Networks and agricultural development: the case of soybean production and consumption in Brazil

Diffusion theory and political economy are the two most common ways by which rural sociologists have explained the widespread use of agricultural innovations. In this paper, the authors argue, borrowing from science studies and using the empirical case of soybeans in Brazil, that the use of agricultural innovations is better understood as the result of the construction of networks of people and things. Thirty years ago soybean production in Brazil was insignificant. However, at that time a soybean network began to be formed. Initial actors were, among others, (1) soybean varieties produced in the United States, (2) wheat farmers located in the south of Brazil, and (3) soil in need of nitrogen, an expensive input for the new wheat varieties. State policies were no doubt important in consolidating that network. However, other persons, institutions and things were also necessary, including science and technology, the creation and/or expansion of farmers' organizations such as cooperatives, and the solution to the problem of photoperiod sensitivity in soybeans. They follow the soybean from its introduction in Brazil to its present status as an important crop. The overall context within which the soybean network began is also described. They conclude that the actor network approach is more congruent with the history of soybeans in Brazil than diffusion theory or political economy.