Home | Collected Papers | Turnover Reflections: Contents | To:BryanBruns

8 - Promoting Participation in Irrigation: Reflections on Experience in Southeast Asia

Bryan Bruns
Published in World Development 22 (11): 1837-1849





Irrigation Organizers


Participatory Siting

Contributing to Construction

Moral Hazards in Maintenance


Research and Action: Fragile Linkages

Evaluation: Missing in Action

Decentralization and Empowerment



REFERENCES #References

Abstract: Comparative analysis of efforts to improve local participation in irrigation in Thailand, Indonesia and the Philippines offers a basis for examining achievements, problems and opportunities for the future. While many alternatives seem to exist for who can carry out the role of organizer more attention should be directed to the problems of training and supervising organizers. Farmers demonstrably can improve the siting of structures but design innovation is still scarce. Requirements for local contributions are more feasible if built into projects from the beginning. Government assistance for maintenance and repair faces problems of moral hazard. Linking research and action is a fragile process. Further progress in participation requires going beyond reforming centralized agencies to create additional means for enabling greater democracy, diversity and self-reliance.


Improving local participation continues to be been seen as one of the most promising ways to make irrigation schemes perform better (Uphoff 1986, Ostrom 1990:4-6). The pioneering efforts of the Philippine National Irrigation Administration (Korten and Siy 1988) helped inspire similar efforts in many countries including Indonesia and Thailand. Irrigation agencies fielded organizers to facilitate farmer involvement in design, construction, operation and maintenance of irrigation systems (Manor and others 1990). Researchers documented activities and took part in refining participatory approaches. Many reports have detailed the experience of individual projects. However, besides Uphoff's (1986) book there have been few attempts to look comparatively at what has been learned about participation in irrigation. Comparison makes it possible to find common patterns, identify more fundamental constraints that otherwise might be blamed on the idiosyncrasies of particular individuals or institutions and look more generally at the feasibility of institutionalizing participatory approaches to irrigation development.

... there have been few attempts to look comparatively at what has been learned about participation in irrigation. ...

This paper reflects on efforts to improve participation in irrigation, drawing particularly on experiences with small scale irrigation in Thailand and Indonesia. Participation as discussed here does not mean simply sharing costs, receiving benefits or being incorporated in a government controlled organization, but means farmers have an active role in making important decisions about irrigation system development and management. The discussion focuses on issues faced by most efforts to improve local participation in irrigation: institutionalizing the role of organizers, increasing participation in design and construction, strengthening the role of water users in operation and maintenance, combining research and action, and the opportunities for more diverse and decentralized approaches to irrigation development. The goal of the paper is to identify conclusions that may be of use in considering future directions for participatory irrigation development in Southeast Asia and elsewhere.

Background <-

Beginning in the 1970s senior officials of the National Irrigation Administration in the Philippines collaborated with researchers in a series of projects to develop institutions through which farmers could take a greater role in developing and managing irrigation systems. These activities took place in the context of reductions in government subsidies to the National Irrigation Agency (NIA) intended to make the irrigation agency more reliant on revenues collected from farmers. Agency officials hoped that changes would make farmers more satisfied and willing to pay toward the costs of developing and managing irrigation systems. Requiring repayment of construction costs increased accountability to farmers. Specific elements of the Philippine reforms are outlined in later sections, in comparison with activities in Thailand and Indonesia.

... the National Irrigation Administration in the Philippines collaborated with researchers in a series of projects to develop institutions through which farmers could take a greater role in developing and managing irrigation systems. ...

In 1985 the Thai Royal Irrigation Department (RID) and Khonkaen University began work to develop methods for improving participation in the development of small scale irrigation. RID had built over two thousand small reservoirs and weirs in northeast Thailand. Even though good sites are now harder to find, construction continues. In Thailand "small-scale" has been defined in terms of construction budget, with RID responsible for most projects costing more than about U.S. $40,000. The dividing line between small projects turned over to farmer operation and medium scale projects where the agency retains a management role has shifted over time, the maximum level for small scale project currently can go as high as $600,000. RID had been criticized for low levels of utilization of small weirs and reservoirs in northeast Thailand. FPSS project efforts in Thailand focused on the use of community organizers to facilitate participation in the design and construction of small weirs and reservoirs. This was based on the assumption that it would be easier to mobilize farmer interest and build a sense of ownership in new projects where farmers could influence design and construction.

The Farmer Participation in Small Scale Irrigation Project (FPSS) is now in its third phase. Using its own resources, RID has expanded beyond the initial FPSS pilot efforts in northeast Thailand and is now undertaking a national pilot project covering all regions of the county which involves hiring recent university graduates as organizers.

... FPSS project efforts in Thailand focused on the use of community organizers to facilitate participation in the design and construction of small weirs and reservoirs. ...

Beginning in the early 1980s the Indonesian Department of Public Works conducted a series of pilot projects and studies directed at improving participation in irrigation (Tobing 1989). The Department of Public Works was concerned that benefits from government construction assistance had been less than expected and hoped that if farmers' sense of ownership could be strengthened this would lead to better development and management of irrigation systems. Community organizers facilitated the participation of farmers in design, construction and management of small and large irrigation projects. University researchers documented the prevalence of traditional irrigation management institutions and the impact of government intervention. Agency officials took an active part in research on how government policy could strengthen the role of water user associations.

This series of pilot projects and studies helped build a foundation for the Indonesian government's decision to establish a policy to gradually turn over all systems irrigating less than 500 hectares to water user associations. Such systems irrigate more than 1.5 million hectares. Encouragement from donors to give a higher priority to operation and maintenance and budget constraints caused by declining oil prices played a major role in this policy shift. Beginning in 1987 the Department of Public Works began developing methods for carrying out turnover.

The International Irrigation Management Institute (IIMI) and an Indonesian non-government organization, the Institute for Social and Economic Research, Education and Information (LP3ES), assisted the Department of Public Works in developing a participatory process of design and construction of minor improvements to irrigation systems and strengthening of formal water user associations in Indonesia (Murray-Rust and Vermillion 1989, Judawinata 1991), Bruns and Dwi Atmanto 1992). By the middle of 1991 the government had transferred more than four hundred irrigation systems covering over thirty thousand hectares to water user associations.

... a participatory process of design and construction of minor improvements to irrigation systems and strengthening of formal water user associations in Indonesia ...

These efforts in Thailand and Indonesia have now been going on for long enough and on a wide enough scale to ask what progress they have made in improving participation. Are methods from the Philippines applicable in other countries? To what extent have relationships between agencies and farmers changed? What are the limitations of participation as a method for improving irrigation development? What directions seem most promising for future efforts?

Irrigation Organizers <-

A principal element of these projects has been the use of an agent trained and paid by the project to work with farmers and assist them in organizing and taking part in project activities. The most common pattern has been to recruit recent university graduates to carry out the role of organizer.

This was the approach pioneered in the Philippines. NIA mainly relies on temporary staff with a fairly high rate of turnover, although it has hired some organizers to become permanent staff. Various titles have been used for the organizer position including irrigation community organizer (ICO) and institutional development officer (IDO). Beginning in the late 1980's NIA began to use farmers as irrigation organizers as part of large projects in national irrigation systems (Angeles and others 1990). They work under the supervision of NIA watermasters and with support from Institutional Development Officers.

In Thailand, as elsewhere, there has been vigorous discussion about who should work as organizers. The FPSS project focused on using recent university graduates working at one to three new sites each. Initially the project also tried training existing staff of other government agencies, such as agricultural extension agents. This approach did not seem promising and did not continue. The project tried training staff from RID provincial offices to work as community organizers (COs). Some performed well but many saw the shift to working in the field as a demotion. Those who had families and homes in the provincial capital often did not spend as much time in the field as was expected. Currently, the main focus is on hiring recent graduates to work as temporary staff.

The effectiveness of recruitment, training and supervision of organizers has varied widely between regions. There have been problems with high turnover, slow payment and low morale. Some RID officials are trying to create permanent positions for organizers. This would offer hope for a career to at least some of the temporary staff. There also continues to be interest within RID in training existing field staff as organizers, perhaps zone men from medium and large scale projects. In one pilot project in Srisaket province RID is experimenting with the use of a farmer organizer, with promising results so far.

Another approach in Thailand has emphasized the ability of farmers to work directly with the agency without requiring an organizer as intermediary or facilitator (Uraivan Tan-Kim-Yong 1987). After many discussions a consensus seems to be emerging that some farmers in north and northeast Thailand are clearly capable of working directly with the agency, if agency procedures allow and encourage substantive participation. In other locations organizers can make a major difference in helping farmers to organize themselves and in improving communications between the agency and farmers.

In Indonesia the first projects exploring participatory approaches used recent graduates as organizers. However, when project funding ended, the positions disappeared. In planning for the turnover project, Public Works officials chose to train existing staff, mainly irrigation inspectors. Despite relatively low levels of education they seem to have performed well, as long as they receive adequate training and support (Helmi and Vermillion 1990). A major advantage of this approach is that the staff and their skills will stay within the agency.

Several feasible alternatives seem to exist for who should work to organize farmers. Recent graduates can work enthusiastically and effectively. Sri Lanka provides the clearest case of explicitly choosing to rely on recent graduates despite very high rates of staff turnover (Uphoff 1991). Indonesian experience shows that retraining existing agency staff is feasible. Using farmers as organizers seems to have considerable potential, though it may depend on local social conditions. There are dangers of undermining the spirit of local voluntary cooperation in irrigation if some people are paid to work and others are not. An outside organizer may be less necessary if farmers are already able to assert their interests while dealing with agency officials.

... Several feasible alternatives seem to exist for who should work to organize farmers. ...

Most of the debate about organizers has focused on factors affecting their performance in the field. While these factors are important, the success of institutionalizing new roles may depend more on the availability of people and budget. If graduates are plentiful, then using them as organizers may be feasible even with high turnover rates. However government hiring restrictions in many countries make establishing new positions extremely difficult. Uncertainty in payment and continuity of salaries and benefits creates major problems for retaining and motivating staff in temporary positions. Even if agency staff or farmers cannot perform as well as recent graduates, the lesser demands they impose on budgets may make it preferable to use them as organizers.

If there are already many agency staff, as in Indonesia, and budget constraints are important, then retraining is effective and helps to reorient the staff towards new ways of working with farmers. If agency staff and recent graduates are scarce, then farmers may the best candidates to work as organizers, when the situation seems to require organizers.

The question that needs further attention is who should train and supervise the organizers. In Thailand and Indonesia there has been a continuing dependence on outside organizations to handle many of these responsibilities. Khonkaen University has continued to play a major role in recruiting and training organizers. It is not clear how well agency staff will supervise and support organizers if they no longer have the stimulus of regular contact with university staff.

During the first phases of turnover, LP3ES staff acted as consultants to help with training, supervising and troubleshooting the activities of agency field staff. Turnover is still a new program, which requires different skills and procedures from the more routine construction, operation and maintenance activities which have occupied provincial irrigation service officials in the past. Annual training of provincial staff who train and supervise the organizers has played a key role in strengthening skills. It remains to be seen how well the provincial irrigation service staff responsible for managing turnover will handle the social aspects of this work once this special consultant assistance is no longer available.

One choice in Thailand and Indonesia would be to make outside support in training and consulting permanent. This is vulnerable to budget cutbacks, similar to the problem of using temporary staff as organizers. However it has the advantage of maintaining links between the agency and outside organizations such as universities or NGOs which can take a more independent view and give farmers' interests a high priority.

Another approach would be for experienced organizers to move into the role of trainers and supervisors. However, even in the Philippines with a long history of using the participatory approach, it has proved difficult to create many permanent positions for organizers. Most continue to be contractual employees. Attrition rates have been high.

Organizers need training, supervision and support. These take substantial time and usually cannot be done well if supervisors also have many other responsibilities. If organizers do not receive support they will not perform well and participatory methods will not succeed.

In all three counties it seems that agency policy, explicitly or implicitly, is that many of the people handling the duties of training and supervising organizers will have engineering backgrounds. If this is the case, then much more attention needs to be directed at ensuring that they will have the time and the skills needed to carry out this work. This requires training, including support for on-the-job learning, short courses and opportunities for some officials to learn about these issues in more depth. Non-government organizations may be able to play an important role in training agency staff and in assisting as consultants, mediators or in other support roles in dealing with communities.

... If organizers do not receive support they will not perform well and participatory methods will not succeed. ...

Even if farmers or agency staff are organizers they still need support from people with the background, experience and attitudes to support participatory approaches. This support can come from the immediate supervisors of organizers as well as staff with more specialized skills. There is a need for a cadre of former organizers or people with a strong capability in the area of social organization and positions. These people could then train and help supervise organizers of whatever variety. Institutionalizing such a cadre requires having positions and a career path, whether located inside or outside the agency. Without positions that recognize and reward the work of supporting organizers it will be difficult to attract and retain qualified people.


Participatory Siting <-

Agencies build many projects at locations where farmers have long experience building and managing irrigation systems. Even in previously un-irrigated areas, farmers have detailed knowledge of property rights, streamflows, cropping patterns and other factors, which is important in planning irrigation schemes. Participatory approaches offer an opportunity to combine this farmer knowledge with the technical and financial resources of an irrigation agency.

The Philippines pioneered the use of socio-technical profiles to gather background information about existing irrigation systems and other local conditions. Community organizers helped farmers to join together to negotiate with the agency concerning sites for structures and canals and other design decisions. Walking through proposed sites together enabled farmers and engineers to discuss design decisions in the field. Farmer committees helped to arrange for rights of way for canals and other structures. These innovations improved the flow of information and established a joint decision-making process for planning.

In the FPSS project COs helped make farmers aware of government plans. They helped farmers prepare for meetings with RID designers and survey teams. COs worked with farmers to develop water user groups and to resolve conflicts concerning land for structures, areas to be flooded, access roads and temporary construction camps. Specific methods such as using stakes to mark out the area structures will occupy have helped to make communication much clearer.

However, it still frequently happens that designs have been prepared before the COs begin work in the community. Sometimes this is because RID prepared designs in earlier years and then had to delay construction. At other times, design has been done informally in advance of budget availability in order to facilitate completing construction within the Cabinet mandated one-year period for design and construction of small scale water resources. Since one CO usually handles at most two new sites there have not been enough COs to cover all sites.

... farmers have long experience building and managing irrigation systems. Even in previously un-irrigated areas, farmers have detailed knowledge of property rights, streamflows, cropping patterns and other factors, which is important in planning irrigation schemes. ...

In Thailand, farmer participation has clearly led to better and more equitable siting of structures and canals. More farmers take part in decisions about where to put weirs and canals. Those who may lose land to flooding or structures are better informed and more adjustments are made, for example sharing losses by siting on property boundaries or making arrangements for compensation. However, there continue to be problems with outlets situated so that they cannot deliver water by gravity to the areas farmers had hoped to irrigate. The project has demonstrated the effectiveness of organizers. Agency officials understand and accept this role. However, for several reasons there are still many cases where organizers have not been able to help farmers take part in design decisions.

In Indonesia, agency field staff collect information using an inventory and a socio-technical profile. However, the written results from these exercises tend to go unused. They still may be useful as a means for the organizer to gain a more thorough understanding of the irrigation system. The main focus has been on the preparation of "farmer design requests" for irrigation system improvements. The organizer helps the farmers to prepare sketches and estimates of material required, in a prioritized list. Farmers discuss these with the technical design staff. The turnover project has institutionalized this approach on a routine basis within the large and nationwide turnover project, but so far the approach seems not to have spread further.

In both Thailand and Indonesia farmers' main priorities concerning design have to do with location. They worry most about what land canals and structures will occupy. They want to have losses of land distributed equitably and water easily accessible. These general ideas seem to hold across many different cases.

There has been substantial progress in participation in siting, deciding where to locate structures. Participation in siting does not particularly threaten agency expertise. It draws on local knowledge of property rights, water flows and soil conditions. It is usually relatively easy for the agency to recognize that farmers possess valuable information which they can contribute and to accommodate farmer requests. The more difficult questions concern design alternatives, the specific types of canals and structures used.

The social and physical conditions of farmer managed systems often differ from the conditions for which most agency standard structures were designed (Yoder 1992). Farmer-managed systems commonly have hilly topography, high runoff and sediment levels, uncertain water supplies and unstable slopes. While perhaps seemingly minor from an engineering point of view, innovation in design can be crucial in enabling participation and more effective local operation and maintenance.

... There has been substantial progress in participation in siting ...

In determining the type of structures it seems more difficult to combine agency and farmer priorities. There is a strong tendency for agencies to continue to rely on existing standard designs and not to feel that there is much need for innovation. The main influence of participation in the Philippines seems to have been on siting decisions, with less influence on the actual types of structures and canals used.

In Thailand there has been little progress in terms of design alternatives. The agency added low stoplogs to the design for weirs. These allow some flexibility in setting the level of water backed up behind the weir. Designers often added pedestrian bridges to weirs at the request of farmers. However, progress, in terms of changing the set of design options open to farmers, has not been great.

The results of some other projects in Thailand have shown that substantial potential exists for reducing costs and improving performance through design innovation (Mayson 1984, Bruns 1990). RID's accumulation of over fifteen years of experience with small scale irrigation should provide a good basis for reassessing many of the design standards which RID established earlier on the basis of very limited information. However government procedures offer little room for testing new approaches, especially ones that might fail conspicuously. The pilot projects discussed here have focused on institutional aspects and mainly been linked with agency officials concerned with operation and maintenance, and so have offered little opportunity for design reassessment, innovation and learning from experience. Since irrigation projects are usually relatively large and expensive there is little activity in this area by non-government organizations and so little scope from learning from the experience of others.

In Indonesia, at farmers' suggestion, some canals are only partially lined according to where water losses are largest. This draws on farmer knowledge to economize on materials. The project has built some proportional division structures and the project design guidelines now include such structures. These fit with existing designs and principles used by farmers to manage water (Ambler 1990). From the farmers point of view they are more intuitive and it is easier to observe whether water is being fairly distributed. There has been little change in the types of drop structures, overflows for disposing of excess water and little progress in clarifying criteria for deciding where to line canals.

In both countries farmers have a greater role in choosing what they want from the menu of potential improvements and where to put them. However, there has been much less innovation in what is listed on the menu. The case for using farmer participation to improve siting decisions seems well established and accepted by many agency officials. Innovation in design of structures and canals for farmer management has had less progress and deserves further support.

The December 1989 workshop on Design Issues in Farmer-Managed Irrigation Systems (Yoder and Thurston 1990) largely emphasized the process of participation in design rather than outcomes in terms of innovative structures or a broader menu of choices from which farmers and design engineers could choose and adapt. The workshop and handbook on design for mountain irrigation should help disseminate more examples of design innovation (Yoder 1992).

... Innovation in design of structures and canals for farmer management has had less progress and deserves further support. ...

Design innovation may help develop simpler, less expensive structures, which make better use of local materials, and a more selective, efficient approach to improving irrigation systems, blending the advantages of outside technical expertise and local experience. Changes in this direction are also likely to help increase the scope for local contributions to construction.

Contributing To Construction <-

One of the most fundamental elements of the participatory approach in the Philippines was that government policy mandated that NIA was supposed to recover costs. Poor payment rates made NIA interested in an approach that promised to make farmers more satisfied. Given a policy framework intended to support farmer management, a major missing link was the lack of institutions through which farmers could take part in planning and construction. Once participatory institutions were established farmers had a very strong basis for having a say in construction, since they were paying. There were implicit subsidies, since farmers did not have to pay interest and were only responsible for direct costs. However this was less important than the fundamental principle that farmers were responsible for repaying a large proportion of costs. In participatory construction of communal systems farmers were active in locating suppliers and monitoring the use of materials. They tried to use their own labor when this would help to reduce the amount they would have to repay.

In Thailand, the government heavily subsidizes construction of small weirs and reservoirs. Cabinet policy requires that landowners and local communities provide land without compensation. The government builds headworks and expects farmers to build canals. The problem with this division of labor between farmers and government is that often farmers do not build canals and little water reaches fields.

Since farmers do not have to repay the construction cost they have no incentive to economize. RID wants to avoid having structures damaged by flooding and prefers to design structures that do not require much active operation and maintenance. The result is a tendency to design structures that are larger and more expensive than necessary (Bruns 1991).

... One of the most fundamental elements of the participatory approach in the Philippines was that government policy mandated that NIA was supposed to recover costs. ...

When the FPSS project began it was hoped that it would be possible to encourage greater involvement of water users in construction. This turned out to be much more difficult than anticipated. The cost of labor is small compared to the total cost of a weir or reservoir. Site engineers are under pressure to complete construction quickly. They usually already have teams of foremen and skilled technicians who move from job to job. RID pays a government mandated minimum wage that is much higher than the local wage rate, so many people seek employment. Field engineers use unskilled labor positions as a way to partially compensate farmers whose land is used in construction. They also often allow village headmen to select workers as a way of facilitating cooperation.

Projects usually only have the potential to provide irrigation for a small fraction of the local farmers. Those who will not get irrigation feel it would be unfair if the irrigation beneficiaries also get the benefit of well-paid construction work.

In northeast Thailand there was little scope for using local labor and materials in construction. The undulating topography and sandy soils of northeast Thailand have few rocks. Most construction is of reinforced concrete. The relatively high minimum wage rates paid by the government mean it is usually much cheaper to move earth using machines. For large earthworks and excavation RID usually hires contractors who use heavy equipment.

The FPSS project mainly worked with the O&M division of RID. No one associated with the project was in a position to require site engineers to give members of the WUA preference in hiring. Given the scale of construction it did not seem as if working as unskilled laborers would actually do much to build farmers' sense of ownership of the structures. Because of all these factors the project staff decided not to push harder for participation in construction, but instead to emphasize participation in other activities. The project did advocate using an open lottery approach to selecting unskilled workers, rather than having all recruitment through personal connections or arbitrary factors of who applied first.

The question of farmer contributions was a subject of considerable discussion in the turnover project in Indonesia (Bruns and Dwi Atmanto 1992, Murray-Rust and Vermillion 1989). The original project design made no explicit provision for local contributions to construction. However, Public Works did encourage purchase of local materials and recruitment of local labor. Much construction used local rock. In some areas local sand was also of adequate quality. Rehabilitation and re-alignment of canals usually required digging and moving earth which unskilled laborers could do. Pre-construction meetings among farmers, Public Works and contractors were a very useful way to arrange for use of local resources and avoid problems that often occurred without such meetings.

A policy that any local unpaid contributions had to be clearly separate from government-funded works complicated and discouraged voluntary local contributions (Bruns and Dwi Atmanto 1992). Since farmers did not have to pay a share of construction costs they often tended to take a wish list approach to planning, asking for many improvements. Despite strong suggestions from IIMI and the World Bank, Public Works officials did not agree with instituting a fixed requirement for local contributions.

Requiring farmers to contribute to construction can have a very good influence not just in terms of an abstract "sense of ownership" but in very concrete terms of reducing costs and improving the effectiveness of government investments. Local contributions can amplify the impact of government investment. However, one of the lessons from these two projects is that project planners should build in requirements for contributions from the beginning. Trying to incorporate requirements for contributions after the project was already being carried out ran into many obstacles.

Political factors and attitudes about farmers' ability to pay may make it difficult to require full repayment by farmers. If there is large scale and effective government investment in the agricultural sector, then many benefits go to consumers in terms of lower relative food prices. This is another justification for paying part of the costs of irrigation from government funds rather than requiring irrigators to pay all costs.

... project planners should build in requirements for contributions from the beginning. ...

However, if farmers only pay a percentage, even as low as 10%, this can have an important influence in encouraging efficiency and legitimizing farmers' role in decision-making. The experience of the Philippines and Indonesia suggests this should be possible not only in farmer managed systems but even within the context of improving tertiary canals in large and medium sized schemes. In contrast to typical subsidized approaches where neither agency nor farmer has incentives to accurately weigh costs and benefits (Repetto 1986), this increases accountability to farmers and encourages farmers to consider what will be the most worthwhile way to make use of government assistance. If farmers have to pay for construction costs this may encourage them to be more careful about operation and maintenance of systems once construction is finished.

Moral Hazards In Maintenance <-

Government agencies suffer from severe limitations in their capability to assist operation and maintenance of small scattered irrigation systems (Coward, Johnson and Walter 1988). Water distribution is highly complex, requiring timely responses to diverse local conditions which local people usually understand well. Operation is closely linked to farmers' interest in providing enough water to insure good yields. Government help may be useful in promoting greater equity in water distribution, though this needs careful attention to existing local water rights. The advantages of government seem to lie more in the area of maintenance rather than operations (Moore 1988). Maintenance can benefit more from technical expertise and is more prone to neglect.

The conventional wisdom seems to be that if farmers feel they truly "own" a system they will take good care of it. This view persists despite the obvious evidence of frequent poor maintenance in villages of both individually owned private property and common property with clear local ownership, such as meeting halls and religious buildings. A pattern of neglect followed by repair and restoration is much more frequent than routine preventive maintenance.

This is not to deny that ownership is an issue. It is important. Institutional changes are probably a necessary part of a solution to the problems of maintenance but seem unlikely to be sufficient in and of themselves.

Given the level of government maintenance under current conditions it would not be difficult for farmers to do better. However, there are also plenty of problems with farmer O&M. This is particularly true when it concerns maintenance of expensive concrete or mortar structures, rather than temporary earth, wood and bamboo structures.

... A pattern of neglect followed by repair and restoration is much more frequent than routine preventive maintenance. ...

It sometimes seems that discussions about farmer-managed irrigation systems and turnover of management assume the possibility of returning to a (supposed) previous idyllic state of local O&M. Farmers in Thailand, Indonesia and the Philippines have developed institutions to mobilize resources to maintain and repair irrigation systems built with traditional materials. These institutions have sometimes developed under circumstances in which outside assistance was not a realistic option. This may have been because there simply was no possible source of assistance or that it would be too slow to help before farmers lost crops. However, history shows that farmers have usually been eager to obtain as much outside assistance as possible to repair and improve their irrigation systems (Maas and Anderson 1978).

In the context of irrigation there are many ways to trade off between routine maintenance and major repairs and rehabilitation. If farmers have to pay the full costs of routine repairs while the government pays fully for major repairs and rehabilitation, then farmers have a strong incentive to delay action until problems qualify for government payment.

Maintenance, repair and improvement face the problem economists refer to as moral hazard. The classic example of this is insurance. If something is fully insured, then the owner has less incentive to prevent losses. The conventional remedies are that there should be standards to prevent negligence. Insurance should not cover all losses. Instead there should be deductible amounts and proportionate shares paid by the insured so that they take costs into account.

This is not just an abstract issue but an accusation many irrigation agency officials make against farmers: that they wait and hope the government will make repairs. On the other hand, the usual assumption seems to be that farmers are not capable of paying for major repairs and rehabilitation and so government should pay for most of these costs.

Thai policy is that farmers should pay the full cost for repairs under $200. The government pays all costs for repairs costing more than $6,000. Repairs costing between $200 and $6,000 should, in theory, be funded through a tripartite approach with one part from farmers, one part from RID funds and one part from provincial government funds. Experience has been that the tripartite approach is basically unworkable. The complexity of coordinating budgets and long lead times involved, and the competition for funds from other activities mean that few repairs are carried out under this approach. In the view of agency officials the result is a tendency to delay until major repairs are needed.

The principle promulgated in Indonesia has been that farmers should pay for repairs that are within their capacity while the government will pay for repairs and improvements that are beyond farmers' capacity. However, there are so far no clear criteria for putting this policy into operation. Thus government responses to farmer requests seemed largely to be shaped by the availability of funds for natural disaster assistance or other sources, and the degree of support which requests may receive from government officials.

In both Thailand and Indonesia the process tends to mean that government officials receive requests and have little clear basis for rejecting them. Politically it is much easier to accept requests rather than turning them down. Villagers are led to hope they may receive help but get little feedback. Their hope, that the government may help, discourages them from making thorough repairs on their own. Instead their incentive is to make stopgap repairs sufficient to keep the system working.

Farmers can and will try whenever possible to obtain outside assistance for repairing and improving irrigation systems. Given the political appeal of conspicuous construction of irrigation works it is unlikely that any policy changes will shut off all avenues for government subsidies to irrigation. A more realistic goal is for users to pay operations costs as much as possible. For repairs, rehabilitation and improvements they should pay a share of costs. During discussion, some Philippine officials felt that in their experience the policy of requiring farmers to repay costs of construction, including repairs and rehabilitation, had encouraged farmers to be more diligent about maintenance.

Procedures for requesting assistance can be more transparent, reducing the uncertainty that often discourages farmers from solving problems on their own. However, it is likely that the government assistance in irrigation will in the future become more politicized rather than less. Democratic aspirations are spreading. Farmers and other local people have better educational qualifications and better linkages with other social groups making them more willing to challenge the actions of technical agencies. Within such a politicized process technical assessments and benefit-cost evaluations can have an effect but will be far from the only influence.

Current policies regarding government assistance can create perverse incentives, discouraging local action to maintain and repair irrigation systems. Clarifying the division of responsibility and the framework for joint action by government and farmers can reduce uncertainties and avoid misunderstandings. Requiring local cost sharing, even for relatively large repairs and improvements can strengthen the incentives to take adequate preventive measures.

... Current policies regarding government assistance can .. discourage local action to maintain and repair irrigation systems. ...

The problem of moral hazard should be seen in proportion. The delays, complications and uncertainties involved in obtaining government help already form a strong incentive for farmers to carry out repairs on their own where possible. The risk of losing crops is a strong motivation for farmers to complete at least the most urgent maintenance and repair activities. The government can assist local efforts by providing technical assistance to help farmers identify maintenance needs and plan how to solve problems. Extension officials, non-government organizations, community development workers and others could play a valuable role in helping farmers understand where they can seek assistance and what types of assistance may be available. Providing technical advice, without providing funding for construction, should be a legitimate and important role for irrigation agencies. This could be a promising area for future research and institutional innovation in irrigation development.


Research And Action: Fragile Linkages <-

One of the most exciting aspects of NIA's participatory approach was the way in which it linked social science research with field activities in a learning process approach to institutional innovation (D. Korten 1980, F. Korten 1987). Researchers observed and recorded the activities of the first irrigation community organizers. They prepared a monthly report, which the project working group used as a basis for improving the participatory approach.

The working group brought together senior agency officials, researchers and donor staff to discuss developments in the field and consider ways to improve the approach (Korten 1988). This method was particularly useful for identifying the obstacles that hindered implementation of participatory policies. Without solutions for these micro-level problems, it would be difficult to implement a participatory approach, however good the macro-level policy. Process documentation research helped to bring these problems to the attention of agency officials who were senior enough to make changes. Researchers and donor staff helped the agency be creative about considering new ideas.

This type of learning process approach offered a way to escape many of the problems that make much conventional research on development projects wasteful and irrelevant. Commonly researchers focus on producing a final report, which usually appears much too late to be of any use. Process documentation research emphasized monthly reports, quickly reporting while there was time to respond constructively to problems. Much conventional research emphasizes questionnaire surveys and other quantitative methods, which often are not relevant to the issues that affect implementation (David Korten 1980). Process documentation stressed qualitative research focused on problems. Researchers were able to be flexible about gathering information from many sources and following up new developments. The principal elements thus were a problem oriented approach, quick reporting of information and responses from senior officials.

IIMI's Philippine field office has conducted an extensive methodological study using workshops and questionnaires to check the representativeness of the information gathered from process documentation sites (Wijayaratana 1991). The results suggest that process documentation information is quite valid despite relying on only a small number of sites.

... This type of learning process approach offered a way to escape many of the problems that make much conventional research on development projects wasteful and irrelevant. ...

In the FPSS project, social researchers from Khonkaen University's Research and Development Institute helped the university's Water Resources and Environment Institute to plan and carry out the project. The project field section was in charge of hiring, training and supervising the activities of the community organizers and coordinating their activities with RID. The research section gathered information on RID policies and procedures, studied sites with COs and comparative sites without organizers. Officials from the RID Region IV office attended working group meetings. Initially meetings served largely as a way for university staff to learn about RID policies and procedures. Over time working group meetings took on more of a problem solving role. However, in the later phases project staff stopped holding special working group meetings and instead used routine meetings with RID staff and special occasions such as seminars and training as the means to discuss issues and refine methods.

There was a high rate of turnover among the social researchers involved in the project. In part this was due to external factors including conflicts with other commitments and opportunities to study or join other projects. It was also linked to the ambiguous status of research activities in a project where field activities clearly took the leading role.

During the first several years, the project had only limited contact with senior RID officials. However, university researchers involved in several irrigation projects, including FPSS, joined with senior RID officials to form a network, the Thailand Research in Irrigation Management Network (TRIMNET) with support from Winrock and the Ford Foundation. This provided a forum for discussing issues of common concern and channeling suggestions from researchers to agency officials. TRIMNET also provided small grants to fund research related to irrigation. The network members held meetings approximately twice a year in various parts of the country. Over time RID officials became more supportive of TRIMNET as they saw it could help them in finding possible solutions to problems, or at least helped show concern about trying to find solutions. For FPSS staff TRIMNET helped provide stronger links with RID. This was particularly important as the RID official who had first sponsored activities retired.

In Indonesia the series of projects beginning in the early 1980s brought together Public Works officials, university researchers and non-government organizations, including LP3ES. A team of Public Works officials, staff of USAID and the Ford Foundation, and staff of LP3ES monitored activities of the High Performance Sederhana Irrigation Systems Project. In the Madiun project in East Java, Satya Wacana university carried out intensive process documentation and took part in working group meetings similar to the Philippine model. University researchers and LP3ES also joined with agency officials to carry out research on policy towards water user associations and traditional irrigation.

In the turnover project, International Irrigation Management Institute staff took part in the working group meetings, which helped develop the initial framework for the project. IIMI then conducted research on implementation. IIMI researchers endeavored to present their findings and recommendations through written reports and memos and in working group meetings and seminars.

The situation in the turnover project was complicated by having three parties: the agency, IIMI and an Indonesian NGO involved in developing the methods for turnover. Agency officials were sensitive to any written reports or formal statements during meetings that might reflect badly on the agency. Meetings and informal contacts did provide a means for communications, particularly at the provincial level.

IIMI's activities finished at the end of 1989. During 1990 and 1991 the Institute for Agribusiness Development (PPA) was responsible for documentation research on the project. In West Java and West Sumatra the universities that had taken part in the research with IIMI continued their cooperation with the provincial irrigation service in each province. In other provinces there were numerous problems concerning the relationship between researchers and agency staff and the timeliness and accuracy of reports. However, agency officials continued to state that they wanted researchers to provide a record of activities, regardless of whether findings were positive or negative.

One conclusion, which arises from comparing these experiences, is that it is much harder than expected to establish an effective learning process involving researchers. It is not simply a matter of providing funding and having periodic contacts between researchers and agency staff. Maintaining a good relationship between researchers and agency staff may be a necessary condition but is not sufficient. Researchers need to be able to exercise a degree of intellectual leadership, without upsetting agency officials or seeking to impose the researchers' recommendations. Agency officials want to learn more about problems, though they prefer reports that they can respond to constructively and which do not cause too much loss of face.

Full scale process documentation is very demanding in terms of both field researchers and rapid analysis and preparation of reports. Most researchers have tried to adopt some of the principles of process documentation but not the whole method. The Institute for Philippine Culture itself has continued reliance on researchers working full time at a few sites but has expanded coverage once the researchers have obtained an in depth knowledge of the initial sites. They have tried make a greater separation between very brief reports targeted at decision-makers and longer academically oriented reports (Romana de los Reyes, personal communication). Others in the Philippines also stress the need for extremely brief reports for agency staff, i.e., one page or so (Robbie Laitos, personal communication). Otherwise, most agency officials will not read them. In all three countries what seems to be much more important is spoken communication, within and outside formal meetings.

In both Thailand and Indonesia working meetings held away from regular offices were an effective way to arrange things so that senior officials could spend substantial amounts of time engaged with project issues. In Thailand and the Philippines regional level working groups were able to move ahead even without much assistance from more senior levels.

However, for all three countries, there seems to have been a large reduction in the integration between research and action over the long term. Research activities have continued but have been much less well linked with implementation. One explanation for this could be a process of maturation. As time goes on, the focus shifts, from learning to make new methods effective and efficient on a small scale, towards institutionalizing the methods and expanding them to larger scale implementation (David Korten 1980). It may be that the process of expansion has less to gain from documentation research. Officials may already be well enough aware of problems and issues and not feel as much need for help from researchers in analyzing them.

... there seems to have been a large reduction in the integration between research and action over the long term. ...

However, this decline in the level of integration between research and action has gone on even where new ideas are being piloted on a small scale. This has been the case for pilot turnover of 500 hectare systems in Indonesia and in the use of farmers as irrigation organizers in the Philippines and Thailand. Therefore some other factors seem to be at work.

The key difference here seems to lie in the role of the donor as facilitator and the level of involvement of senior agency officials (Korten 1988: 82-83). In earlier work in the Philippines and Indonesia the role of donor as facilitator involved not just attending meetings but talking individually to prepare for meetings, helping frame issues in useful ways, encouraging participants to pay attention to reports and facilitating consensus.

Because of workload, changes in personnel, changes in source of funding and other factors this donor role as facilitator has not been as strongly present in more recent activities. In Thailand the project had limited access to senior officials and only intermittent involvement of donor staff. In Indonesia the increase in project funding for O&M meant it became much more difficult to gain access to senior officials. Researchers and consultants can take on some of the responsibilities in managing a learning process approach, but much of the essential role of facilitator may be almost impossible to delegate.

The difficulties of the various attempts to set up learning process approaches involving researchers raise serious questions about whether such approaches are replicable unless the donor is willing and able to invest a large amount of time in the process. Without the direct involvement of the donor it is much harder to get time from senior officials who have many competing activities.

What seems to have happened is a cross national example of the classic pilot project problem. Something succeeds under favorable conditions after strenuous effort. Attempts are made to replicate this, usually under less favorable conditions. At the same time a lower level of resources is invested. It is somewhat less than surprising that the results are often below expectations. This process seems to have occurred with attempts to use learning process approaches incorporating researchers as part of projects to promote participatory approaches to irrigation development.

This does not mean that the research work is completely wasted. Lower level working groups may use it, as happened during the earlier phases of FPSS and in some of the work done by IIMI with NIA in the Philippines. Such groups can address operational issues but are less able to deal with more basic policy constraints. Meetings primarily concerned with implementation, training sessions or the preparation of manuals and guidelines also provide forums for discussion in which information from the field can be introduced and analyzed. Conventional final reports may still have influence, though usually after long lags. Informal contacts may also allow researchers to share information, though in a much less structured form. However, this is quite different from the original model of a learning process approach involving researchers in a high level working group. The question is not one of whether to use part-time or full time researchers in the field. The issue is how and when senior level officials receive and respond to research results.

... questions about whether such approaches are replicable unless the donor is willing and able to invest a large amount of time in the process. ...

The links between research and action are fragile. A learning process approach that involves researchers potentially offers large gains in speeding up learning compared to the waste and long lags involved in conventional research or the limited information available if specialized researchers are not included in the project. However, there seem to be several prerequisites. One is strong support from senior levels in the agency. Another is diplomacy on the part of researchers, to cultivate a good working relationship with the agency and present findings in ways that facilitate a constructive response. Most crucial seems to be support on the part of the donor, in terms of ample time spent facilitating a learning process.

Research is still worth conducting if conditions are less favorable but expectations should be lowered accordingly. In particular the amount of time spent by policy level staff of the donor and agency will be a crucial factor in determining the extent to which research findings can enhance implementation in a rapid way.

Research is not the only need for a learning process. Particularly as potential solutions are identified and a project evolves much work is needed in figuring out how to institutionalize new methods. This includes such things as new policies, manuals and training programs. Researchers can contribute to this process, as can agency officials, donors and NGO staff.

Learning process approaches can be undertaken without involving researchers. This may seem easier and more comfortable for some agency officials. However this loses the valuable contributions researchers can make in gathering information about conditions in the field, analyzing problems, looking at things in a larger context and being creative in helping search for potential solutions. A learning process without specialized researchers involved is likely to be less productive and have fewer opportunities for making major progress. The problems of linking research and action for institutional innovation are part of larger issues concerning the flow of information in irrigation agencies and mechanisms for monitoring, evaluation and accountability.

Evaluation: Missing In Action <-

Perhaps the most telling indicator of the construction bias in irrigation development is the lack of systematic evaluation. Not only is timely information of the sort provided by process documentation and similar methods rare, but there is little information in general about the performance of completed irrigation systems. The Philippines has an elaborate system for advance assessment of projects. NIA officials feel that this has given them a better basis for screening out requests for projects at sites that are technically or economically unsuitable. However, there seems to have been little systematic follow-up by NIA of how actual performance compares with the initial estimates. There seems to be a general acknowledgment that actual irrigated areas are often much smaller than planned areas. Payment rates for amortization in communal systems and irrigation service fees in national systems do provide one indicator with some relationship to farmer satisfaction.

In Thailand the operations and maintenance division of RID has made some attempts to monitor utilization of sites. These do not seem to be integrated into any attempt to improve the design and construction process.

In Indonesia some information is routinely collected on irrigated areas and yields. However, the accuracy of the information is uncertain. It is not used as part of any process to improve planning or to diagnose sites needing government assistance in O&M.

In all three countries use of evaluation results to improve construction or O&M is not a routine part of agency operations. Information collected during monitoring activities does not seem to be used for improving planning of construction or for targeting government assistance to improve O&M. The orientation continues to be towards construction with little or no feedback regarding performance.

... Perhaps the most telling indicator of the construction bias in irrigation development is the lack of systematic evaluation. ...

Two conclusions flow from this. One is that more evaluation would be useful. Some has been done. In the case of northeast Thailand much of it consists of case studies at model sites, which are not representative of general conditions (Bruns 1991:15). In all three countries the large amounts of funding from bilateral and multilateral donors seem to have brought little evaluation of what benefits farmers receive from these massive investments.

Internal and external evaluation both have a role. The innovative activities of NIA's participatory approach have been evaluated by researchers from the Institute for Philippine Culture who took part in developing the approach (de los Reyes 1988). They found that performance was much better at sites that used the participatory approach. The credibility of this finding would be much higher if there were a matching study by independent researchers who had not been associated with implementing the approach.

FPSS researchers are currently examining the impact of participation in sites where the project worked earlier. The standard argument for participation in design and construction has been that if farmers have a greater role in the design and construction of irrigation works, then they will feel that the irrigation system belongs to them and therefore will operate and maintain it better. The justification for this assumption is the example of farmer built irrigation systems where farmers do take a very active role in operation and maintenance. However there are many major differences between large agency managed systems and farmer managed systems which make this assumption questionable (Hunt 1988).

COs in the FPSS project seem to have had some impact in stimulating farmers to build canals and to think about how to distribute water more equitably. Beyond this it is not clear how much impact the participatory approach has had on operation and maintenance.

The next phase of LP3ES consulting for the turnover project will work on mechanisms for monitoring the project, with an emphasis on methods for self-assessment by water users associations.

External evaluation alone will never be enough. Structural mechanisms must encourage attention to performance. Requiring farmer repayment of construction costs is one method, as is the institution of irrigation service fees. These have helped to reorient NIA, although there are still high levels of subsidy involved. However, such fees are politically difficult to establish and maintain. Other mechanisms would also be useful. Turnover of responsibility to farmers is clearly one of these. This is linked to issues of decentralization and governance. As long as centrally funded technical agencies with little accountability dominate irrigation development it is likely to continue to be oriented towards construction with only weak incentives for improving the performance of investments.

Decentralization And Empowerment <-

The participatory approaches to irrigation development discussed here have essentially concentrated on ameliorating the consequences of centralized intervention by technical line agencies in irrigation. Bilateral and multilateral funding has helped build up such centralized agencies. The global shift towards democratization should encourage exploration of new alternatives.

One choice is to simplify the design process to the point where it requires little or no professional engineering work. This was one aspect of the People's Volunteer weir program in Thailand (Mayson 1984, Bruns 1990, 1991). A flexible, standardized design adapted to the specific conditions of the region only required adjustments in terms of the width and height of the weir. Technicians based in district offices could prepare the design. Farmers working under the supervision of a trained foreman could build the weirs themselves. Similarly for rehabilitating small tanks the Department of Agrarian Services in Sri Lanka evolved a set of designs that now require little engineering work. Such simplification makes it possible for many agencies, including some non-government agencies, to assist in irrigation development.

Turnover of all or part of irrigation systems to water users can be a part of decentralization and empowerment (Vermillion 1991). Policy in the Philippines, Thailand and Indonesia now largely aims at having farmers manage and operate small irrigation systems. The Philippines has begun to turn over responsibility for managing parts of large irrigation systems. In the Philippines the pace of turnover within national systems largely follows the rate at which existing staff retire or transfer.

Some reforms in management in Thailand and Indonesia are moving towards giving water user associations greater responsibilities and authority within larger systems. Thailand has many medium-sized irrigation systems which farmers could probably operate with a very modest amount of technical advice. Indonesia has some farmer managed irrigation systems irrigating thousands of hectares. Probably farmers could manage many systems larger than the current 500 hectare ceiling for turnover.

There is considerable potential for continued turnover, if it can be done either with strong support from policy-makers and donors or else carried out in ways that are not too threatening in terms of agency budgets and staffing. It seems likely that the government can turn over most responsibility for operations to users, with government remaining as a mediator and if necessary enforcer of last resort to help encourage fairness. However, turnover is a transitional process. The more fundamental question is the framework within which management occurs after turnover.

Another way to increase local roles in irrigation is to shift the source of funding. Measures of this sort include direct grants to villages or subdistricts, which they may spend on irrigation, among other activities. Such block grants from central government funds still maintain dependency and encourage intervention by central government agencies. However, they do give local officials stronger say in the process. In this case the role of the technical irrigation agency can shift from one of controlling funds to one of providing technical services. Requiring greater transparency in budget allocation can create a more open process than the usual patterns of agency intervention. Transparency can also help to counteract tendencies towards corruption and abuse, which exist under either agency or more decentralized investment.

... Bilateral and multilateral funding has helped build up such centralized agencies. The global shift towards democratization should encourage exploration of new alternatives. ...

The requirement that farmers repay costs in the Philippines gives farmers a strong voice in construction. Thus cost sharing is an important mechanism for decentralizing decision-making. It allows farmers a much stronger basis for saying no to agency intervention, unless it is done in ways acceptable to farmers.

The Philippines is currently facing the questions of how to decentralize responsibilities in irrigation because of the recently passed Local Government Code. This seeks to establish much stronger local government units, which would directly retain a large share of any taxes they collect. This raises the question of whether some of NIA's current responsibilities would be transferred to these local government units.

Thailand has a long history of mostly ineffective attempts to decentralize power (Anderson 1985). The present government has given funds to the provinces as a step towards allowing more decision-making at the provincial level. The tambon (subdistrict) development fund in the middle seventies was an important innovation in giving power over budget to local units. However subsequent governments have been reluctant to entrust more power to local people and their representatives at this level and so the impact of this approach has been diluted rather than extended.

Where external funding is important, one part of a more decentralized approach will be to develop mechanisms through which lower levels of government can have better access to funding and responsibility for repaying costs. This would apply both for foreign loan funding and to cases where regional or local level authorities are responsible for raising money to cover construction costs. Requirements for following government accounting procedures should not prohibit decentralization in management and financing.

Further movement in this direction would involve giving greater taxing powers to local government units. This would give representative local authorities the means to carry out O&M and minor improvements on their own. With additional institutional changes they could use tax revenues to repay bonds, which would fund larger investments.

Where farmers have the resources the government should provide an enabling framework for them to work together to develop irrigation systems on their own. This might require something similar to the legislation allowing for irrigation districts in the U.S. This would specify procedures for holding a vote to make a binding decision to borrow money to develop the irrigation system. Once the decision had been made then a local authority would have the necessary legal powers to carry out the work and collect payments from farmers. This type of governance structure would help farmers to be able to help themselves. It could operate in parallel with government direct assistance. It would be useful in cases such as high value crops (for example, onions or garlic) where farmers are prepared to invest substantial amounts in improving irrigation but could do this more effectively collectively than individually.

A purely free market version of this would require that farmers put up their own land or other assets as collateral. Alternatively the government could provide partial guarantees and enforce sanctions.

... Where farmers have the resources the government should provide an enabling framework for them to work together to develop irrigation systems on their own. ...

In a climate of budgetary austerity such an institutional framework would provide a way for irrigation development to continue where potential payoffs were attractive. Even where the government subsidizes irrigation it would provide a fast track process for areas where the economic returns to irrigation are high. This could allow government efforts to concentrate on poorer areas. It would give farmers an alternative to waiting for government help. Not creating such a framework is equivalent to endorsing only agency managed investment in irrigation.


The approach to participation exemplified in the projects discussed above evolved in the context of centralized agencies and a predominance of external funding. More participatory approaches have been institutionalized within national irrigation agencies. Many of these were accomplished under authoritarian regimes where there was little scope for more democratic, decentralized approaches to the management and development of local resources. However, in changing political environments it is past time to ask whether priorities need to be reconsidered and new goals set. Answers are likely to lie not in a simplistic devolution of responsibility to the lowest level but in figuring out how best to combine the roles of farmers, people's organizations, non-government organizations, technical agencies and local government.

There are many measures available for encouraging a more participatory, decentralized and democratic process of irrigation development. Much of the work of operating irrigation systems can be turned over to farmers. Rather than channeling all funding through a single bureaucratic quasi-monopoly many types of investment can be opened up to more self-reliant efforts. Simplification of designs can allow more organizations to be active in irrigation development. Local cost sharing and more transparent procedures can help promote equity and honesty in decentralized decision-making. Financially capable farmers can be given the means to borrow and invest in irrigation development on their own initiative.

... There are many measures available for encouraging a more participatory, decentralized and democratic process of irrigation development. ...

Progress has already been made in improving irrigation through greater participation. Further efforts can develop institutions for financing and managing irrigation that increase accountability and reinforce local participation in design, construction, operation and maintenance of irrigation. There is much scope for encouraging more dynamic, diverse and democratic institutions for the development of irrigation.



A shorter version of this paper was published as "Promoting Participation in Irrigation: Reflections on Experience in Southeast Asia." World Development v. 21, No. 11 (November 1993): 1837-1849. The writing of this paper was supported by a grant from the Ford Foundation to the Water Resources and Environment Institute of Khonkaen University. I wish to express my thanks for comments on earlier drafts of this paper by Sanguan Patamatamkul, David Thomas and Frances Korten. Views expressed in the paper are the responsibility of the author. Comments concerning this paper are welcome and should be sent to Bryan Bruns 39/1 Ban Daun Ngeun, A. Pong, Phayao 56140 Thailand or Internet: 70702.1154@compuserve.com.

May 23, 1992

Comments and questions regarding this paper should be directed to Bryan Bruns at 39/1 Ban Daun Ngeun, A. Pong, Phayao 56140 Thailand, or: BryanBruns@BryanBruns.com

Next -> Chapter 9: Distributed Information Systems for Farmer Managed Irrigation


Ambler, John. 1990. "The Influence of Farmer Water Rights on the Design of Water-Proportioning Devices." In Design Issues in Farmer-Managed Irrigation Systems, ed. Robert Yoder and Juanita Thurston. 37-52. Colombo: International Irrigation Management Institute.

Anderson, Edward. 1985. The Political Economy of the Tambon Development Program. Master's thesis, Cornell University.

Angeles, Honorato L., Ireneo C. Agulto, Eduardo G. Marzan Jr., and Pastora S. Coloma. 1990. The Farmer As Irrigators' Organizer: Lessons from the BPEPIS Experience. Irrigation System Management Improvement Program, Department of Agricultural Engineering, Central Luzon State University. ISMIP Monograph.

Bruns, Bryan. 1990. "Design for Participation: Elephant Ears, Crocodile Teeth and Variable Crest Weirs in Northeast Thailand." In Design Issues in Farmer-Managed Irrigation Systems. ed. Robert Yoder and Juanita Thurston. 107-119. Colombo: International Irrigation Management Institute.

Bruns, Bryan. 1991. The Stream the Tiger Leaped: A Study of Intervention and Innovation in Small Scale Irrigation Development in Northeast Thailand. Ph.D. Dissertation, Cornell University.

Bruns, Bryan and Sudar Dwi Atmanto. 1992. How to Turn Over Irrigation Systems to Farmers? Questions and Decisions in Indonesia. Overseas Development Institute Network Paper 10 (April 1992): 2-23.

Coward, E. Walter, Jr., Robert L. Johnson, and Michael F. Walter. 1988. Other Channels: Improving Public Policies and Programs for Small Scale Irrigation in Asia. Cornell University. Working Paper, Irrigation Studies Group. Ithaca, New York.

De los Reyes, Romana and Sylvia Jopillo. 1988. "The Impact of Participation: An Evaluation of the NIA's Communal Irrigation Program." In Transforming a Bureaucracy: The Experience of the Philippine National Irrigation Administration. ed. Frances F. Korten and Robert Y. Siy. 90-116. West Hartford, Connecticut: Kumarian.

Helmi and Douglas Vermillion. 1990. "Using Irrigation Agency Staff as Institutional Organizers: The Small Systems Turnover Program in Indonesia." In Role of Social Organizers in Assisting Farmer Managed Irrigation Systems. ed. Shaul Manor, Sanguan Patamatamkul, and Manuel Olin. 41-47. Colombo: International Irrigation Management Institute.

Hunt, Robert. 1989. "Appropriate Social Organization? Water User Associations in Bureaucratic Canal Irrigation Systems." Human Organization 48 (1): 79-90.

Judawinata, Renyasih. 1991. A Study of Public Works Small Scale Irrigation System Turnover to the Water Users: The Case of Four Systems in Cianjur, West Java, Indonesia. Master's Thesis, Ateneo de Manila University.

Korten, David C. 1980. "Community Organization and Rural Development: A Learning Process Approach." Public Administration Review 40 (5): 480-510.

Korten, Frances F. 1987. "Making Research Relevant to Action: A Social Learning Perspective." In Public Intervention in Farmer-Managed Irrigation Systems. ed. Ministry of Water Resources, IIMI and Water and Energy Commission Secretariat Government of Nepal. 293-306. Sri Lanka: International Irrigation Management Institute.

Korten, Frances F. 1988. "The Working Group as a Catalyst for Organizational Change." In Transforming a Bureaucracy: The Experience of the Philippine National Irrigation Administration, ed. Frances F. Korten and Robert Y. Siy. West Hartford, Connecticut: Kumarian.

Korten, Frances F. and Robert Y. Siy Jr. 1988. Transforming a Bureaucracy: The Experience of the Philippine National Irrigation Administration. West Hartford, Connecticut: Kumarian.

Maas, Arthur and Raymond L. Anderson. 1978. ...and the desert shall Rejoice: Conflict, Growth and Justice in Arid Environments. Cambridge, Massachusetts: MIT Press.

Manor, Shaul, Sanguan Patamatamkul, and Manuel Olin, ed. 1990. Role of Social Organizers in Assisting Farmer Managed Irrigation Systems. Colombo: International Irrigation Management Institute.

Mayson, Evan. 1984. "Assisting Villagers to Build Their Own Irrigation Scheme: An Example From Thailand." Overseas Development Institute Irrigation Management Network Paper 9e.

Moore, Michael. 1988. "Maintenance Before Management: A New Strategy for Small Scale Irrigation Tanks in Sri Lanka?" ODI-IIMI Irrigation Management Network Paper 88 (2e):

Murray-Rust, Hammond and Douglas Vermillion. 1989. Efficient Irrigation Management and System Turnover. Volume 3: Small Scale Irrigation Turnover Program. Colombo: International Irrigation Management Institute.

Ostrom, Elinor. 1990. Crafting Irrigation Institutions: Social Capital and Development. Workshop in Political Theory and Policy Analysis. Decentralization: Finance and Management Project. Indiana University.

Tan-Kim-Yong, Uraivan. 1987. "Problems and Strategies in Management of Communal Irrigation Systems: The Experience in Joint Decision-making by Farmers and Agencies." In Public Intervention in Farmer-Managed Irrigation Systems. ed. International Irrigation Management Institute. 277-291. Sri Lanka: International Irrigation Management Institute.

Tobing, Hanna. 1989. NGO Strategy in Irrigation Development: The Case of LP3ES of Indonesia. M.S. Thesis, Ateneo de Manila University.

Uphoff, Norman. 1986. Improving International Irrigation Management with Farmer Participation: Getting the Process Right. Boulder, CO: Westview Press.

Uphoff, Norman. 1991. Learning from Gal Oya: Possibilities for Participatory Development and Post-Newtonian Social Science. Ithaca, NY: Cornell University Press.

Vermillion, Douglas L. 1991. "The Turnover and Self Management of Irrigation Institutions in Developing Countries." Colombo: International Irrigation Management Institute.

Yoder, Robert. 1992. "Environmental Factors and Observed Practices." In Designing Irrigation Structures for Mountain Environments: A Handbook of Ideas., ed. Robert Yoder. Colombo, Sri Lanka: International Irrigation Management Institute.

Yoder, Robert and Juanita Thurston, ed. 1990. Design Issues in Farmer-Managed Irrigation Systems: Colombo, Sri Lanka: International Irrigation Management Institute.


Next -> Chapter 9: Distributed Information Systems for Farmer Managed Irrigation




C-723 Lanna Condo, T. Pa Tan A. Muang Tel 66-53 872-225

Chiang Mai 50300 Thailand Fax 66-53 872-226