Thoughts on the history, principles and practices of SRI, and on its importance for the present scenario*

Norman Uphoff

CIIFAD, Cornel University, USA

Participants in this symposium know much more about the application of SRI in Indian contexts than I do. What perhaps I can contribute to this review of the present status and future prospects of the System of Rice Intensification in India are some reflections on how SRI came into being, and how its arrival on the Indian and world scene is very timely, given large-scale trends that confront the agricultural sector in this country and around the world.

Thinking ‘Outside the Box’

SRI is available to us today only because Fr. de Laulanié and others who have followed in the directions that he mapped out were willing and able to ‘think outside the box,’ to use a term from popular culture. Those who want to understand and utilize SRI need to have this capacity as well.

Few beliefs in agriculture have been as long-standing or as deeply held as the idea that rice as a crop performs best when it is grown under continuously flooded conditions. In his well-known and widely-cited volume on /_Principles and Practices of Rice Production_/, S.K. De Datta has written: “A main reason for flooding a rice field is that most rice varieties maintain better growth and produce higher grain yields when grown in flooded soil than when grown in non-flooded soil” (1981, pp. 297-298). This has been the accepted conventional wisdom among scientists and farmers for many generations, even millennia.

The idea that rice plants perform best when grown in standing water is now being displaced by the better, often spectacular performance of SRI rice plants, which are grown under aerobic or at least intermittently aerobic soil conditions. Rice plants which are raised in soil that is kept moist but not continuously saturated are more productive, and indeed, the other inputs used in rice production also become more productive when paddy soils are kept unsaturated most of the time.

The best soil conditions for rice are thus now seen to be aerobic rather than hypoxic. Indeed, some periods of soil drying are beneficial for plants since putting some water-stress on plants induces their roots to grow deeper and more profusely. This has many benefits: more drought-tolerance, uptake of micronutrients, resistance to lodging, enhanced plant health, better grain quality, etc.The advantages of growing rice in better-drained rather than saturated soil should have been recognized long ago. In poorly-leveled rice paddies where some plants are growing in higher, well-drained areas and others in lower, waterlogged areas, it is evident that the former fare better than the latter. It should have been noticed long ago that while rice plants can /survive/ under flooded conditions, they do not /thrive/ there. A mental ‘box’ has constrained our thinking.

“Everybody knows that rice plants grow better in standing water” I was once told by a former director-general of the International Water Management Institute (IWMI), inferring this from the fact that irrigated (flooded) rice varieties produce more than their upland (water-stressed) cousins. Because “everybody knew” this presumed fact, there was no investigation of alternatives, and no assessment of any /intermediate/ or optimizing conditions. Intermittent wetting and drying or light irrigation that met plants’ minimum water requirements without interfering with their access to oxygen was not evaluated, because everyone was thinking inside this particular ‘box.’

Fr. de Laulanié demonstrated a remarkable ability to think outside the box. SRI changes rice-growing practices that have prevailed for hundreds, even thousands of years, mostly because they appeared to reduce farmers’ risks of crop failure. However, evaluations by IWMI in Sri Lanka (Namara et al., 2004) and by GTZ in Cambodia (Anthofer et al., 2004) concluded based on random samples of SRI users and non-users that the SRI methods actually reduce risk.

People ask why, if SRI has greater productive potential, it was not developed before the 1980s. My response is that each of the practices which it brings together appears to be risky, and farmers were unlikely to try these out separately, let alone to utilize them all together and reap the benefits of their synergistic promotion of plant root and shoot growth.

Seedling Age: Why would a farmer plant a ‘young, tiny seedling’, only 8-12 days old, when he/she could plant one 3-4 weeks old or more? Older plants appear and often are better able to withstand various stresses. However, young seedlings if carefully planted into an aerobic soil environment, not submerged in hypoxic soil, have more vitality and resilience, and more growth potential as can be explained in terms of phyllochron analysis (Stoop et al., 2002).

Number of Seedlings per Hill: Why would a farmer plant just ‘one seedling per hill’ when at little extra cost he could put in 3, 4, 5 or 6 plants together? This larger number of bigger plants looks much better (safer) than having one tiny plant all by itself. Yet this intuitive notion is contradicted by the known fact that when any plants are crowded together, their root growth is inhibited, something that applies

to rice as much as any other plant species.

Plant Density: Why should a farmer deliberately plant his field sparsely with ‘wide spacing’ between seedlings when it seems that a larger plant population will give more total yield? In fact, high density among plants inhibits their growth and performance, although spacing is a parameter to be optimized rather than maximized so as to get the greatest number of large and fertile panicles per unit area. What is called ‘the edge effect’ was widely recognized and cautioned against when crop-cut samples are taken to estimate the yield in a field. But why did not someone think outside the box enough to ask: although ‘the edge effect’ should be avoided when making yield estimates, shouldn’t we try to achieve it throughout the entire field by exposing plants more to the Sun and air if this leads to their being more productive?

Water Management: Why should a farmer manage water carefully to give his plants just an */optimum amount of water/* rather than keeping the field always flooded if water is available and if rice plants can tolerate this? Besides, flooding makes weed control easier. Keeping a lot of water in the field seems likely to buffer the crop against water shortages later in the growing season. However, this ignores the fact that hypoxic conditions cause rice plant roots to degenerate (Kar et al., 1974). Continuous flooding early in the growth cycle will diminish the size and health of plants’ root systems, making them less able to tolerate water stress later in the cycle. Plants with truncated roots cannot access the residual soil moisture in lower horizons that is accessible to plants which have large and functioning roots systems to maintain their growth and productivity. So always flooding plants makes them vulnerable.

Fertilization: Why should a farmer use ‘organic nutrients’ if chemical fertilizer is available? The latter is so convenient compared with collecting, processing and applying biomass as compost or mulch to add organic matter to the field. Use of inorganic nutrients was made all the more attractive if fertilizer was subsidized by the government, as often occurred, at least early in the Green Revolution. However, the adverse impacts of fertilizers on soil biota, suppressing or unbalancing soil communities, was not considered seriously.Fr. de Laulanié thought outside the box in his experimentation that developed a system for rice production which would be minimally dependent on external inputs and maximally driven by plant-soil system interactions, now better understood in terms of the contributions that soil biota make to plants’ growth and health. He used very young seedlings, one per hill, widely spaced, with no flooding, using the rotary hoe to control weeds which also aerated the soil and promoted the growth of aerobic soil organisms, and applied nutrients in organic form as much as possible.

The result was that SRI plants had ‘much larger and healthier root systems’, evident if one took the trouble to look at them, seeing how deep they grew and how they maintained a white color. This was the result of demonstrable effects, nothing mysterious, of wide spacing, of soil aerated and rich in organic nutrients, and of starting with young seedlings that retain their potential for vigorous root growth as well as for more profuse tillering, explained by phyllochron dynamics.

Larger root systems with larger canopies lead to more photosynthate, some of which goes into the soil through */root exudation/* and other forms of rhizodeposition. The carbohydrates, amino acids and other compounds that plant roots put into the rhizosphere support *larger, more diverse and more active populations of soil organisms*, ranging all the way from the tiniest microbes (bacteria and fungi) up to visible and obviously beneficial earthworms (macrofauna).

These are the foundation of SRI success, though they are ‘out of sight’ and usually ‘out of mind.’ Most soil organisms are invisible because of their minute size, so they are easily ignored. However, plant roots and soil organisms taken together are the foundation of better crop growth. The relationships are indeed symbiotic. Organisms provide services to plants such as N fixation, N cycling, P solubilization, phytohormone production, protection against pathogens, etc., while plants provide soil organisms with needed nutrients and habitat, as noted above.

The earlier concept of microorganisms being concentrated in the rhizosphere, that thin coating of biologically-rich and biologically-active soil surrounding the root system much as a rubber glove fits on a hand, is giving way to a broader understanding that such organisms live not only /on/ and /around/ the roots but also /inside/ the roots as endophytes. It has further been discovered that soil organisms such as rhizobia live also in and on the leaves, in what is known as the phyllosphere. They move from the below-ground rhizosphere up through plant’s roots and stem to get there/ Once established, they enhance the leaves’ chlorophyll reserves and photosynthesis, contributing thereby to greater grain production as well as to plant health (Feng et al., 2005).

Ten years ago, who would have thought that rice yields would be increased, without any external investment of resources, by soil organisms that live intimately and productively with plants in such a beneficial symbiotic relationship? The effects of microbial habitation within plants is not just the mechanistic one of helping plants acquire nutrients and water, but a more holistic effect, changing plants’ physiology and functioning, e.g., inducing systemic resistance in plants to the effects of pathogens (Habte, 2006) or affecting root physiology and function (Smith et al., 2004).

There are many ‘boxes’ that have constrained our understanding of how to practice agriculture in a productive, efficient, profitable and sustainable ways. Fr. de Laulanié set a good example for us of ‘thinking outside the box.’ Such a frame of mind is not anti-scientific, but rather a requisite for making scientific advances.

Unexpected or unanticipated results and relationships should be investigated systematically to grasp their causal mechanisms or processes. Knowing these can help us (a) improve upon and optimize practices that have been derived inductively, such as SRI; (b) anticipate and possibly remedy adverse consequences that might arise in the future with new practices; and (c) extrapolate what is learned from a crop like rice to others such as millet, sugar cane and others.

Persons who have difficulty thinking outside the box, whether farmers, scientists or extension personnel, will not find SRI particularly congenial. But persons with a steadfastly empirical bent of mind, who are willing to be good observers and seek greater understanding (theory) without being too constrained by /a priori/ thinking -- will be drawn to SRI. Such a stance and orientation will be important if we are to modify and improve our agriculture for success in the 21^st century.

Thinking about post-modern agriculture

‘Modern’ agriculture, a name often applied to the technologies and techniques developed and used in the latter half of the 20^th century, was immensely successful, and its productivity gains fueled widespread economic growth in many countries at the same time it forestalled growing poverty and hunger. So the following comments are not a critique or rejection of what was done, often under the banner of ‘the Green Revolution.’ Its contributions were great and needed, and represented that best that our scientific endeavors could propose for practical application.

However, the Green Revolution has been losing momentum since the mid-1990s, and the question arises: */what will we do for an encore?/* Can we succeed in meeting the world’s need for food and fiber, and at the same time continue reducing poverty and hunger and avoid further degradation of our environment, by */doing more of the same?/* These are not rhetorical questions, but rather real and important ones.

In some ways, what is ‘modern’ refers to whatever is most contemporary and up-to-date. But with respect to agriculture, this term became associated with agricultural production which was (a) highly *mechanized and land-extensive*, depending heavily upon energy inputs from fossil fuels, (b) heavily dependent on *agrochemical inputs* in the form of inorganic fertilizers and also a wide variety of biocides (pesticides, insecticides, herbicides, fungicides, etc.), many derived from petroleum, and (c) based on *improved varieties* of crops and animals having better genetic endowments through breeding programs or, now, genetic engineering and modification. Not necessarily, but in practical fact, this modern agriculture was also *very ‘thirsty’* -- depending upon the expansion of irrigated areas, requiring much investment of energy as well as capital.

This combination of technologies, though costly in economic and environmental terms, was very productive and, for many if not all, quite profitable. As we have left the 20^th century and begin the 21^st however, we note there are trends which will make these technologies less profitable and indeed less sustainable (Uphoff, 2006).

Arable land per capita is declining as populations continue to grow while land for cultivation declines year by year. More and more agricultural land is being lost to urban expansion. Irrigation and/or land management practices are contributing to loss of millions of hectares of cultivable land through salinization and erosion. The kind of land-extension production which was successful in the 20^th century is becoming less viable as productive land needs to be utilized with greater care and attention.

The volumes of fresh water available for agriculture are similarly declining as other competing uses expand, for industrial production, for domestic uses of a growing population, to meet expectations of improving life styles. While the volumes of natural rainfall are probably not changing much, rainfall patterns are shifting. Global warming will increase evaporation rates over large areas. So water must be used more wisely and carefully.

While the nature and extent of climate change remains uncertain, there is less and less doubt that weather in the 21^st century will be less benign for agriculture than in the 20^th century. The frequency and severity of ‘extreme events’ – droughts, flooding, hot spells, cold snaps, irregularity of seasonal patterns of rainfall and temperature – will probably plague farmers more than before. Accordingly, farming systems need to be more climate-proof, which works against large-scale monocropping with its higher risks, which could include more pest and disease problems associated climatic variation.

Energy costs are likely to be higher in the 21^st century than in the 20^th century. Nobody can know for certain what will be the price of petroleum. But modern agriculture was developed with input prices shaped when the cost of petroleum was $25 per barrel. That era is unlikely to come again. Petroleum prices of $50-100 per barrel are more likely to predominate in the future. Hopefully, renewable energy sources will expand greatly in the decades ahead, just as hopefully, technologies for desalinization and recycling of water can be greatly improved and spread. But henceforth, agricultural technologies will need to plan and expand with a very different set of factor prices shaping the sector’s path of development (Hayami and Ruttan, 1985).

Environmental impacts will need to be given more thought and attention as agrochemical use and soil management practices were adversely affecting soil and water quality. Build-up of nitrates in groundwater in China is already becoming serious, with levels 5-6 times higher in some places, where 500 kg of N are being applied per hectare, than the maximum allowed by the U.S. Environmental Protection Agency. Excessive pumping of groundwater is lower water tables in parts of India and China by 1 m per year, or more. Deforestation and expansion of cultivate areas is undermining vulnerable ecosystems and reducing biodiversity. The cavalier attitude taken toward natural resources during most of the 20^th century needs to change.**

Labor productivity will become a huge issue, both for reducing poverty in rural areas and for maintaining food security in urban areas. Depopulation of rural communities is accelerating in many countries, including India, with urban ‘pull’ often outweighed by rural ‘push,’ due to miserable and insecure living conditions. Within a decade or two, the aging of the rural labor force will turn into a stark crisis. The ‘modern’ response of accelerating the displacement of rural workers by mechanization will be less feasible economically as energy costs rise. The ‘free market’ response of growing more food in more-favored environments like Canada or Brazil and then shipping it to distant markets will be constrained by the same cost considerations, and by the fact that if the hungry are not gainfully employed, they cannot afford to buy imported food. The only sustainable and equitable solution is to find ways to raise labor productivity so that rural livelihoods can be more remunerative and attractive.**

SRI appears to be the harbinger of ‘post-modern agriculture,’ a new set of technologies and practices that is more based on biological solutions to the problem of raising agricultural productivity, superseding (though still building in many ways upon) the engineering, chemical and genetic solutions of the 20^th century.

SRI raises not just productivity of land, but also of water, of labor, and of capital, something not seen in agricultural innovations before. It does this despite the economists’ admonition that there is ‘no free lunch.’

SRI does require better water control – to be able to provide rice crops a smaller but more reliable amount of water – which can often be achieved through investments in physical infrastructure (hardware) and in social organization (software), as addressed in Uphoff (1986); Uphoff et al. (1991) and Uphoff (1996).

SRI does require more skill and knowledge on the part of farmers, but anyone who knows how to grow irrigated rice and who is motivated to achieve higher productivity, given water control, can learn what is necessary very quickly and easily. Some system of training and technical support to provide backstopping will make SRI spread quicker and more reliable, especially if coupled with a system and process of farmer-to-farmer dissemination. Use of farmer field school methods is particularly appropriate for SRI since it emphasizes farmers engaging in experimentation with and evaluation of the methods on their own fields.

Possibly some crop protection measures may be needed, with the larger production. However, as a rule, farmers report that pest and disease problems are reduced with SRI methods, possibly explainable in terms of the theory of /trophobiosis/ (Chaboussou, 2004). Initially there is more labor is required, while the new methods are being learned, but contrary to earlier understandings of SRI – suggesting that it is intrinsically more labor-intensive – reports are now coming in that SRI can even be /labor-saving/, e.g., Li et al. (2005) and Sinha and Talati (2005).

As noted above, SRI concepts and practices are now being extended to other crops beyond rice; and also, NGOs working with upland farmers in eastern India, northern Myanmar and southern Philippines are finding that methods derived from SRI experience elsewhere can be usefully adapted to rainfed rice production, with yields in the 6-8 t/ha range. The ‘trick’ is to avoid flooding the rice crop in its early stages, draining off water rather than retaining it – as farmers feeling insecure about future water supplies are prone to do. Keeping soil saturated stunts root growth, whereas making the soil aerobic encourages root growth, that later in the season gives rice plants resilience and makes them more drought-proof than if their roots have degenerated. Given the likelihood that rainfall patterns will become more adverse for crop production in the decades ahead, SRI and other practices that can nurture root growth will become essential for meeting future food needs.

SRI and its extension into what can be called ‘post-modern agriculture’ are coming on the scene at a very appropriate time. Crop demands for water need to be curtailed, especially for rice. Otherwise, there will be suboptimal supply for the agricultural sector, and it will be taking water that is needed for other uses, including preservation of natural environments. Agricultural systems cannot continue their ‘appetite’ for agrochemicals without compromising the health of sol systems, the environment and human beings. The use of chemical biocides creates a kind of ‘treadmill’ where new and more powerful chemicals are continually needed to replace ones that lose their efficacy or are found to have unacceptable effects on plant, microbial and/or animal life. The use of chemical fertilizers also encounters, at some point, diminishing returns. In China, whereas 40 years ago, the addition of 1 kg of N fertilizer was associated with 15-20 kg of additional rice production, this ratio has now declined to 5:1 (Peng, 2004) and will probably continue falling as levels of application are pushed higher and higher in response to lower marginal productivity. Irrigated production similarly needs to be weaned from its high consumption of water as water tables drop. Heavy use of chemical fertilizers in various parts of India and Pakistan has begun to force agricultural land out of production as this is sabotaging the soil fertility through salinity that it was intended to increase.

The adverse impact of N fertilizers on nitrate concentrations in groundwater was noted above. When such considerations are set alongside the changing economics of ‘modern agriculture,’ it becomes clear that technologies and factor proportions used in 21^st century agriculture will need to evolve in new directions. There will still be a role for external inputs of energy and nutrients and some crop protection methods. But as knowledge is growing about soil microbiology and soil ecology, it should be possible to manage and mobilize more of the endogenous processes and potentials of soil systems (better said, soil-crop-animal-microbial systems), being less dependent on exogenous resources. These resources will be regarded not as a substitute for endogenous ones but as a complement and a catalyst for them.

This is a perspective on 21^st century agriculture that is still only a vision. But the scientific foundations for this have been growing in the last half of the 20^th century (Uphoff et al., 2006). SRI gives impetus to this perspective, demonstration how more output can be achieved using reduced rather than increased external resources, but the perspective is a broad and encompassing one. For the present, we will do well to engage our minds and practical efforts with the opportunities of SRI. But we should maintain a broader view.

SRI is more than a set of practices (young seedlings, wider spacing, less water, etc.). Rather it is a set of insights and principles, even a philosophy, which can help to reshape the paradigm of agricultural production and its associated sciences. As noted in the keynote, part of SRI is methodological in the sense of encouraging farmer participation and innovation to make SRI more appropriate to local conditions and more owned by the users, who become more than adopters and indeed partners in the process of agricultural post-modernization. As a civil society innovation, stemming from the work of Fr. Henri de Laulanié, SJ, it embraces farmers and researchers, NGO workers and policy-makers, extension personnel and teachers, all walks of life.