Chapter 14
Alternative research strategies

Like other elements of a SARM strategy, research should as far as possible be managed by land users themselves, with the appropriate support from both the extension and research services. Experimentation in farmers' fields ensures that improved practices are developed under farmers' conditions and overcomes the difficulty of using experiment station results to make farmer recommendations.

The research which is most valuable for SARM is primarily that which is carried out on farmer's own land, designed in consultation with them, and as far as possible monitored and supervised by them. An advantage of conducting on-farm research, particularly in a participatory manner, is that during the research phase the experiments can be used to familiarize farmers with the technologies involved and the successful trials will be useful demonstrations of the results. Likewise if a significant proportion of the farming community has been involved in the trials, either in field implementation or discussion of the results, there may be little need for a formal extension programme as the benefits will have already been disseminated to the others by word of mouth and their own observation of the trials.

Agricultural research has suffered in the same way as extension in often being compartmentalised on a separate commodity basis. It is critical that research for sustainable agriculture takes a holistic viewpoint which means that it should be undertaken by inter-disciplinary teams rather than narrowly specialised individuals. Such teams should not just contain natural science specialists (agronomists, soil scientists etc) but also social science specialists (sociologists, economists etc).

Considering the research needs

When reviewing alternative technical options for tackling SARM problems there is a need to determine whether all, or only some, are sufficiently tried and tested to be directly disseminated to farmers. For some options there may be a need if research is: (1) technically sound, sustainable and adoptable; (2) financially attractive and affordable; and (3) socially and culturally acceptable.

The aim of such a review is to be able to classify each option according to the following categories:

• Notional. A notional option is one that has had very little if any prior research or field experience. It is included because it is thought to have a theoretically high potential to improve significantly the productivity and sustainability of the local farming system(s), while being financially beneficial and socially acceptable, however the theoretical technical feasibility and financial benefits have still to be proven. Initially research and development of notional options would need to be conducted on-station as not enough is known about them to involve farmers in their testing.
• Preliminary. A preliminary option is one that has received some previous research or is known to exist elsewhere in traditional farming systems which are thought comparable to one or more of the farm household common resource management domains (see chapter 2) in the area of concern. Past experience suggests that it has definite potential to improve the productivity and sustainability of the local farming system(s), while being financially beneficial and socially acceptable, however this needs to be verified before dissemination. Preliminary options are sufficiently developed to be included in on-farm trials with the aim of developing and improving them further as well as testing them under the bio-physical and socio-economic conditions typical of the intended users.
• Validated. A validated option is one that has a well founded basis in past research or existing usage (under conditions very similar to the area of concern) and is considered suited to the bio-physical and socio-economic circumstances of one or more of the farm household socio-economic groups within the area. Past experience shows it to have the capability to improve the productivity and sustainability of the local farming system(s) while being financially beneficial and socially acceptable. It can therefore be recommended to the intended users with confidence without the need for further testing.

In an ideal world enough there would be sufficient validated SARM technologies to make research unnecessary. In practice many of the newer ideas on conservation and particularly those that aim to tackle land degradation through the development of productive farming systems e.g. agroforestry, minimum tillage, intercropping, organic matter management etc. are based on practices which are likely to be notional or preliminary with regard to the bio-physical and socio-economic circumstances facing small-scale farmers within a specific locality. It is important to be aware of this. Whereas it may be possible to immediately recommend the dissemination of some field level technical solutions, other potential recommendations can be expected to require an initial research and field testing programme in order to validate them.

On-farm or on-station research

On-farm research is research conducted in farmers' fields with or without the active participation of farmers. On-station research is conducted by researchers without the direct participation of farmers and usually within the grounds of regional, national or international experiment stations.

There is an on-going debate as to the relative merits of undertaking research on-station or on-farm. In the past too many inappropriate recommendations have been made by research workers working solely within the confines of a research station, thereby isolated from the socio-economic realities of real farming conditions. Hence the present recognition of the need for undertaking some, if not all, of the experimentation within farmers' fields so as to ensure new recommendations are developed under conditions representative of those experienced by the target farmers.

Experimentation in farmers' fields is intended to ensure that the proposed improvements are developed under farmers' conditions and to overcome the difficulty of using experiment station results to make farmer recommendations. Important where the existing experimental stations are not representative of the local farming conditions, for instance because of their high management levels or geographic location. It also aims to promote effective communication between farmers and researchers so that the researchers, have a greater awareness and understanding of the problems and constraints faced by farmers, when they seek to improve their land use enterprises and farm management practices. As the following table shows, the fact that research is conducted in farmers' fields does not mean that it is automatically participatory research.

Types of on-farm research

Source Pretty 1995

There is a role for both on-station and on-farm research in SARM. The need is to get the right balance between them. There are no hard and fast rules on this as it will depend on the nature and severity of the problem, the circumstances of those farmers experiencing the problem, and what work has already been done to solve the problem, or similar ones. The general consensus would appear to be that on-station research should be kept to a minimum, and that it should support and compliment on-farm research and not be undertaken in isolation of it. It should be comprehensible to visiting farmers and extension technicians so academic complexity should wherever possible be avoided in designing such research.

On-station research should be restricted to those topics that cannot effectively be undertaken within farmers' fields. For instance when the development of new technologies and farming practices requires controlled conditions in order to determine which variables are critical in determining the final outcome. Suitable topics for on-station research will include breeding new crop varieties, testing alternative conservation practices for their technical potential to control runoff and prevent soil loss, monitoring soil changes under different agroforestry interventions etc.

Farmer participation in agricultural research

To date on-farm research is still largely a top-down process with the technologies to be tested having been decided on by the `outside expert' (the researcher) even though these may be based, as in farming systems research, on a detailed assessment of farm household needs and constraints. So far there has been little recognition of the role farmers can play as technology developers in their own right. Likewise the extent to which farmers actively engage in developing their own improved technologies (suited to their individual household bio-physical and socio-economic circumstances) has been largely underestimated and their research capability under utilised.

These days it is common to find that all on-farm research is described as farmer participatory research however as one recent report noted there is often a contradiction between the "rhetoric and reality" at the field level (Okali et al 1994). In reviewing farmer participation in agricultural research they identified the following four modes of participation:

Source: Biggs 1989

Farmers' experiments

There is now an increasing body of evidence to show that farmers conduct their own experiments (see Chambers et al 1989, Haverkoort et al 1991). Farmers experiments are generally one of three types (Rhoades and Bebbington 1991):

• Curiosity experiments - i.e. trying out a new idea purely out of curiosity, such as planting a few rows of a food crop variety acquired during a visit to another area to see how it performs and tastes;
• Problem-solving experiments - i.e. testing possible solutions to old and new problems, such as tie-ridging a small part of the field as an experiment to conserve moisture and improve yields; and
• Adaptation experiments - i.e. testing a new technology or modifying an existing one to see if it works within the known environment of their farm household system, eg trying out a new crop variety or farming practice promoted by the local extension service.

Strengths and weaknesses of farmers' experiments

The experimental methods used by farmers vary widely. As they are specific to the local communities and rooted in long history, their validity and limits will vary and may be difficult to assess. Some strengths of farmers' experiments are that (Haverkoort et al 1991):

• The focus for experimentation is on issues directly relevant to the farm household's needs and circumstances;
• They start with the knowledge of farm household members and their results expand and deepen this knowledge;
• They are usually directed to improving the use of locally available resources;
• They use evaluation criteria which are directly related to the values and objectives of the farm household, particularly regarding taste and utilisation;
• Observations are iterative and made from within the household, as they take place during routine farmwork, so evaluation is on-going and not only based on the end result, eg farmers look at more than just the final yield.

However farmers' experiments as an alternative to conventional research experiments also have their methodological limits (Haverkoort et al 1991):

• The search for improved technologies may be based on limited scientific understanding of the processes involved and technical knowledge of alternative options;
• Farmers may use a new technology over their whole field so that comparison can only be made with a crop of a previous year or in a neighbours field;
• Errors in experimental design such as poor replication of trials may lead to non statistically justified conclusions;
• Methods of measuring and weighing may not be scientifically accurate;
• Farmers, because of their limited theoretical understanding of biological or other processes, may attribute crop performance to one obvious factor, and not realise this is the result of several interrelated factors, or the result of the intervention of less observable factors;
• Communication about the results may be limited to certain geographical areas, gender and/or socio-economic groups.

Participatory technology development

The term `Participatory (or people-centred) Technology Development' (PTD) refers to approaches that aim at strengthening local capacities to experiment and innovate. Farmers are encouraged to generate and evaluate indigenous technologies and to choose, test and adapt external technologies on the basis of their own knowledge and value systems.

PTD is not a substitute for station-based research or researcher- managed on-farm trials. It is a complementary process which involves linking the power and capacities of agricultural science to the priorities and capacities of farming communities, in order to develop productive and sustainable farming systems. PTD is seen as a way of enabling farmers to further develop and validate what would be considered potential options. PTD could be used as a means to encourage farmers to confirm for themselves, and perhaps modify, options that the PRA team believe to be sufficiently validated for immediate adoption.

The PTD approach

PTD is a process of purposeful and creative interaction between local communities and outside facilitators which involves (after Reijintjes et al 1992):

• Gaining a joint understanding of the main bio-physical properties and constraints of the different land management units used for agricultural and forestry purposes by a particular rural community;
• Gaining a joint understanding of the socio-economic circumstances of the different rural household groups within the community;
• Defining priority problems;
• Experimenting locally with a variety of options derived from both indigenous knowledge (i.e. from local farmers and farmers elsewhere) and formal science; and
• Enhancing farmers' experimental capacities and farmer-to-farmer communication.

PTD not only seeks to generate technologies adapted to local environments, but also seeks to develop the local capacities, socio-cultural structures and organizational linkages necessary to sustain the process.

General sequence of PTD activities

PTD is a key component of the participatory approach to better land husbandry. It would be undertaken when the appraisal of the `best bet' options for tackling the production and sustainability problems reveals a need for potentially promising options to be locally validated prior to dissemination and widespread adoption. The sequence of activities for PTD (see table 12) can be grouped under the following headings (after Reijintjes et al 1992):

• Looking for things to try - gathering information for detailed analysis and prioritisation of locally felt problems and identifying promising solutions, in order to set up an agenda for experimentation;
• Designing experiments - developing patterns and methods of trying things out, which give reliable results and can be managed and evaluated by the farmers themselves;
• Trying things out - carrying out, measuring and assessing the experiments, and simultaneously building up farmers' experimental skills and strengthening their capacity to conduct and monitor experiments;
• Sharing the results - amplifying spontaneous diffusion of the results of farmer experimentation by building up a programme to improve farmer-to-farmer communication;
• Keeping up the process - activities that lead to replication of the PTD process and create favourable conditions for ongoing low external input sustainable agriculture technology development.

Table 9. Participatory Technology Development Activities

Source: Reijntjes et al 1992.