The social construction of weeds: different reactions to an emergent problem by farmers, officials and researchers P.V. Vissoh 1,4, à , R. Mongbo 1 , G. Gbe `hounou 2 , D. Hounkonnou 3 , A. Ahanche ´de ´ 1 , N. Ro ¨ ling 4 and T.W. Kuyper 5 1 Faculte ´ des Sciences Agronomiques, Universite ´ d’Abomey-Calavi, Cotonou, Benin; 2 Laboratoire de De ´fense des Cultures, Institut National de Recherche Agricole du Be ´ nin, Cotonou, Benin; 3 Private consultant, CDP-International, Cotonou, Be ´nin; 4 Communication and Innovation Studies Group, Wageningen University, Wageningen, The Netherlands; and 5 Soil Quality Group, Wageningen University, Wageningen, The Netherlands Rapid population increase in southern Benin has changed the prevailing system of shifting cultivation into one of more permanent land use. New herbaceous weeds exacerbated rural poverty through crop failure, higher labour inputs, rising costs of production and reduced availability of suitable land. We investigated how different actors reacted to the emergence of weeds, in terms of the construction of knowledge, labour practices and technology development. Weeds have become an important cause of rural poverty. Farmers have actively engaged in technology development and new labour practices have emerged. Officials early on did report weed problems, especially where export crops were concerned. Researchers have not translated the new weed problem into a research priority until very recently, resulting in limited and inappropriate weed management technologies. The challenge of the research of which this study is part is to optimize weed management, by combining emergent indigenous weed management practices with scientific knowledge. Keywords: Indigenous technologies, labour shortage, permanent land cultivation, population pressure, rural poverty, stakeholder diversity Introduction Weeds induce substantial crop losses and have an important impact on smallholders’ revenue. Weed control is labour intensive. It can consume be- tween 30 and 54 % of the total amount of labour used on an African farm, depending on the crop and the level of other available resources, and it takes about 280 hours of labour to weed one hectare twice (Okon & Amalu, 2003). In all, weed reduction is an important component of the quest to reduce rural poverty in Africa. However, formal science has made insufficient impact in addressing and solving weed-related problems. Weed-related problems raise interesting ques- tions with respect to how different groups perceive and react to weeds. In this paper, we focus on farmers, (colonial) officers and agricultural researchers. At a more theoretical level, the paper discusses the issue how the weed problem became socially constructed (Berger & Luckman, 1966; Hacking, 1999) by different groups and what the consequences were and are for dealing with the problem. What makes weeds in Africa especially interesting from a constructivist point of view is that within living memory, the issue of weeds has undergone a transformation in African farming systems. Population pressure on land has led to a very rapid decline in fallow periods from some- times more than 20 to four or five years (Mateete à Corresponding author. Email: pierrevissoh@yahoo.fr INTERNATIONAL JOURNAL OF AGRICULTURAL SUSTAINABILITY 5(2&3) 2007, Pages 161–175 # 2007 Earthscan www.earthscanjournals.com