I’d like to challenge one of Prof. Collier’s key points: small farmers are failing to keep up with the pace of change. “Innovation is hard to generate through peasant farming”, he writes. “Their mode of production is ill-suited to modern agricultural production in which scale is helpful”.  On the contrary, small farmers have shown time and again a capacity to rapidly evolve technologies and systems; in this their greater number and their closer interactions with their land, crops, animals and each other, relative to large farmers, are key advantages. This extends to achieving scale economies – where these are attractive – through cooperation. In a context of rapid change, small farmers’ capacity to evolve is critical. However, it is far more often ignored or suppressed than supported and fed.

An illuminating case comes from the highlands of southern Rwanda, one of the most densely populated parts of the most densely populated country in Africa.
Soon after arriving in the late 1980’s, I took up an initiative to advance sustainable intensification of highland valley bottoms thru farmer-led experimentation. Farmer groups in 3 valleys tried out and modified technical options they or we suggested; they bore all risks, we provided initial seed and advice. Within 2 seasons the valleys were transformed (photos). Rice, previously only grown 200 m lower, spread rapidly. Farmers identified varieties that tolerated cold and developed cropping patterns adapted to their economic orientation and the hydrology of their valleys. By the second season, all the groups had constructed sandbag-reinforced diversion dams and peripheral irrigation canals. Farmers who had never before seen a need to farm cooperatively were now electing coordinators to organize tasks that benefited all, like irrigation maintenance, and, when necessary, to enforce penalties. Appropriate scales of cooperation were quickly found for different tasks: larger for maintaining canals, smaller for managing a seedbed, still smaller for scaring off birds. “Traveling seminars” in which the groups showed and explained what they were trying were crucial for the evolution of these lumpy options. How to maintain functional diversity was a constant topic of conversation.  Rice was proving very productive (appreciated at home and with a ready market) but its spread threatened other elements.  Sweet potato especially: growing it in the valley provided cuttings for the hillsides and made year-round cultivation possible – an enormous boost to food security. One solution was to grow rice in paddies then rebuild raised beds for sweet potato, beans and e.g. out-of-season maize for the market: tremendously labour-demanding but evidently feasible for farmers with a few hundred sq m of land. Innovation was driven by necessity, which was hardly in short supply. But the context was less than supportive: markets functioned poorly, extension was demeaning and the state apparatus hostile to any autonomous initiative. Discussion in policy circles favoured scale and specialization – fewer people, growing one or a few crops, either in the valleys or on the hills. This much of the story was recounted in Agricultural Systems (1994, attached). I left a year before the genocide broke out in 1994. I visited in 1996: despite upheaval and more than 4 years without support of any kind, the groups had survived and rice cultivation had spread up the valleys.  It was more difficult to make out what had happened to other innovations. A few months ago, a colleague visited the area. The groups are all still active, 20 years on, and he found rice dominant over many kilometres of valley (photo).  It’s unclear to what extent farmers have been supported in this by public or private sector institutions (I know some are active in the area) and whether farmer innovation is being recognized. I’d love to find out more. A final thought. New cultivation techniques for familiar crops may prove an important production frontier, particularly as climate change accelerates. The System of Rice Intensification and related approaches are notable examples. The wheel is still very much in spin but evidence suggests a potential for significant gains in production and water use efficiency along with an inescapable need for local innovation and adaptation around the basic principles. Supporting the innovative capacity the Rwandan groups demonstrated would seem essential if that potential is to be captured.  

Michael Loevinsohn, Applied Ecology Associates