In this essay we present three biases that make it difficult to represent farmer voices in a meaningful way. These biases are information bias, individual bias, and short-term bias. We illustrate these biases through two case studies. One is the case of Golden Rice in the Philippines and the other is the case of Bt cotton in India.

Despite its use to exemplify how the world is “flat,” India is in many ways “spiky.” Hyderabad is a prosperous hub of information–communication technology while its impoverished agricultural hinterland is best known for dysfunctional agriculture and farmer suicide. Based on the belief that a lack of knowledge and skill lay at the root of agrarian distress, the “e-Sagu” project aimed to leverage the city’s scientific expertise and ICT capability to aid cotton farmers. The project fit with a national surge of (...) “last mile” projects bringing ICT to the village, but it was unique in using ICT to connect farmers directly with agricultural scientists acting as advisors. Such projects fit the interests of many actors, which has led to an unrealistic national enthusiasm about their impacts. This article uses the first five years of the project as a lens to view the cultural nature of both indigenous agricultural knowledge and “scientific” agricultural advising. Unlike lay publics whose uptake of science is better known, with farmers the invention and adoption of agro-scientific knowledge is deeply embedded in daily productive activities and sociocultural interactions. E-Sagu eventually had to abandon its construction of agricultural science as objective and acultural, resorting to rural methods of persuasion. It also found that it could only survive by joining forces with companies promoting commodification of agricultural inputs, which was a cause of the agrarian distress it sought to alleviate. (shrink)