programs

The need for researcher-farmer involvement was given high priority in the recent farming systems research/extension literature. However, it is practically difficult for research station scientists to conduct research involving farmers all the time due to the insufficient human resource capacity of regional research stations (Rajasekaran and Martin, 1990; Warren, 1991). For instance, there is only one research station in Pondicherry region, India, which is expected to cater to the entire agricultural research needs of the entire region. There are approximately twenty scientists working in this station. This number is far too low when compared to the number of farming communities in the region. Keeping this low researcher-farm family ratio in view, the framework advocates the use of academically well-trained and "research minded" extension personnel to validate farmer experiments.

5.1 Subject matter specialists as researchers

Recent statistics show that most of the divisional-level subject matter specialists (SMSs) are post-graduates in different disciplines such as agronomy, soil science, entomology, and plant breeding. Moreover, the department of agriculture is sponsoring extension personnel to undergo post-graduate training in the specialized disciplines mentioned above. The advanced knowledge they acquire during this training period along with their field experience as SMSs should be used for validating farmer experimentation.

It was found that SMSs spend most of their time in headquarters assisting their heads of offices, and preparing periodical reports to be sent to their higher authorities (Rajasekaran and Martin, 1990). In other words, the academic training acquired by the SMSs is rarely exploited. They should spend at least one day in a week on activities such as: (1) problem identification; (2) recording relevant Indigenous knowledge systems; and (3) presenting the problems and Indigenous knowledge systems to the technology development consortium.

5.2 Developing extension programs to validate farmer experiments

Farmers are not passive consumers, but active problem solvers who develop for themselves most of the technology they use. For many hundreds of years before today's national agricultural research systems were set up, farmers did their own research (Pretty, 1991; Prain, 1992). And, by integrating technology from different sources and continuing to adapt it on their farms, they still do so today (Roling, 1989; Warren, 1991). Indigenous knowledge systems form the basis for informal experimentation of farmers. The factors which influence farmer experimentation according to Gupta (1990) are: (1) ecological: innovations that result due to interaction among crops, soil, and climate; (2) historical: a major happening such as crop failure or year of glut or scarcity; (3) serendipity: a practice discovered by farmers accidentally; (4) economical: farmers innovate new practices taking advantage of government subsidies for flood and drought relief activities.

Gupta and Saha (1989, p.15) provided an interesting case from Gujarat state, India, to illustrate how farmers conduct informal experiments by deviating from a conventional transfer of technology approach:

While it is a usual practice to irrigate urd crop (black gram) during 15-25 days after sowing, farmer Ajoy Kumar (21 years old) did something very exceptional. He applied water when the crop was two months old and almost dried up in the field. He did not think that it would survive. But to his surprise, he observed that after receiving water it regained its life. The green leaves appeared and the pod formation started. Farmers could not normally keep the urd crop deprived of water for two months. He explained that when the whole plant has much of the vegetative growth, the pod setting was poor. Checking the vegetative growth later on fosters the reproductive growth (fruit formation). He reaped the best harvest in the village.

Adja farmers of Benin have found through experimentation that improved maize varieties are not drought-resistant, require fertilizer or fertile soil, do not store well, and are not suitable for consumption (Dangbegnon and Brouwers, 1990). Hence, these farmer experimenters need to be involved in informal research and development activities (Biggs, 1990). Roling and Engel (1992, p.127) warned that, "to look at farmers only as users neglects the important fact that farmers are experimenters and that farmers have developed most of the technology used on the farm today." Specific extension programs should be targeted towards strengthening what farmers are already experimenting. Farmer experimenters are those farmers who conduct experiments in order to evaluate certain indigenous technical practices in their own way. Validating farmer experiments is an extension process in which SMSs encourage farmers to replicate their own experiments in their own environment in order to: (1) understand experiments in the socio-cultural and agro-ecological environments, and (2) determine the impact of the experiments on productivity, profitability, and sustainability of the agricultural system.

During bi-weekly training programs, separate sessions should be allocated to develop extension programs for validating farmer experiments. The various steps involved in the process of developing the extension programs are: (1) selecting "research minded" village extension workers; (2) identifying "research minded" farmers who are already involved in farmer experiments; and (3) establishing programs for validating farmer experiments.

Selection of farmers is one of the crucial activities during the process of validating farmer experiments. The various steps involved during the process of validating farmer experiments are: (1) Understand the rationale behind farmer experimentation. Examples are testing varieties for yield increase, blending local and external inputs, avoiding risks by adjusting sowing and harvesting periods, and testing new varieties for local adaptation; (2) Recording the mode of conducting experiments. For instance, some farmers conduct varietal trials by raising local and high yielding varieties in two different plots. Others establish experiments by planting the local and new varieties in alternate rows; and (3) Identifying farmers' evaluation criteria. The criteria used by farmers to evaluate their own experiments differ from farmer to farmer and also for the same farmer, from crop to crop. Physical stand of the crop and the way it bears the earheads is one of the major criteria for rice farmers in the Eastern Visayas region of Philippines (Tung, 1992). In Pondicherry, India, farmers randomly uproot one or two groundnut crops and shake the pods by holding them close to their ears. If they hear any sound, it indicates that the pods are unfilled. If they do not hear any sound, it indicates that the pods are filled.

Understanding, identifying, recording, and evaluating farmer experiments form the various stages of validating farmer experiments. It is important that extension personnel must understand the farmers' criteria when they explore indigenous approaches to experimentation.

5.3 Facilitating village-level experimenter workshops

Experimenter workshops should be conducted immediately after validating farmer experiments. The village extension workers should facilitate the experimenter workshops by involving farmer experimenters as resource persons. The SMSs should act as semi-silent observers during these workshops. This process is a way of empowering and respecting village-level extension workers and farmers. Farmer experimenters should be encouraged to share their experiences while conducting the experiments. They are expected to answer specific questions raised by other participant farmers. After the formal discussion, the SMSs should wrap up the workshop by sharing their experiences during the process of validating farmer experiments. The village extension worker should act as a facilitator by bringing farmers to the subject of discussion when conflicts arise and also monitor the time.

5.4 Evaluating technological options

Finally, farmer experimenters with inputs from other farmers should evaluate the technologies that have been tested during the farmer experimentation procedure in terms of their contribution to: (a) productivity of crops and associated livestocks, (b) sustainability of the agricultural system, (c) complexity (e.g., ease of experimentation), and (d) labor intensity. They are expected to arrive at any one of the following decisions:

1. Drop the technological option that has been tested;

2. Technological option needs long-term research; and

3. Technological option is ready for further dissemination.

Technological options that need long-term research should be communicated to researchers through the technology development consortium. Technological options that are ready for further dissemination but require additional resources and infrastructural facilities should be discussed with appropriate departments. Technological options that are ready for further dissemination can be communicated to their colleagues through zonal workshops.

6. Strengthening Indigenous Organizations by Utilizing Non-Governmental Organizations

Indigenous organizations are found to play important roles in facilitating non-agronomic activities such as the following:

1. Off-farm income-generating activities, especially for landless laborers including women;

2. Cooperative marketing of agricultural produce, e.g., tapioca and sugarcane;

3. Farmer-to-farmer seed exchange;

4. Cooperative marketing of milk;

5. Fodder tree planting;

6. Raising and marketing of flower seedlings; and

7. Maintenance of irrigation tanks.

6.1 Identifying indigenous organizations

Indigenous organizations are crucial for sustainable resource use and development because they can act as institutions for resource management and control. They enforce rules, provide incentives, and apply penalties for eliciting behavior conducive to rational and effective use of local resources. Local associations are embedded in local social structures and characterized by voluntary, personalistic, face-to-face transactions; hence, they tend to be highly participatory and reflect well with their members' interests (Cook and Grut, 1989). Leaders of local organizations have a comprehensive understanding of existing strengths and weaknesses in their own organizations and are exceptionally open to trying new management and planning mechanisms for development (Warren, 1992c).

Indigenous organizations also play a developmental function within the community. Strengthening the capacity of these existing organizations can greatly facilitate sustainable approaches to development (Warren, 1992c; Atteh, 1992). Identifying and strengthening indigenous organizations are challenging tasks for sustainable development of the Indian villages. Landless laborers including women who represent more than 50 percent of village population always keep their eyes open in identifying off-farm income generating activities. These activities are carried out either through formally established indigenous organizations or informal networks. Hence, it is essential to strengthen the indigenous organizations and informal networks that support and encourage these activities. Non-governmental organizations (NGOs) are found to play significant roles in strengthening informal local networks as well as indigenous organizations of local people.

Locally managed projects are often implemented by local people through informal networks. For instance, groups of women laborers in the study villages organized a duck-rearing activity using common property resources. Women's role in off-farm agricultural activities has been discussed in detail elsewhere (Poats et al., 1986). A formally established milk cooperative society in the study villages pooled and transferred the milk to central freezing plant located in Pondicherry city. NGOs can concentrate on identifying these indigenous organizations and informal networks.

6.2 Analyzing the structure and functions of indigenous organizations

The second step is to analyze the structure and function of the indigenous organizations. Well-established indigenous organizations have a clearly defined organizational structure. For instance, the village-level milk cooperative society has a president and locally elected governing body. Farmers and farm laborers who rear cattle form the members of the organization. They are formally linked to the Central Milk Cooperative Society located at Pondicherry city.

There are also informal networks without any formally organized structure that still possess well-defined functions. For instance, the women laborers joined together to form a loosely structured informal network. They jointly obtained credit from one of the large-scale farmers for whom they work as laborers. They raise ducks on a rotational basis, each member taking care of the ducks for one day. The large-scale farmers provide initial investments for the purchase of ducklings and maintenance support such as provision of shelter for the ducks during night times. Thirty percent of the output goes to the large-scale farmers and the rest is shared among the women members. Common property resources such as water streams and public lands are used to search for feed for the ducks. There exists a mutual understanding between the women laborers and large-scale farmers. By helping these women laborers to find such off-farm income sources, large farmers secure their labor resources. In other words, the large-scale farmers do not face labor problems since these women laborers come to their rescue especially during peak labor demand periods. Hence, understanding the structure and function of indigenous organizations and how they fit into the socio-cultural environments of the villages needs indepth investigation.

6.3 Identifying constraints in indigenous organizations

Identifying the constraints of the indigenous organizations form the next step of the process. NGOs can participate in informal as well as formal meetings of the indigenous organization to identify the constraints faced by the organizations. Followed by this, NGOs can work with individual members to identify their own perceptions about their problems. Examples of constraints in indigenous organizations include the following:

1. Conflicts due to social groupings in the organization, e.g., caste groups;

2. Influence of power brokers on the decision-making systems of the organization, e.g., large-scale farmers have a profound influence on the women's informal duck keeping network;

3. Non-availability of sufficient funds to run the organization; and

4. Cultural change inhibits the growth of some organizations, e.g., younger generations of cattle rearers who have some years of formal education are not willing to become members of the village-level milk cooperative society.

6.4 Developing and implementing off-farm action programs (OFFAP)

The purpose of this stage is to develop specific action programs to overcome potential constraints in the indigenous organizations and informal networks. Action programs can be developed to strengthen and empower indigenous organizations by effectively mobilizing resources from the outside. For instance, women laborer networks in the study villages face severe constraints in obtaining credit for duck rearing activity. They have to depend entirely on large-scale farmers. Action programs can also identify locally-manageable projects, such as off-farm income-generating activities for men laborers during the lean periods.

Implementing action programs on a pilot-basis is one of the most challenging tasks for NGOs. Only during actual implementation, can constraints that have not been identified in the earlier stages be understood. The members of the indigenous organizations and informal networks should be encouraged to actively participate during the process of implementation. NGOs should provide supportive roles by linking external institutions with indigenous organizations. For instance, NGOs can work with women laborers to obtain credit from credit institutions such as lending banks for purchasing ducklings. The NGOs can also help the indigenous organizations to identify profitable outlets for marketing the ducks, e.g., cooperative societies.

6.5 Evaluating action programs

Evaluating the action programs forms the final stage of strengthening indigenous organizations. Representatives of NGOs should evaluate the program in terms of its contribution to: (a) sustainable off-farm income, (b) ease of implementation, (c) limited dependence on external institutions, e.g., lending banks and marketing societies, (d) minimal social conflicts, (e) low interference with farm labor, i.e., the activities should not hinder their on-farm labor activities, (f) equity, and (g) social profitability. As a final part of the evaluation, the NGOs, in coordination with the members of indigenous organizations, should arrive at one of the following decisions:

1. Drop the program;

2. Extend the program to the rest of the population in the village; or

3. Extend the program to the rest of the population in the same village as well as implement the program in other villages provided similar agro-ecological and socio-cultural environments exist.

The NGOs are expected to share their experiences in zonal workshops if the program falls under category two or three.

7. Technology Dissemination by Linking Research, Extension, and NGOs

Technologies that are identified, developed, modified, and evaluated should be disseminated using two mechanisms: (1) informal farmer-to-farmer communication and (2) the agricultural extension system.

7.1 Informal farmer-to-farmer communication

Informal farmer-to-farmer communication forms the major source of technology dissemination in the same village and neighboring villages where technologies are developed. Informal indigenous communication systems in agricultural communities work incredibly well for the spread of farmer-selected rice and cotton varieties in India (Antholt, 1992). Nayman (1988) reported that 91 percent of the farmers in Punjab, India, sought other farmers as a source of agricultural innovation. Outsiders will play less of a role in this process. Extension has a role to play during the process of informal communication by organizing field days where on-farm research and validation of farmer experiments were conducted.

7.2 Using the existing agricultural extension system

Compton (1989) stated that extension personnel blanket the countryside. This enormous human resource capacity should be effectively utilized for disseminating technologies to distant locations and other villages. In spite of the continuous debate regarding the effectiveness of the Training and Visit (T&V) extension system, the T&V stands as the single major source for formal technology dissemination in many developing countries. The T&V system of extension has sought to operationalize a strong and regular link between research and extension, and between extension and farmers (World Bank, 1990). The salient features of the T&V such as (1) monthly zonal workshops; (2) biweekly training programs; (3) village extension workers contact with farmers; and (4) maintaining extension worker-farm family ratio can be effectively utilized. The potential of the T&V system of extension in increasing agricultural productivity has been clearly demonstrated (Antholt 1992; Feder, Slade and Sundaram, 1986).

7.3 Reorienting monthly zonal workshops

Presently, the extension personnel are entirely dependent on research station scientists for technologies. As explained earlier, it is practically difficult to rely on research stations alone for technological innovations. Moreover, NGOs, in spite of their contribution to strengthening of local networks and indigenous organizations, never form part of the monthly zonal workshops, the meeting point of researchers and extensionists. The framework aims at bringing the NGOs into the zonal workshop.

The technological options that are developed by research, extension, and NGOs should be communicated to the extension system during the zonal workshops. Research station scientists should present the technological options that are finalized from OFFOR. The SMSs of extension should share the technological options that are developed by validating farmer experiments. Representatives of NGOs should share their experiences in strengthening and empowering indigenous organizations.

7.4 Bringing original innovators to zonal workshops

Monthly zonal workshops are the important points where farmer experimenters as original innovators of technologies need to be recognized. It is essential for agricultural extension personnel to listen to the farmer experimenters whose raw materials (indigenous knowledge systems) contributed to the development of finished products (technological options). Encouraging the farmer experimenters by offering cash prizes is one of several ways of providing recognition and compensation for their contribution to the development of technologies. Such rewards also encourage their colleagues to share their knowledge by participating in the process of developing technological options.

7.5 Screening technological options

The SMSs receive technologies from zonal workshops and relay them to their village-level extension workers without tailoring these technologies to the agro-ecological and socio-cultural conditions of their own division (Rajasekaran and Warren, 1993). Once the technological options are disseminated to extension personnel, it is their responsibility to screen those options by considering the following factors:

1. SMSs should select those technological options that fit into agro-ecological environments of their division; and

2. SMSs should work with village-level extension workers in understanding the socio-cultural factors that have a negative impact on selected technological options.

7.6 Disseminating to village extension workers

After screening, the technological options should be disseminated to village extension workers. During the process of dissemination, SMSs should act as facilitators rather than simply conducting training programs for the village extension workers. The adaptability of technological options should be discussed with village extension workers. The technological options that are disseminated to village-level extension workers using these steps differ from the existing system of delivering technologies in the following ways:

1. Technologies delivered by existing research-extension system are fixed packages and rarely provide any options to farmers. The system expects the farmers to adopt an entire package. On the other hand, the technologies that are developed using the proposed framework provides diversified technological options which enable farmers to choose using their own decision-making system;

2. Presently technologies rarely build on Indigenous knowledge systems of farmers. In the new approach, technological options presented to farmers originate from the farmers' own knowledge; and

3. Under the conventional system, technologies come from only one source, the research stations. In the suggested system, the technological options are developed using diversified sources such as extension agents, NGOs, farmers, and research stations in active participation with "research minded" farmers.

7.7 Using indigenous communication channels

Village extension workers should be encouraged to follow certain guidelines while disseminating the technological options. The agricultural officers should be made responsible for providing institutional support for the village extension workers during the process of disseminating the technologies. Organizing training programs to explore indigenous communication channels for disseminating the selected technological options is essential (Mundy and Compton, 1991). Village extension workers should be encouraged to use delivery points other than farms such as shandis (market days), koil thiruvizha (village temple days), magalir mandram (a village-level women's society), and cooperative marketing points.

The following guidelines are necessary for the village extension workers while disseminating the technologies:

1. Decisions to choose a particular technology from the set of technological options should be left to the farmers;

2. If the farmers are not choosing an option from the technologies, the extension worker should encourage the adoption of the farmers' own practices since there may be some rationale behind it; and

3. The extension workers should provide relevant information to the farmers who decide to choose a technology from the technological options provided to them.

7.8 Evaluation

Evaluating technologies is the last but essential stage of the technology dissemination process. The agricultural officers should conduct the evaluation. Under the conventional system, the monitoring and evaluation (M&E) unit conducts this evaluation. Three major constraints of the existing evaluation process are (Rajasekaran and Martin, 1989):

1. Results of evaluation are rarely communicated to the field-level extension personnel. Most of the evaluation reports are circulated to extension administrators working at headquarters;

2. The M&E officials have a special status since they evaluate the work of extension personnel, thus creating fear among village-level extension workers;

3. Crop yield is the only factor considered for evaluation.

Evaluation should be conducted by middle-level extension personnel. The results should be used to refine technologies and not used merely for writing reports. The following factors should be taken into consideration during the process of evaluation:

1. Productivity (both land and labor)

2. Profitability

3. Compatibility of technologies with the farming system

4. Risk

5. Need for external resources

6. Need for institutional support (extension, credit, cooperatives)

7. Ease of testing by farmers

8. Labor intensity

9. Sustainability of agricultural system.

10. Conclusion

This framework is just a starting point in the process of identifying and developing sustainable agricultural technological options by keeping farmers' knowledge as the focal point. This framework is not a rigid `blue print' that needs to be implemented but provides essential principles to be kept in mind during the process of sustainable agricultural technology development. The framework will be of immense value to extension administrators and research policy makers while preparing their extension programs and research agendas for a 21st century focused on sustainable agricultural development. Moreover, this framework is an attempt to narrow the existing gap between researchers involved in indigenous knowledge systems and agricultural development practitioners.

The following conclusions can be made based on the framework:

1. There is much to be learned from indigenous knowledge systems of local people. If we are to move towards interactive technology dissemination from the conventional transfer of technology approach, it is feasible, efficient, and cost-effective to learn from village-level experts;

2. Devaluing indigenous knowledge systems as "low productive," "primitive," and "old" is no longer a useful attitude;

3. Establishing national indigenous knowledge resource centers is inevitable for strengthening the capacities of agricultural research and extension systems;

4. Bringing a desirable change in the attitudes and behaviors of researchers and extensionists would stimulate the process of incorporating indigenous knowledge systems into agricultural research and extension;

5. Keeping indigenous knowledge as a basis during the process of developing technologies would result in a basket of sustainable technological options rather than fixed packages;

6. Validating farmers' experiments would create an environment of respecting local people and village-level extension workers thus leading to their increased participation and empowerment; and

7. Using indigenous communication channels and farmer-to-farmer extension strategies would increase the rate of dissemination and utilization of technologies that are built on indigenous knowledge.

---

References

Amanor, K. and J. Farrington. 1992. "NGOs and Agricultural Technology Development." pp.243-256, In W.M. Rivera and D.J. Gustafson (Eds.), Agricultural Extension: Worldwide Institutional Evolution and Forces for Change. New York: Elsevier Science Publishing Company.

Antholt, C. 1992. "Agricultural Extension in the 21st Century: Lessons from South Asia." pp.203-216, In W.M. Rivera and D.J. Gustafson (Eds.), Agricultural Extension: Worldwide Institutional Evolution and Forces for Change. New York: Elsevier Science Publishing Company.

Atteh, D. 1989. Indigenous Local Knowledge as a Key to Local-level Development: Possibilities, Constraints and Planning Issues in the Context of Africa. Studies in Technology and Social Change, No.20. Ames, Iowa: Technology and Social Change Program, Iowa State University.

Baker, D. 1990. "Reorientation, Not Reversal: African Farmer-Based Experimentation." Paper presented at Tenth Annual Farming Systems Research/ Extension Conference. East Lansing, Michigan, October 14-17, 1990.

Bentley, J. and W. Melara. 1991. Experimenting with Honduran Farmer-Experimenters. ODI Agricultural Administration (Research and Extension) Network Newsletter 24: 31-48. London: Overseas Development Institute.

Bentley, J. and K.Andrews. 1991. "Pests, Peasants and Publications: Anthropological and Entomological Views of an Integrated Pest Management Program for Small-scale Honduran Farmers." Human Organization 50 (2): 113-124.

Biggs, S. 1990. "A Multiple Source of Innovation Model of Agricultural Research and Technology Promotion." World Development 18 (11): 1481-1499.

Butler, L. and J. Waud. 1990. "Strengthening Extension Through the Concepts of Farming Systems Research and Extension (FSR/E) and Sustainability." Journal of Farming Systems Research-Extension 1(1): 77-98.

Chambers, R., A. Pacey and L. Thrupp. (Eds.) 1989. Farmer First: Farmer Innovation and Agricultural Research. London: Intermediate Technology Publications.

Chambers, R. 1989. "Reversals, Institutions and Change." pp.181-195, In R. Chambers, A. Pacey and L. Thrupp (Eds.), Farmer First: Farmer Innovation and Agricultural Research. London: Intermediate Technology Publications.

Chambers, R. 1990. Micro-Environments Unobserved. IIED Gatekeeper Series No. 22. London: International Institute of Environment and Development.

Chambers, R. and J. Jiggins. 1987. Agricultural Research for Resource Poor Farmers: A Parsimonious Paradigm. IDS Discussion Paper No. 220. London: International Development Studies.

Chand, S.P. and B.D. Gurung. 1991. "Informal Research in Nepal." Journal for Farming Systems Research-Extension 2(2): 69-80.

Colfer, C., J. Pierce, D.W. Gill and F. Agus. 1988. "An Indigenous Agricultural Model from West Sumatra: A Source of Scientific Thought." Agricultural Systems 26 (2): 191-209.

Cook, C. and M. Grut 1989. Agro-forestry in Sub-Saharan Africa - A Farmers' Perspective. World Bank Technical Paper No. 112. Washington, D.C.: The World Bank.

[Compton, L. 1989. "Strategies and Methods for the Access, Integration, and Utilization of Indigenous knowledge in Agriculture and Rural Development." pp. 21-32, In D.M. Warren, L.J. Slindigenous knowledgekerveer and S.O. Titilola (Eds.), Indigenous knowledge Systems: Implications for Agriculture and International Development. Studies in Technology and Social Change Series, No. 11 . Ames, Iowa: Technology and Social Change Program, Iowa State University.]

Dangbegnon, C. and J. Brouwers. 1990. "Maize Farmers' Informal R&D." ILEIA Newsletter 6(3): 24-25.

Doorman, F. 1991. "Linkages between Research, Extension and Farmers: The Case of Rice in the Dominican Republic." Agricultural Systems 37: 107-118.

Dvorak, K.A. 1988. Indigenous Soil Classification in Semi-Arid Tropical India. A Progress Report. Patencheru, India: International Crops Research Institute for Semi-arid Tropics (ICRISAT).

Ezaza, W.P. 1989. "The Main Environmental Issues and Options for the Sustainable Development of Agriculture Associated with the Use of Mineral Fertilizer and Agro-Chemicals." Journal of Eastern African Research and Development 19: 73-82.

Farrington, J. and A. Martin. 1987. Farmer Participatory Research: A Review of Concepts and Practices. ODI Agricultural Administration (Research and Extension) Network Discussion Paper No. 19. London: Overseas Development Institute.

Feder, G., R. Slade and A.K. Sundaram. 1986. "The Training and Visit Extension: An Analysis of Operation and Effects." Agricultural Administration and Extension 21: 33-59.

Gupta, A. and M. Saha. 1989. "Local Innovations and Farming Practices in Rainfed Eastern Uttar Pradesh: An Annotated List of Farmers' Knowledge." as cited in A. Gupta, J. Capoor and R. Shah. 1990. Inventory of Peasant Innovations for Sustainable Development: an annotated bibliography. Ahmedabad, India: Center for Management in Agriculture, Indian Institute of Management.

Gupta, A. 1990. "Documenting Farmers' Innovations: or How do People Survive through Innovations in Risky Regions." Unpublished Paper. Ahmedabad: Indian Institute of Management.

Jorgensen, D.L. 1989. Participant Observation: The Methodology for Human Studies. London: Sage Publications.

Fernandez, M. and H. Salvatierra. 1989. "Participatory Technology Validation in Highland Communities of Peru." pp.146-150, In R. Chambers, A. Pacey and L. Thrupp (Eds.), Farmer First: Farmer Innovation and Agricultural Research. London: Intermediate Technology Publications.

Frio, A.L. 1991. "Gender Analysis: The Farming Systems Research-Extension Training Experience." Journal for Farming Systems Research-Extension 2(3): 1-10.

Haverkort, B. 1993. "Agricultural Development with a Focus on Local Resources: ILEIA's View on Indigenous Knowledge." In D.M. Warren, D. Brokensha, and L.J. Slikkerveer, (Eds.), Indigenous Knowledge Systems: The Cultural Dimensions of Development. London: Kegan Paul International. (forthcoming).

Haverkort, B. and H de Zeeuw. 1992. "Development of Technologies towards Sustainable Agriculture: Institutional Implications." pp.231-242, In W.M. Rivera and D.J. Gustafson (Eds.), Agricultural Extension: Worldwide Institutional Evolution and Forces for Change. New York: Elsevier Science Publishing Company.

Haugerud, A. and M. Collinson. 1991. Plants, Genes, and People: Improving the Relevance of Plant Breeding. IIED Gatekeeper Series No. 30. London: International Institute of Environment and Development.

IFAP, 1990. Sustainable Farming and the Role of Farmers' Organizations. Wageningen, The Netherlands: International Federation of Agricultural Producers.

Kaimowitz, D. 1992. "The Evolution of Links between Research and Extension in Developing Countries." pp. 101-112, In W.M. Rivera and D.J. Gustafson (Eds.), Agricultural Extension: Worldwide Institutional Evolution and Forces for Change. New York: Elsevier Science Publishing Company.

Kerr, J. and N.K. Sanghi. 1992. Indigenous Soil and Water Conservation in India's Semi-Arid Tropics. IIED Gatekeeper Series No.34. London: International Institute of Environment and Development.

Moris, J.1991. Extension Alternatives in Tropical Africa. London: Overseas Development Institute.

Merrill-Sands, D. and J.McAllister. 1988. Strengthening the Integration of On-Farm Client Oriented Research and Experiment Station Research in National Agricultural Systems: Management Lessons from Nine Country Studies. OFCOR Comparative Study Paper. No. 1. The Hague: International Service for National Agricultural Research.

Mundy, P. and L. Compton. 1991. "Indigenous Communication and Indigenous knowledge." Development Communication Report 74 (3): 1-3.

Nayman, O.B. 1988. Distant Cousins: An Evaluation of Official and Non-official Sources of Agricultural Information. Unpublished Manuscript. Punjab, Pakistan.

Norgaard, R.B. 1984. "Traditional Agricultural Knowledge: Past Performance, Future Prospects, and Institutional Implications." American Journal of Agricultural Economics 66:874-878.

Oritz, R., S. Ruano., H.Juarez., F.Oliver and A. Meneses. 1991. A New Model for Technology Transfer in Guatemala: Closing the Gap between Research and Extension. OFCOR Discussion Paper. No. 2. The Hague: International Service for National Agricultural Research.

Poats, S., A. Spring and L. Schmidt. 1986. Gender Issues in Farming Systems Research and Extension. Boulder, Colorado: Westview Press.

Prain, G. 1992. "Local Knowledge and Global Resources: User Participation in Crop Germplasm Research." Paper presented at the International Symposium on Indigenous knowledge and Sustainable Development. Silang, Cavite, Philippines, September 20-26, 1992.

Pretty, J. 1991. "Farmers Extension Practice and Technology Adaptation: Agricultural Revolution in the 17th-19th Century Britain." Agriculture and Human Values 8 (1&2): 132-148.

Pretty, J. and R. Sandbrook, 1991. "Operationalising Sustainable Development at the Community Level: Primary Environmental Care." Paper presented at the DAC Working Party on Development Assistance and the Environment, London, October 1991.

Rajasekaran, B. 1993. A Framework for Incorporating Indigenous knowledge System into Agricultural Research and Extension Organizations for Sustainable Agricultural Development in India. Ph.D. Dissertation, Iowa State University, Ames, Iowa.

Rajasekaran, B. and R. A. Martin. 1989. "Organization and Management of Training and Visit Extension System in Developing Countries." Paper presented at the Ninth Farming Systems Research/ Extension Conference. Fayetteville, Arkansas, October 18-21, 1989.

Rajasekaran, B. and R.A. Martin. 1990. "Evaluating the Role of Farmers in Training and Visit Extension System in India." Paper presented at the Tenth Farming Systems Research/Conference. East Lansing, Michigan, October 14-18, 1990.

Rajasekaran, B., D.M. Warren and S.C. Babu. 1991. "Indigenous Natural-Resource Management Systems for Sustainable Agricultural Development - A Global Perspective." Journal of International Development 3 (4): 387-402.

Rajasekaran, B. and D.M. Warren. 1993. "Indigenous Rice Taxonomies and Farmers' Rice Production Decision-Making Systems in South India." In D.M. Warren, D. Brokensha, and L.J. Slikkerveer, (Eds.), Indigenous Knowledge Systems: The Cultural Dimensions of Development. London: Kegan Paul International. (forthcoming).

Raman, K.V. and T. Balaguru. 1990. "Human Resources Development as a Key to the Success of Farming Systems Research in India." Journal of Farming Systems Research-Extension 1 (1): 153-164.

Reijntjes, C., B. Haverkort and A. Waters-Bayer. 1992. Farming for the Future: An Introduction to Low-External Input and Sustainable Agriculture. London: Macmillan.

Richards, P. 1985. Indigenous Agricultural Revolution: Ecology and Food Production in West Africa. London: Hutchinson.

Richards, P. 1989. "Agriculture as a Performance." pp.39-42, In R. Chambers, A. Pacey and L. Thrupp (Eds.), Farmer First: Farmer Innovation and Agricultural Research. London: Intermediate Technology Publications.

Rhoades, R. and R. Booth. 1982. "Farmer-Back-to-Farmer: A Model for Generating Acceptable Agricultural Technology." Agricultural Administration 11 (1): 127-137.

Rocheleau, D. 1987. "The User Perspective and the Agroforestry Research and Action Agenda." Paper presented at Workshop on Farmers and Agricultural Research: Complementary Methods. Institute of Development Studies, University of Sussex, July 28-31, 1987.

Roling, N. 1989. "Misleading Metaphors." In D. Merrill-Sands, and D. Kaimowitz (Eds.), Technology Triangle: Linking Farmers, Technology Transfer Agents and Agricultural Researchers, The Hague: ISNAR.

Roling, N. and P. Engel. 1992. "The Development of the Concept of Agricultural Knowledge Information Systems (AKIS): Implications for Extension." pp.125-138, In W.M. Rivera and D.J. Gustafson (Eds.), Agricultural Extension: Worldwide Institutional Evolution and Forces for Change. New York: Elsevier Science Publishing Company.

Salas, M.A. and J. Tillman. 1989. "What Should be Taken into Consideration in Training and Extension? Example Thoughts on Training Programmes for Small Farmers in Peru." Entwicklung-Landlicher-Raum (in German) 23(5): 25-28.

Scoones, I. 1989. Patch Use by Cattle in Dryland Zimbabwe: Farmer Knowledge and Ecological Theory. ODI Pastoral Development Network Paper No. 28. London: Overseas Development Institute.

Thrupp, L. 1989. "Legitimizing Local Knowledge: Scientized Packages or Empowerment for Third World People." pp.138-153, In D.M. Warren, L. J. Slindigenous knowledgekerveer, and S.O. Titilola, Indigenous knowledge Systems: Implications for Agriculture and International Development. Studies in Technology and Social Change No. 11. Ames: Iowa State University.

Thurston, D. 1992. Sustainable Practices for Plant Disease Management in Traditional Farming Systems. Boulder, Colorado: Westview Press.

Tripp, R. and J. Woolley. 1989. The Planning Stage of On-Farm Research: Identifying Factors for Experimentation. Mexico, Colombia: CIMMYT and CIAT.

Tung, L. 1992. "Farmer Experimentation." On-Farm Research Notes 11:2.

Venkataratnam, L. 1990. "Farmers' Wisdom for a Sustainable Agriculture." Kisanworld 7(12): 22.

Warren, D. M. 1987. "Editor's Notes." CIKARD News 1(1): 5.

Warren, D.M. 1989. "Linking Scientific and Indigenous Agricultural Systems." pp. 153-170, In J.L. Compton, (Ed.), The Transformation of International Agricultural Research and Development. Boulder: Lynne Rienner Publishers.

Warren, D.M. 1990. "Indigenous knowledge Systems and Development." Background paper for Seminar Series on Sociology and Natural Resource Management. The World Bank, Washington, D.C. December 3, 1990.

Warren, D.M. 1991. Using Indigenous Knowledge in Agricultural Development. World Bank Discussion Paper No. 127. Washington, D.C. : The World Bank.

Warren, D.M. 1992a. "Indigenous knowledge, Biodiversity Conservation and Development." Keynote Address at International Conference on Conservation of Biodiversity in Africa: Local Initiatives and Institutional Roles. Nairobi, Kenya August 30 - September 3, 1992.

Warren, D.M. 1992b. "Indigenous knowledge and Sustainable Development: A Review of Critical Research Areas and Policy Issues." Paper presented at the International Symposium on Indigenous knowledge and Sustainable Development. Silang, Cavite, Philippines, September 20-26, 1992.

Warren, D.M. 1992c. Strengthening Indigenous Nigerian Organizations and Associations for Rural Development: The Case of Ara Community. Occasional Paper No. 1. Ibadan: African Resource Centre for Indigenous knowledge.

Warren, D.M. 1992d. A Preliminary Analysis of Indigenous Soil Classification and Management Systems in Four Ecozones of Nigeria. Discussion Paper RCMD 92/1. Ibadan, Nigeria: Resources and Crop Management Division (RCMD), International Institute of Tropical Agriculture.

Warren, D.M., L.J. Slikkerveer and S.O. Titilola. (Eds). 1989. Indigenous knowledge Systems; Implications for Agriculture and International Development. Studies in Technology and Social Change No. 11. Ames, Iowa: Technology and Social Change Program, Iowa State University.

Warren, D.M.and B. Rajasekaran. 1991. Utilizing and Integrating Indigenous knowledge Systems for Agricultural and Rural Development: Training Manual and Guide for Extension Workers. Ames, Iowa: CIKARD, Iowa State University.

Waters-Bayer, A. 1987. Participatory Technology Development in Ecologically Oriented Agriculture: Some Approaches. ODI Agricultural Administration (Research and Extension) Network Discussion Paper No. 7. London: Overseas Development Institute.

Waters-Bayer, A. and J. Farrington, 1990. "Supporting Farmers Research and Communication: The Role of Grass-roots Agricultural Advisors." Paper presented at the 10th Annual Conference of the Association for Farming Systems Research/ Extension. East Lansing, Michigan. Oct.14-17. 1990.

World Bank 1990. Agricultural Extension: The Next Step. Policy and Research Paper No. 13. Washington, D.C. : The World Bank.