Introduction

Irrigation systems in Southeast Asia have undergone significant changes in the last 50 years.¹ Technological interventions aiming at increased efficiency and crop productivity were accompanied by the implementation of centralized, government-led irrigation bodies, detaching local users from the management of their resources (Groenfeldt, 1997). Many state-operated irrigation infrastructures, however, deteriorated because social aspects, at the centre for local users, were neglected in the project designs (Meinzen-Dick, 1997). In consequence, the focus shifted to decentralization to include local water users in participatory management (Dayton-Johnson, 2003).

Nevertheless, there is still an underlying disposition that greater formalization of farmer managed irrigation systems (FMIS) will lead to even better water-flow management. Irrigation experts, non-government as well as government, find it difficult to grasp the 'informality' of irrigation management, which is part of local resource users' management strategies.

Informality is the enemy of government improvement projects. Informality implies a shared understanding of the resource users, not only on how they manage the resource, but also on how the resource is embedded in the local environment. Good knowledge of rainfall patterns, soil qualities, water dependencies, religious activities, and many more variables anchors the informality in such a complexity that it is extremely difficult and time consuming for the outsider to penetrate into this complexity. Irrigation improvement projects, therefore, tend to reduce the complexity to a few variables. What Scott (1998) calls the hegemony of scientific knowledge over local, practical knowledge allows putting the complexity of local practices aside by labelling them 'underdeveloped'. This, then, is the determining attitude of developers in regard to simplifying local practices by emphasizing a few variables, which they define as important. Such a simplification – often accompanied by introducing more formal institutions – can undermine the complexity that local users employ. It takes management variables out of the context and gives them a higher priority. For local users this complicates resource management rather than simplifying it because the context is disrupted.

The Balinese subaks – loosely translated as socio-religious irrigation societies – are examples of how irrigation management is embedded into the social organization. Although subaks have formal arrangements concerning membership, territory, and members' rights and duties, these arrangements are embedded into social and ritual interactions between members and between adjacent subaks (Sutawan, 2000). Water distribution, the main concern of the subak, is negotiated on a daily basis between farmers, sub-subak heads, and subak heads. Social relationships, informal agreements, even disputes are seen as an integral part of the daily working of the subak as a whole. For Balinese farmers, the subak is the framework in which the informal management can take place. External agencies tend to concentrate on the formal subak structure such as its law, internal hierarchy, and irrigation infrastructure, whereas subak members use both formal and informal subak institutions to optimize irrigation. The following paragraphs present a case study where irrigation improvement experts' perceptions collided with local users' attitude towards their irrigation system.

One size does not fit all

Major irrigation infrastructure improvements in Indonesia began in the early 1970s, almost at the same time as the introduction of high-yielding rice varieties. The development of new irrigation regions and the improvement of already existing ones aimed at increasing rice production to meet the needs of the fast-growing Indonesian population. The tremendous increase in irrigated regions in Indonesia brought with it immense operation and maintenance (O&M) costs for the government because the irrigation management was centralized in government offices. In 1989 the O&M costs amounted to 65 billion Rupiahs (US$ 37 million at that time) per year (Gerards, 1992: 224).

The growing financial burden on the Indonesian government led to the introduction of irrigation service fees (ISF).² The objective of the ISF, to be paid by the irrigation water users, was to obtain full cost recovery for O&M cost of the main irrigation infrastructure for systems larger than 500 hectares (Bruns, 1992). However, such a decentralization of responsibilities and financial arrangements required localized irrigation management institutions not yet established in many parts of Indonesia.

The government acted quickly and created water user associations (WUAs) to unite farmers who share a common irrigation infrastructure. Several pilot projects in West, Central, and East Java as well as in South Sulawesi and North Sumatra were established in the years 1989–1991. By 1992, the government had transferred more than 400 irrigation systems to WUAs (FAO, 1993). It was seen as a big success, a win-win situation. On the one hand, the government could pass on O&M costs to the users and on the other hand, an institutional strengthening resulted by creating WUAs that worked closely with government officials in monitoring the irrigation infrastructure. As a consequence, the government extended the ISF to become a nationwide programme.

In Bali, the Bali Irrigation Project (BIP), financed by the Asian Development Bank, was launched in 1979. Irrigation facilities were rehabilitated or expanded, integrating smaller irrigation infrastructures into large systems sharing a common permanent weir. Here, there was no need for establishing WUAs. In the subak, farmers were already cooperating in irrigation management.

The pilot site selected for the ISF programme in Bali happened to be the site where we did research in 2004–2005. These subak situated in the south central area of Bali were chosen because they are located close to urban centres, already share an irrigation infrastructure, and have an irrigation area of more than 500 hectares. The government was interested in increasing the economic benefits by strengthening the subak because this region is renowned for producing the highest yields in Indonesia. After the selection of the pilot site, the 'socialization' phase of the ISF project began within the subaks. This meant informing the subak heads that they were chosen for the pilot project and explaining what the project was about. The subak heads were then asked to distribute the information among the farmers at the subak meetings.

An important part of the project was the establishment and strengthening of infrastructural-management institutions: the creation of a regional ISF council consisting of government officials and subak representatives and the administrative body.³ The six subaks that were chosen for the pilot project were to establish a federation with the assistance from the government. This subak federation was to strengthen subak interaction and, as a result, optimize irrigation management. The ISF council was set up by the end of 1995. The subaks were combined into a federation by the end of 1996. Since the six subaks share a dam and a primary canal, no infrastructural adjustments had to be made. In the beginning, the government was to play the role of the catalyst for the changes with the aim to eventually retreat to form an advisory and supervisory position (Bappeda, 1996).

Three of the six subak heads still held their positions by the time we arrived in the field in September 2004. However, it took a while until we found out that they were actually combined into a federation. None of the subak heads ever mentioned it to us. We learned about it two months later when we first met with the local government official involved in the project. During the interview he mentioned the subak federation to us. He explained that the subak federation was a corporate body that manages water, monitors the infrastructure, has three-monthly meetings, and even has a written set of laws. In response we followed up on his information when interviewing the subak heads. It seemed from their responses that the subak heads were not so interested and had difficulties putting into words what the subak federation was about.

Over time, it became clear to us why the subak heads did not show any great interest in their federation. While the creation of WUAs and the introduction of ISF in Java, Sulawesi, and Sumatra had shaped a sense of belonging among the farmers, for Balinese farmers there was no need. In Bali that was already part of the duty of every subak member towards their subak and it already embodied principles of equality and unity. A sense of belonging had already been part of Balinese subaks as long as the farmers could remember. However, one major difference between the other islands and Bali was that while the Javanese farmers saw the changes as acts of decentralization giving them more power, the subak heads and subak members in Bali saw the institutional strengthening as a disempowering interference in their affairs. They did not aggressively oppose the government project, nor did they support it. Indeed, in 2004 the project was in limbo. Nobody really knew what was going on – and the subak heads did not seem to care.

Basically, the creation of the subak federation was just putting a formal name on interactions that had already existed for a long time. This made little sense to the subak heads – particularly the creation of a set of laws for the subak federation to guide the subak heads in their interactions. Their interactions were embedded in the social texture of the local environment. The subak heads did not see any benefit in formalizing these interactions. However, as it is a government requirement that a subak federation has a law book, they formulated some regulations by copying the law book from another subak federation – just changing the names of the subak. It took them nine years to hand in the set of regulations in order to meet the government requirements.

The problem with ISF

In 1996, shortly after the subak federation was established the government went about trying to gather the new ISF from the subak members. The heads of subak were supposed to get their sub-subak heads to gather the money from the sub-subak members. But, the majority of the subak members were not willing to pay.⁴

The farmers did not see why they should pay an ISF if they were doing the irrigation servicing themselves and already paid several fees to the subak to maintain religious, infrastructural, or social order in the subak.⁵ The subak members also maintain that they carry out O&M and would like to continue to do so in order to ensure the O&M is of good quality.

Although the subak heads expected problems from the government, nothing happened. The ISF was never enforced. One can only assume that the responsible government agencies understood the concerns of the subak members. While the WUAs never paid any fees for O&M up until the introduction of the ISF in other parts of Indonesia, the Balinese have always paid fees and been engaged in canal repairs in exchange for receiving their share of irrigation water.

O&M responsibilities: Confusing simplifications

Part of the national ISF project was to return O&M responsibilities to WUAs. The legal basis was provided by a national regulation of the Ministry of Home Affairs (law 6/1992) (Gerards, 1992: 227). Translated into a Balinese context this meant that the government was responsible not only for the dam, but also for the primary canal – which the subak members believe belongs to several subaks – and the secondary canal – which lies within subak territory. Subak members see it as their responsibility to operate and maintain these canals. There is, therefore, a conflict between national law and the Balinese farmers' perception. This conflict continues to exist with the national government recently reinforcing the regulation. The new national water law 7/2004, article 41, for example, stipulates that the primary and secondary irrigation infrastructure falls under the responsibility of the provincial or regional government. The subak heads say it is quite confusing to know who is responsible for what, and as long as this confusion remains unsolved, the subak members will have to continue to do the O&M of the entire infrastructure just to make sure that everything is running to their full satisfaction.

Usually, the regional government financially supports larger canal repairs. If the financial support is below 100 million Rupiahs (US$11,000) the subak head can, if he wants, lead the maintenance work and involve the subak members if the assembly agrees on it. The subak is then responsible for all aspects of the repair work including the bookkeeping, the purchase of building materials, the organization of labour requirements, and the reporting to the department of public works. The subak also carries full responsibility for the success of the repairs, which is particularly related to the finances. If the repairs are done for less than the financial support received, the subak can keep the surplus, but if the costs exceed the given amount, the subak has to pay the excess out of its own pocket.

If the maintenance work exceeds 100 million Rupiahs, the Department of Public Works directly tenders the project to a private business and the subak is not involved at all. Consequently, the subak heads try to keep the proposals below this amount because this allows them to be directly involved in the infrastructural repairs. It again shows how centrally important the irrigation infrastructure is for subak internal organization. It represents the subak as a whole and binds adjacent subaks together. Naturally the infrastructure is the carrier of the most valuable resource in irrigated agriculture – water. As protective as the subak members are in keeping their autonomy in O&M, they are similarly concerned with water flow management.

The national plans to improve irrigation infrastructures by introducing ISF, creating WUAs, and introducing new national regulations to support all of these measures have resulted in chaos. The heads of subak ask why things have to be made so complicated when, in the end, they operate and maintain their infrastructure like they have always done. Moves to simplify irrigation management by the government have had quite the opposite effect leading to a rather confusing situation among local users.

Conclusion

Government agencies understand that improved irrigation management requires a unilateral move from informal to formal institutional settings. Informal institutions are seen as weak links. They believe that to strengthen irrigation management, there would have to be an emphasis on the formal arrangements in decision-making, focusing on representation, regularization, and formalization. This implies that informal institutions should be gradually replaced by 'modern' arrangements in order to reduce the deficiencies of the 'traditional' (Seabright, 1993).

Underlying these approaches are functionalist assumptions that better institutions can be crafted by policy makers. Government officials praise the subak for its existing structures, such as clear boundaries of the irrigation area, obligatory membership for those who cultivate rice within these boundaries, and clear rules and regulations that subak members have to adhere to.⁶ However, they complain that Balinese farmers are strong-headed, which leads to non-compliant behaviour and regular conflicts that in turn negatively affect optimal irrigation management. They lament that authority structures are unclear in a subak and that decision-making lacks formal regulations.

Yet, for Balinese farmers the authority structures are not ambiguous, although they might lack formal construction. Every subak member has the authority to make decisions regarding his own rice terraces, respecting of course the jointly agreed arrangements on the sub-subak and the subak level. The sub-subak heads have the formal authority to represent the sub-subak members at subak level and the informal authority to negotiate water with each other. The subak heads have the formal authority to represent their subak at inter-subak meetings and in inter-subak activities. This can be for ceremonies or in dealings with government officials. Informally the subak heads have the authority to negotiate water among each other in a beneficial way for the subak members. In these dealings they are expected to act in the best interest of the subak members they represent.

Government officials also tend to perceive the subak as hierarchically organized. The subak head is perceived to be at the top. He has his sub-subak heads who enforce his decisions on the sub-subak level, and the subak members obey these orders. In reality, however, authority is diversified and most power lies on the terrace level. Such decentralization is perceived to be impractical if the government wants to have control over irrigation management by using a bureaucratic approach. It seems impossible to situate a government official somewhere in the subak management system without disturbing already existing management arrangements. The subak members are too intertwined with subak matters and their irrigation management is not just a practical matter that can be easily modified. Rather, it is embodied in complex relationships focussed on ideas of equality and unity.

Government intervention would be far more efficient and effective if it encourages a minimum of formal organization on the part of the farmers. The aim for subak development should be to avoid unnecessary and expensive institutional overhead. Bruns (1992) argues that an active, successful WUA is one that efficiently and fairly distributes water and maintains and improves the physical infrastructure. The Balinese subak already accomplishes these elements – partly in formalized procedures, partly in informal interactions. It is questionable whether it is worthwhile to build up excess capacity to deal with contingencies. Government intervention should be selective and responsive rather than extensive and top-down. Reorganization is not beneficial unless it is based on local initiative and addresses specific local issues. As far as the Balinese subak is concerned, it would be wise to assume that current formal and informal subak institutions are adequate for effective irrigation management.

Notes

¹ On a comprehensive assessment on the evolution of irrigation in South and Southeast Asia, see Barker and Molle (2004).

² The Indonesian word for the Irrigation Services Fee is Iuran Pelayanan Irigasi.

³ The ISF council was called BAMUS, which is the abbreviation for Badan Musyawarah. The administrative body was called SETTAP, which stands for Sekretariat Tetap IPAIR.

⁴ Subaks are sub-divided into smaller units. Each of these smaller units comprises an average of 40 subak members who share a tertiary irrigation canal. The sub-subak head elected by members out of their ranks represents the sub-subak members at the subak level and can be consulted when there are conflicts.

⁵ The farmers pay a ritual fee to cover ritual costs; they pay a water fee that is used for infrastructural repairs calculated on how much water their fields receive; and they pay labour-substitute fees if they do not come along to communal labour activities.

⁶ While government officials regard subak laws as rigid and formal, subak members see them as "behavioural" guidelines. Subak rules are seldom applied in a formal way. Rather, if a problem occurs that cannot be resolved at the terrace level, the head of subak can be asked for advice. He can consult the subak laws, but usually they are put into the particular context. Elder subak members are often asked for advice on how such matters were handled before. Decisions can only be made if the majority of the subak members agree with them. This, again, shows the unfamiliarity with practices where a single person such as the head of subak can make coercive decisions in a top-down matter.

References

1. Bappeda (1996) 'Laporan Hasil Pelaksanaan Proyek Pendataan Obyek Pelaksanaa dan Sosialisasi IPAIR di Kecematan Mengwi', report, Bappeda Badung [Regional Development Planning Board]: Badung, Bali.
2. Barker, Randolf and François Molle (2004) 'Evolution of Irrigation in South and Southeast Asia', research report 5, Comprehensive Assessment Secretariat: Colombo.
3. Bruns, Bryan R. (1992) 'Just Enough Organization: Water users' associations and episodic mobilization', Visi: Irigasi Indonesia 6: 33–41.
4. Dayton-Johnson, Jeff (2003) 'Small-Holders and Water Resources: A review essay on the economics of locally-managed irrigation', Oxford Development Studies 31(3): 315–339. | Article |
5. FAO (1993) 'The State of Food and Agriculture', FAO Agriculture Series No. 26, Food and Agriculture Organization of the United Nations (FAO): Rome.
6. Gerards, Jan L.M.H. (1992) 'Introduction of Irrigation Service Fee (ISF) in Indonesia: Institutional development in action for resources management', Irrigation and Drainage Systems 6(3): 223–247. | Article |
7. Groenfeldt, David (1997) 'Transferring Irrigation Systems from the State to Users: Questions of management, authority, and ownership', paper given at the 96th annual meetings of the American Anthropological Association, Washington, DC, USA, 19 November.
8. Meinzen-Dick, Ruth (1997) 'State Administration, Devolution, and Water Markets in Irrigation Management', DVWK Bulletin [Bulletin of the German Association for Water Resources and Land Improvement] 20: 5–24.
9. Scott, James C. (1998) Seeing Like a State: How certain schemes to improve the human condition have failed, New Haven: Yale University Press.
10. Seabright, Paul (1993) 'Managing Local Commons: Theoretical issues in incentive design', Journal of Economic Perspectives 7(4): 113–134.
11. Sutawan, Nyoman (2000) 'Negotiation of Water Allocation Among Irrigators' Associations in Bali, Indonesia', in Bryan Bruns and Ruth Meinzen-Dick (eds.) Negotiating Water Rights, New Delhi: Vistaar Publications.