Livestock Research for Rural Development 3 (1) 1991

Citation of this paper

Adding a learning to a blueprint approach - or what a small amount of flexible money can do -

Frands Dolberg

University of Aarhus, Aarhus V, Denmark


Project design and execution in the area of rural development, and in agriculture generally, classically follow the "blueprint" (top- down) approach. The alternative, non-conventional way is by "learning" (from below or "bottom-up" approach). An example is given of the results obtained in Bangladesh when these two approaches were taken sequentially in a freshwater fish culture project. The "learning" approach involved the use of University students to study and research a range of topics related to village fish culture on which it was considered there was insufficient local knowledge. The work was done in villages with the collaboration of the local farmers and landless labourers. The information that was gained was incorporated into guide books and handouts for use in face-to-face discussions with small groups of fish farmers. The "learning" approach proved to be very successful, measured in terms of the expansion (by a factor of four) of fish farming in the project area. It is argued that development cannot be considered as simply the need for transfer of technology, on the basis that knowledge already exists on which such technologies can be built. More often, the reverse is the case, and it is the build up of knowledge through farmer-participatory research which leads to more rapid project implementation and greater impact.

KEY WORDS: Farmer participatory research, blueprint, learning, fish culture, training, extension, Bangladesh, villages.


A number of publications have recently been presented describing experiences with complementary methods of, or even alternatives to, conventional research and extension methods for farmers in developing countries (Farrington and Martin 1988).

At a general level a distinction has been made between a "blueprint" and a "learning approach" to rural development. This includes approaches to agricultural development.

New methods and trends

The contrasts (table 1) have been elaborated by Chambers (1986).

Table 1: The Blueprint and Learning Process Approaches in Rural Development
idea originates in capital city village
first steps data collection and plan awareness and action
design static, by experts evolving, people involved
supporting organisation existing, or built top-down built bottom-up, with lateral spread
main resources central funds and technicians local people and their assets
staff development classroom, didactic field-based action learning
implementation rapid, widespread gradual, local, at people's place
management form spending budgets, completing projects on-time sustained improvement and performance
content of action standardised varied
communication vertical: orders down, reports up lateral: mutual learning and sharing experience
leadership positional, changing personal, sustained
evaluation external, intermittent internal, continuous
error buried embraced
effects dependency-creating empowering
associated with normal professionalism new professionalism


Source: Chambers (1986), p. 23 & 227.


More specifically for agriculture, a distinction has been made between three types:

The third type has also been termed undervalued agriculture (Chambers 1988). The three types have been characterised (table 2). Having identified the main types of agriculture and understood the important differences between them, it becomes easier to understand why there are problems in applying explanations and prescriptions developed in the context of one system to the circumstances of another.

It is also this gradual realisation, which underlies the sequence of dominant attitudes, that have been used in the explanation of non-adoption by farmers in the developing countries since the Second World War (table 3).

Most of the agriculture in Bangladesh falls into the category of "undervalued" agriculture, with too much water in the wet season and too little in the dry being the main problem. Although in a general sense there is a widespread understanding that approaches under these circumstances have to be different to conventional ones, the conventional way prevails as norms and practices deeply engrained in curricula, textbooks and technical premises for project preparation and implementation. Among people responsible for decision-making, project preparation and implementation, non- conventional attitudes are represented but they are rarely foremost.

What follows is a description - with the benefit of hindsight - of an attempt to build up expertise "from below" within fresh-water fish culture in an integrated rural development project in Bangladesh. The project started implementation in 1978.



In 1983-84, marine fishery contributed 23% and inland fishery 77% to the total national catch in Bangladesh (Ameen 1987). The potential for increased fish production is considerable. It has been found that with fish culture of a low intensity of fish stocking, limited fertilizer and feed application, a harvest of 1150 kg fish/acre can be achieved. This figure has to be compared with 150 in derelict and 340 kg/acre in traditionally cultured ponds (Ameen 1987). The potential has also been demonstrated in experiments involving polyculture of carp with prawn. It was found in one trial that, by investing taka 370 (US$ 12) in extra feed, the value of additional prawn harvest was taka 2100 - just to mention an example.

Table 2: Three Types of Agriculture Summarised



Main locations Industrialised countries Irrigated and high rainfall, high potential areas in the South Rainfed tropics, hinterlands, most of sub-Saharan Africa, etc.
Climatic zone Temperate Mainly tropical Mainly tropical


Condition Over-developed Developed Underdeveloped
Current production as percentage of sustainable production Far too high Often near the limit Low
Priority fo production Reduce production Maintain production Raise production



Topography usually Flat or undulating Flat Undulating
Farming system, relatively Simple   Complex
Environmental diversity relatively Uniform   Diverse
Relative stability Low risk   High risk
Use of external inputs Very high High Low



Similarity of farmers’ and research station conditions High   Low
Farmers consulted about research priorities Richer farmers sometimes   Rarely
Number of scientists/extensionists per farming system More   Fewer


CDR = complex, diverse, risk-prone
Source: Chambers (1988).

Table 3: Non-adoption: Changes in Explanation and Prescription
Stage of explanation Model Period when dominant Explanation of non-adoption Prescription
Transfer of Technology
TOT1 1950s,
Ignorance of farmer Agricultural extension to transfer the technology
Farmer-First-and-Last approach
TOT1 l970s
Farm-level constraints Ease constraints to enable farmers to adopt the technology
Resource-Poor Farmer
FFL2 Latter
1980s for
The technology does not fit RPF conditions FFL to generate technology which does fit RPF conditions


Source: Chambers and Ghildyal (1985).

In spite of this obvious potential, very little literature existed in 1978 recording practical field experience with inland fish culture. For fish culture in small to medium-size ponds, such literature for all practical purposes did not exist.

According to Gill and Motahar (1982) a comprehensive survey of fisheries research in Bangladesh conducted by Bangladesh Agricultural Research Council revealed that, of more than 700 ongoing projects and scientific papers, only seven dealt with social or economic aspects. Of these seven, five dealt exclusively with distribution and marketing. The survey concluded that the studies that had been carried out thus far, were more academic than applied and that social (and by inference therefore also rural) aspects hardly had been touched.

With hindsight and reference to tables 1, 2 and 3, this was only to expect. However, that was the situation by the beginning of the project.

Content of the project - the blueprint

Re-excavation of government-controlled ponds, formation of fish pond cooperatives of landless people, extension work and credit were the main activities listed in the original project document.

However, without a solid knowledge of fish culture in place, too great an emphasis on re-excavation and credit contained all the dangers inherent in spending money in order to reach a physical target, which may not be reached anyway (Wood 1988) and without any guarantee of more fish production. Formation of cooperatives at predetermined points (where the ponds were) is a top-down approach beset with its own problems although the large dairy project in India known as "Operation Flood" has shown that a top-down approach can work, when the responsible organisation is efficient and the leadership dedicated to the cooperative course and itself having a "bottom-up" experience (Kurien 1987).

Shortly after implementation started, the original targets were scaled down and money shifted to training. Much emphasis was placed on short 1-2-week training courses of a very practical nature, sometimes involving successful fish culturists as trainers. Some of the topics covered were: fish fry (sources of supply, quality, transport, liberation in ponds); fish pond management; diseases in fish, their treatment and artificial spawning. Some related topics were: poultry and duck farming in relation to fish culture; cultivation of vegetables and fruits on pond embankments.

Extension work was maintained as originally outlined to be carried out by the staff of the Department of Fisheries.

Students - the learning

To overcome the shortage of relevant knowledge, collaboration with a group from the Zoology Department of Dhaka University, lead by Professor M. ul-Ameen, was started in 1980. As it is commonly held that university students of developing countries are not willing to work in rural areas, it is only right to record that there was no major difficulty in recruiting students who were willing to board the night-train, sample data for their thesis over the weekend in a village pond in the project area, to return by train the following night to Dhaka to be ready for the morning classes. For an M.Sc. thesis, they typically did this once a month for a year. Their travel expenses and food were paid by the project, but nothing else - an expense of some US$ 300 per student per year.

Over the period 1982-1985, 17 students completed their M.Sc. thesis. However, students are still getting involved every year.

The topics they studied were for example:

These problems relate to some of the common management issues of fish ponds arising at the start of a programme. As the programme grew older, the topics changed and in 1988 the important ones for research were listed as:


The importance of getting students involved is not only the opportunity it offers for them to develop expertise on a technical subject, but also that they get the experience of cooperating with pond owners and work in a rural and a project situation.

Extension material

Based on this learning process (Ameen et al 1983, and Davis et al 1983), the training centre of the project developed an educational set, consisting of five lessons (Pond Fish Culture 1984) designed to support face-to-face discussions with relatively small groups of 20-30 fish farmers. It consists of three main parts:


A teacher's guide is designed to allow the instructor to be familiar with the main points of the lessons and to promote discussion with the participants.

Flip chart sets.

The lessons are structured through the use of five flip chart sets. Several practical demonstrations are included as outlined in the teacher's guidebook.


The five lessons contain many pieces of information and data which will be difficult for the participants to remember after the lesson. Handouts with key-points are therefore a natural part of the material. They are distributed to the participants after each lesson and not too difficult for even school children to read.

For efficient extension work other tools than these materials are needed. They include manpower, transport facilities, equipment, etc. However, the point to emphasize is that producing, using and - not least - keeping such an extension material up-to-date puts the objectives into sharp focus for which the other tools are meant. And it is very efficient in identifying gaps of knowledge, where further research is required.

Impact and effect

The critical question is of course whether the demonstration of fish culture according to the principles developed by the project has led to "spread and take up".

The answer can be very simple, because more than 20 private fish hatcheries have been built in the area and their production of fish spawn has expanded as shown in table 4.

Table 4: Production of Spawn in Private Hatcheries






kg spawn





Source: NRDP/Danida Project Monitoring and Evaluation Unit, personal communication.

Elsewhere in Bangladesh it has been estimated (Deppert et al undated) that 1 kg spawn leads to a value in terms of table fish of 800,000 Bangladesh Taka (24,353 US$). For 738 kg the value is approximately 18 million US$. However, according to the information provided in table 4, the trend is towards expansion. Basically, expansion would not take place if fish culture were not profitable. However it is profitable, and, within that framework, impact studies suggest that input demonstrations and training are making significant contributions to the expansion. But it is, first of all, the results of the students' engagement that decide what can be demonstrated and assist in providing information for the training.

Obviously there are many other factors contributing to the adoption of modern fish culture practices. However, the distinctive feature in this project has been a continuous learning process, facilitated by a small amount of flexible money, which apart from its technical contribution has also helped to show that it is a myth that students in a developing country are not prepared to participate in the development efforts. The critical factors are appropriate guidance by the teacher, senior officials and expatriate advisers - and continuity, to achieve results. In general, a lack of conducive attitudes and skills at that level may be what limits student involvement.


The conventional approach to rural development is to assume that knowledge exists. The task is therefore to reinforce the implementation mechanism and support it with planning, accounting, monitoring and evaluation, training and communication. In this arrangement, research is the task of separate institutions. At times the opposite view is also presented - that if appropriate knowledge exists, farmers will learn from each other without any particular support from extension. On the latter view, a close link between farmer and research becomes the key to rapid implementation. From the real world, examples can be drawn which support both schools of thought.

It is interesting that an independent group, sponsored by the UNDP, in its sector review of agriculture in Bangladesh in 1988 concludes that a closer farmer-research-extension linkage is required. In another review from 1988 conducted on behalf of the International Service for National Agriculture Research, Stoop (1988) concludes:

  • "The implications are that the research-extension interface functions of testing and adaptive research, input and feedback, regulation, training, communication, and services (library, laboratory) may be conducted by either research or extension, depending on their relative strengths, or even to some extent by farmer organisations. Over time certain shifts in responsibility will therefore be required. Consequently, blueprints of the organization of the research-technology transfer interface become rather irrelevant."
  • Perhaps the point to make is that in 1989 it is being realised that systematic learning (research) is also important for successful project implementation.


    Ameen Mahmud-ul 1987 Fisheries, Resources and Opportunities in Freshwater Fish Culture in Bangladesh, Bangladesh, Noakhali, Project Advisory Team NRD-II/Danida.

    Ameen Mahmud-ul, Ali S , Z. Begum Z, Munyem F M A, Rahman M, Ray T, Ahmed N, Bhuiyan A R and Davis C H 1983 "Traditional versus semi-intensive fish culture in ponds: remarkable differences in fish production", in Maximum Livestock Production from Minimum Land, 1983 (Editors: C H Davis, T R Preston, Mozammel Haque and M Saadullah). Bangladesh Agricultural University.

    Chambers R 1988 Farmer-first: A Practical Paradigm for the Third Agriculture, England, University of Sussex, Institute of Development Studies.

    Chambers R 1986 Normal Professionalism, New Paradigms and Development, England, University of Sussex, Institute of development Studies, Discussion Paper No. 227.

    Chambers R and Ghildyal R P 1985 "Agricultural Research for Resource-poor Farmers: The Farmer-First-and-Last Model", Agricultural Administration Volume 20, pp 1-30.

    Davis C H, Bhuiyan A R and Ameen Mahmud-ul 1983 "Fish production in managed farmers ponds with different feeding and stocking regimes", in Maximum Livestock Production from Minimum Land, 1983 (Editors: C H Davis, T R Preston, Mozammel Haque and M Saadullah). Bangladesh Agricultural University.

    Deppert D, Perschbacher P, Rahman M and Rahman A (undated) Preliminary Survey and Analysis of the Impact of Hatchery Fry (Spawn) Production on the Economy of the Greater Mymensingh Area, in a collection of technical reports from Freshwater Aquaculture Research Station Mymensingh, Bangladesh.

    Farrington J and Martin A 1988 Farmer Participation in Agricultural Research: A Review of Concepts and Practices, England, Overseas Development Institute, Agricultural Administration Unit, Occasional Paper No 9.

    Gill G J and Motahar S A 1982 "Social factors affecting the use of the aquatic system in farming in Bangladesh", in Maximum Livestock Production from Minimum Land 1982. (Editors: T R Preston, C H Davis, F Dolberg, Mozammel Haque and M Saadullah). Bangladesh Agricultural University.

    Kurien V 1987 From a Drop to a Flood, India, National Dairy Development Board.

    Pond Fish Culture 1984 Guide to the Fish Pond Educational Programme. NIRDP/Danida Media Unit, Noakhali. (Editor: E Wickman).

    Stoop W A 1988 NARS Linkages in Technology Generation and Technology Transfer, The Netherlands, International Service for national Agriculture Research.

    Wood G D 1988 "Plunder without danger: Avoiding responsibility in rural works administration in Bangladesh", IDS Bulletin, Sussex, Volume 19, Number 4.


    (Received 1 March 1990)