The System of Rice Intensification (SRI).
Modified from a contribution from Dr Norman Uphoff, Director, Cornell International Institute for Food, Agriculture and Development (CIIFAD).

The system of rice intensification (SRI) was developed in Madagascar about 20 years ago, by Fr. Henri de Laulanie, SJ, after 20 years of his observation and experimentation, working with farmers to develop a low-input strategy for raising the yields and productivity of irrigated lowland rice. The ideas and techniques are now being adapted also to upland rice production mutatis mutandis.

SRI methods have now been tried and validated in 16 other countries beyond Madagascar: China, Indonesia, Philippines, Cambodia, Bangladesh, Sri Lanka, Myanmar, Laos, Nepal, India, Sierra Leone, The Gambia, Benin, Guinea, Cuba and Peru. Contrary to the assertion made by Dr. Achim Doberman in Agricultural Systems (without providing any empirical evidence for this assertion), SRI is not a 'niche' innovation, suitable just for certain kinds of soils or climate. SRI methods have been productive in a wide variety of agroecosystems, from tropical to temperate climates and at various elevations, from sea level to 1200m, with all kinds of rice cultivars, both improved varieties (HYVs or hybrids) and traditional ones (local landraces).As indicated above, SRI was developed by Father de Laulanie over 30 years before his death in 1995. SRI was being promoted in Madagascar by a small NGO that he and several Malagasy friends established in 1990, Association Tefy Saina, when CIIFAD learned about this new methodology in 1993. After three years of remarkable results (1994/95 to 1996/97) with average yields of 8 t/ha on farmers' fields where previously they had gotten 2 t/ha, CIIFAD began working with Tefy Saina and other Malagasy partners to evaluate and disseminate SRI methods. These do not require any external inputs (no new seeds, no chemical fertilizer, no agrochemical use) and reduce water requirements by about 50%.

The first evaluations of SRI outside of Madagascar were in 1999 (Nanjing Agricultural University in China, and the Agency for Agricultural Research and Development in Indonesia), and since then, SRI has been extended more and more widely. There are trials being undertaken in many countries. As an initiative rather than a project, there is no planned end. SRI will continue to spread and to evolve (improved in various ways by farmers and researchers) as long as the genetic potentials of rice can be further exploited.

Specific data from five years of adoption and evaluation with farmers in the peripheral zone around Ranomafana National Park in Madagascar, assisted under an integrated conservation and development project funded by USAID. The SRI extension work was done by Association Tefy Saina, a Malagasy NGO. Note that the soils of this zone had been evaluated for an agronomy PhD thesis for North Carolina State University in 1992 with the conclusion that these were some of the poorest soils NCSU had ever evaluated: pH 3.8-5.0, available P 3-4 ppm, low to very low CEC in all horizons. Rice average yields before SRI interventions were in the region of 2 t/ha (documented by NC State Univ). Rice Average yield with SRI interventions: 8 t/ha (detailed records kept by Tefy Saina). Similar results from a five-year French-aided project on the high plateau, where conditions are somewhat better: average yield with peasant practices 2.5 t/ha; with SRI 8.8 t/ha.

Impact in terms of increased water availability - SRI is beneficial because of its water-saving. With SRI methods, paddy fields are not kept continuously flooded during the vegetative growth phase. Instead, fields are just kept moist, not flooded, with periods of drying of 3-6 days; or fields are flooded for 3-5 days and then drained and kept un-flooded for 3-5 days. Overall saving of water has been measured between 40 and 60%.

Adoption: SRI adoption has been most rapid in Cambodia. The uptake in Madagascar has been 20,000-100,000, depending on one's definition of what constitutes SRI, which is a set of practices. The full adoption is around the lower number, the use of SRI practices and ideas in some combinations is around the higher number. In Sri Lanka, SRI is being used by about 5,000 farmers we estimate; in Bangladesh, also about 5,000. The government in Indonesia is promoting SRI as part of its Integrated Crop Management strategy, in 20 provinces, but we have no data.

Farmers who have made SRI work for them, getting yields in the 10-15 t/ha range, have had their lives transformed. One farmer who first used SRI in 1994/95 near Ranomafana, on 1/4 ha now uses SRI on 8 ha and has become a rich man, owning all his paddy land and also three houses, including one in the provincial capital of Fianarantsoa. There have been no systematic evaluations of household/community impact.

In some studies, there has been evidence of dis-adoption in Madagascar, usually by farmers who have a labor constraint and who are so poor that they do not have enough household liquidity to defer income streams until harvest time. In Cambodia, where we have seen a smart dissemination strategy used, dis-adoption has been negligible. In the French-aided project in Madagascar referred to above, SRI use spread from less than 50 ha to almost 600 ha over the five years for which data were reported (Hirsch, 2000).

Quick and tangible benefits: SRI has many other benefits for the environment and human health (reduced use of agrochemicals, for example, that become important for its sustained used. That SRI requires much less use of seed (as much as 90% less) means that farmers can save immediately as much as 100 kg of rice per hectare, a significant benefit - very quick and tangible.

Low risk of failure: SRI has the disadvantage of appearing 'risky' at first. The main changes in plant, soil, water and nutrient management that constitute SRI appear risky with respect to transplanting a tiny, young seedling only 8-12 days old, only one per hill, wide spacing between plants which leaves a lot of field unplanted, and no standing water. Farmers need to be willing to take some risk at first, and some ridicule from other farmers who think the practices look 'crazy.'

Market opportunities: Most of the farmers who have taken up SRI are small farmers, who are not even self-sufficient in rice production. In addition, rice is one commodity with well-established markets, even if these are often extractive and exploitative. Certainly having good market opportunities for sale of any surplus is a bonus for SRI adoption. This probably is more important a factor for other crops than rice.

Aspiration for change: SRI has thus far required very innovative persons to get it started. As more and more farmers make SRI methods work for them, and there is a broader cultural acceptance of its changes, there should not need to be as much dependence on individual initiative and chutzpah.

Innovation and appropriate technologies: The main technological constraint for SRI spread has been availability of suitable 'rotating hoes' for weeding paddies that are not kept continuously flooded. The continuation requires more because we expect (and are getting) a lot of farmer innovation, to save labor, to enhance seedling vigor, etc. SRI is promoted to encourage changes in farmers' thinking and attitudes, for adoption of improved practices, as it is to achieve higher yields.

Leadership: SRI has spread only because of leadership. It has attracted a very diverse set of people in different countries: NGO leaders, innovative farmers, independent-thinking bureaucrats, curious researchers who care about getting results that benefit the poor and the environment, retired teachers, environmentalists, etc. There has been a remarkable mobilization of altruism
in the set of persons who have been attracted to SRI, not rejecting it on a priori grounds as have most agricultural scientists, economists and administrators. As SRI spreads and gains credibility, it will not take as high an order of leadership as in the first 3-5 years.

Social Capital: Answering this question depends on how social capital is defined. What I see is that the start of SRI does not depend on social capital so much as on leadership (previous question), but then the spread depends on friendship networks and on norms, values, attitudes and beliefs that reinforce collective action for mutual benefit.

Participatory approach: SRI is not a 'technology' since it is not to be 'adopted.' Rather, farmers are encouraged to understand the principles and basic concepts, and to try the practices out with some adaptation, evaluating the results of different spacings, different ages of seedlings, etc. Our Cambodian colleague who leads the SRI work there says that SRI helps CEDAC to get more local participation because it encourages a more cooperative spirit, so that their community development efforts are benefited by extension of SRI. But there is a social and cognitive side of SRI that its proponents find very gratifying.

Property rights: SRI can be used by tenants as well as landowners. People who need to get the highest possible productivity from small amounts of land, labor, water and capital will find SRI very advantageous. However, since SRI builds up soil quality the more these methods are used, it does improve soil, and this would give an incentive for the spread of SRI among landowners.

Supportive policies: In many countries, SRI has been promoted without any support from the government, and in the case of Sri Lanka, with active and vocal criticism and opposition (from the government's rice research establishment). SRI has spread in spite of this (partly because of the courageous leadership of Dr. Gamini Batuwitage, at the time senior assistant secretary in the Ministry of Agriculture, who refused to be halted by the research establishment, having seen the benefits of SRI for himself). In Indonesia and China, there has been government support, which will probably pay off with rapid spread now that researchers have satisfied themselves of its merit. In India, getting central government support, despite three years of efforts, has been difficult, but two state governments (Andhra Pradesh and Tamil Nadu) are promoting SRI. So, the conclusion is that supportive policies are beneficial, but absolutely essential - provided that in the latter case, there are strong NGOs to work with. SRI has been largely a 'civil society' innovation, started by a priest and carried on by an NGO in Madagascar, and then supported by a university agency (CIIFAD), with NGOs and individuals taking the leadership in specific countries more often than government agencies have.