Twenty-first century agriculture needs low-input advances like the System of Rice Intensification, says Norman Uphoff.
According to the principle of diminishing returns, continuing to produce something in the same way, with the same inputs and technology, usually becomes less productive over time. This appears to apply to agriculture's 'Green Revolution', as yield improvement has slowed in recent decades, no longer lowering the real price of basic foods as it did in its first three decades. Diminishing returns may not be a universal principle — but agronomists should take it seriously.
Although there is talk of another Green Revolution, the approaches being proposed are essentially more of the same. This technological strategy for raising production is running into major economic and environmental constraints, however. The costs of fossil fuel-derived inputs keep rising, while impaired soil health and degraded water quality are growing concerns.
Fortunately, there are agroecological options available, such as conservation agriculture and the System of Rice Intensification (SRI), which differ from high-tech and input-dependent production systems.
Agroecological management systems capitalise on the potential for more productive and robust crops from existing plant genomes and on their intricate, symbiotic associations with other organisms, particularly microorganisms — in what is now becoming better understood as the plant-soil microbiome. 
A paradigm shift — from regarding non-crop organisms as mostly pests or pathogens and treating plants as carbon-based machines, to understanding networks and webs of symbiotic relationships — can help us 'rebiologise' agriculture and adopt alternative methods that are better suited to current and foreseeable challenges.
More for less
By changing how plants, soil, water and nutrients are managed, SRI practices grow larger, better-functioning root systems that interact with a bigger and more diverse soil biota, also promoted by these same practices.
Compared with standard crop management methods, SRI practices raise yields usually by 50-100 per cent and sometimes more. These gains are achieved with less water, greatly reduced seed rates, less or no inorganic fertilisers, and often even with less labour once the methods have been mastered. [2,3,4,5]
In 2011, a farmer in the Bihar state of India who had adopted SRI reportedly surpassed the world-record paddy yield previously set in China. This stirred controversy — but having analysed how this yield of 22.4 tonnes per hectare was achieved and measured, with data from the Indian government's Directorate of Rice Development, I am satisfied that the farmer achieved this yield. 
This record yield is less significant, however, than two other statistics. First, the rice area under SRI methods in Bihar has risen from 30 hectares in 2007 to more than 300,000 in 2012, a 10,000-fold increase in five years. Second, even without all of the farmers following SRI recommendations fully, their average SRI yield in 2012 was calculated by government technicians as 8.08 tonnes per hectare — three times the usual yield in Bihar.
These figures and differences are so large that SRI can no longer be ignored by sceptics and critics. They come from farmers' fields and from official reports, not from experimental stations and partisan sources.
Winning the argument
Published criticisms of SRI have ebbed since 2006, with the benefits from SRI now demonstrated in more than 50 countries. But initial opinions die hard, even as governments in Cambodia, China, India, Indonesia and Vietnam, where two-thirds of the world's rice is produced, have begun supporting the spread of SRI, based on farmers' experiences and scientific evidence.
It is time to put the 'controversy' over SRI behind us. And it is time to begin learning more about how these new ideas and methods can help get more from less.
For instance, we have begun to learn how microorganisms that live mutualistically within plant organs and tissues, and even cells, bring benefits such as increasing the chlorophyll levels in leaves and protecting against pathogens in roots. [7,8]
Learning from farmers
Importantly, experience with SRI in countries such as Burundi, Cuba, India, Madagascar, Nepal and Rwanda is helping us to better understand how to learn from and with farmers. Some very informative and impressive videos are now available where farmers themselves explain their good experiences with these new methods. 
Farmers have been adapting and improving the methods to which they were introduced, and they have been disseminating their knowledge and experience to other farmers — changing the usual linear 'from lab to land' model of developing and transmitting innovations.
SRI is one of the few innovations where scientists have had difficulty replicating farmers' results in their on-station trials — usually the situation is reversed. Farmers may be getting higher yields than the researchers do because, more often than not, farmers' soils have less impairment from fertiliser and agrochemical applications than on experiment plots.
Changing times need changing practices
Agriculture in the twenty-first century will need to change considerably from the technologies and paradigms that evolved in the preceding century. Conditions are becoming increasingly different from the past. The need to change should not be taken as derogation of past research and practices.
Both scientists and farmers need to evaluate alternative methods that could be more suited to our new realities. Proponents of agroecological alternatives welcome systematic evaluation that is conducted rigorously, with open minds and with farmers' participation.