Our interest in trying to find a way to analyze the ecological effects of the traditional management system was stimulated by the crisis in rice production triggered by the advent of the "Green Revolution" in Bali. During the 1950s Indonesia (figure 7.1) was forced to import nearly a million tons of rice each year. The government of Indonesia was thus very receptive to the promise of higher yields from the "Green Revolution" in rice, and in 1967 launched a major program called BIMAS (Bimbingan massal, or "massive guidance") to boost rice production by providing farmers with high-yielding rice seeds and access to fertilizers and pesticides. The new rice varieties grew faster than the native crops, and farmers were encouraged to triple-crop the new rice whenever possible. BIMAS reached pests. in 1971 and by about about 70 percent of the rice terraces in South Bali were planted with Green Revolution rice. At about the same time, the Asian Development Bank began a major irrigation development project in Bali. Rice production increased, but as early as 1974 field-level agricultural officials in Bali were reporting "chaos in water scheduling" and "explosions of rice pests. In 1984 I wrote an unsolicited report to the Asian Development Rank, in which I tried to show how these problems were linked to their disruption of the traditional system of water management.

My report to the bank emphasized the ecological role of water scheduling and pest management by water temples. Before the Green Revolution, Balinese farmers met annually in regional water temples to set cropping patterns, which often involved staggering irrigation schedules from one irrigation system to the next. Ritual ties between water temples emphasized the interdependency of upstream-downstream relationships, and the temples also helped solve quarrels over water rights. With the arrival of the Green Revolution, religious ceremonies continued to be held in the water temples, but farmers were encouraged to plant rice as often as possible and the temples lost control of cropping schedules. Yet these traditional schedules had important effects on both water sharing and pest control. By enabling the farmers to synchronize cropping patterns, the temple networks provided a mechanism to facilitate water sharing, and also enabled the farmers to synchronize harvests and thus create brief fallow periods over large areas, thus reducing rice pest populations by depriving pest populations of their habitat. The success of fallow periods as a pest control technique depended on the extent and duration of the fallow period. Unless all of the fields in a large area were fallow at the same time, pests could simply move from field to field. I urged the Asian Development Bank to consider trying to strengthen (rather than weaken) this traditional system of water temple scheduling as an ecologically sound system of water management. But my advice was formally rejected by the director of irrigation and rural development projects in a memorandum to the vice-president of the bank in 1984:

We do not fully share the expressed concerns of Mr. Lansing. Certainly there is a direct relationship between large areas of fallow land for a considerable period and the population of pests. However, pest control programs carried out efficiently and effectively will control the pest population and allow growing of rice year-round if adequate water resources are available as is done, for example, in certain areas of Central and East Java where farmers grow three rice crops per annum. It should be noted, that there is no development without affecting traditional systems or customs. Everybody can criticize and damage a project, but only a few people can overcome those difficult problems and make the project viable.
The Asian Development Bank continued to advocate the use of pesticides rather than synchronized fallow periods as the right way to control pests. Four years later. a study by World Bank officials reported that the use of pesticides had by then "pervasively polluted the island's soil and water resources." Meanwhile, I had become convinced that the water temple system had evolved to optimize the tradeoffs between water sharing and pest control in different regions of Bali. In the traditional farming system, groups of subaks (local irrigator associations) belonging to local water temples adjust cropping patterns cooperatively to achieve fallow periods over sufficiently large areas to minimize dispersal of pests, but coordination of rice planting over too large a scale would create inefficient peaks of water demand. Simulation modeling provided a way to analyze this trade-off, and in so doing gain a deeper insight into the effectiveness of the temple system as a system of ecological management.


J. Stephen Lansing (1996) Simulation Modeling of Balinese Irrigation. In Canals and Communities: Small-scale Irrigation Systems, ed. by J. Mabry, pp. 139-156. Univ of Arizona Press, Tucson.