ment extension personnel served also as pesticide distributors (Pemsl et al., 2005; Williamson, 2005) to supplement low government wages. Smaller pesticide production and distribution companies grew rapidly in developing countries such as Argentina, India, China and South Africa, often producing cheaper but more hazardous pesticides than their multinational counterparts (Pawar, 2002; Bruinsma, 2003).

Impacts of the chemical pest control approach. The significant yield gains and achievements in food security obtained in many countries in the 1950s and 60s have been closely linked to the use of hybrid seeds, synthetic fertilizers and other inputs including pesticides and to high levels of political and institutional investment in public sector research and extension (Bhowmik, 1999; Evenson and Gollin, 2003a; Lipton, 2005). Yield losses owing to disease and weed infestations have been reduced through chemical pest control (Bridges, 1992; CropLife, 2005ab); animal health has improved where insect-vectored diseases have been successfully controlled (Singh, 1983; Windsor, 1992; Kamuanga, 2001) and soil resources have been conserved through notill practices, which sometimes rely on herbicide use (Lal, 1989; Holland, 2004). Some have speculated that widespread famines and devastation of crops from outbreaks of disease and pests have been prevented (Kassa and Beyene, 2001); from a historical evidence-based approach it is difficult to assess the validity of these claims. As early as 1950, evidence of pest resistance to pesticides, resurgence where natural enemy populations had been destroyed and secondary pest outbreaks began to accumulate (Stern et al., 1959; Smith and van den Bosch, 1967; van Emden, 1974). Pesticide resistance (including cross-resistance to new products) became extensive and has been thoroughly documented in the scientific literature (MSU, 2000; Bills et al., 2003).

     By the 1960s the adverse environmental and human health effects of pesticide exposure had become known. The impacts, widely documented in the scientific and medical literature and popularized (e.g., Carson, 1962), affected not only pesticide applicators but entire rural communities and diverse biota in aquatic and terrestrial ecosystems and watersheds (reviewed in Wesseling et al., 1997, 2005; Hayes, 2004; Kishi, 2005; Pretty and Hine, 2005; Relyea, 2005; USGS, 2006; Desneux et al., 2007). Acute poisonings by pesticide residues have had immediate adverse effects, including death (Chaudhry et al., 1998; Rosenthal, 2003; Neri, 2005). Social and environmental justice cases have been documented regarding the inequitable distribution of the benefits of chemical control (largely accruing to better resourced farmers and manufacturers) and the harms in actual conditions of use that are experienced disproportionately by the poor and disadvantaged and the "ecological commons" (Wesseling et al., 2001; Reeves et al., 2002; Jacobs and Dinham, 2003; Reeves and Schafer, 2003; Harrison, 2004; Qayum and Sakkhari, 2005). A significant portion of the chemicals applied has proved to be excessive, uneconomic or unnecessary in both industrialized (Pavely et al., 1994; Yudelman et al., 1998; Reitz et al., 1999; Prokopy, 2003; Pimentel, 2005) and developing countries (Ekesi, 1999; Adipala et al., 2000; Jungbluth, 2000; Sibanda et al., 2000;

Asante and Tamo, 2001; Dinham, 2003; Nathaniels et al., 2003). Pesticide reliance has also been linked to agricultural deskilling (Vandeman, 1995; Stone, 2007), evidenced by subsequent erosion of farmers' knowledge of crop-insect ecology and reduced ability to interpret and innovate in response to environmental cues at field level (Thrupp, 1990; Pemsl et al., 2005).

Chemical control remains the cornerstone of pest management in many parts of the world, sustained by its immediate results, the technology treadmill (see 2.1) and path-dependency (wherein a farmer's accumulation of equipment, knowledge and skills over time conditions her potential to change direction). It is also upheld by the professional cultures and training of most advisory and extension programs (Mboob, 1994; Sissoko, 1994; Agunga, 1995; FAWG, 2001; Sherwood et al, 2005; Touni et al., 2007); the dominance of institutions promoting technologydriven intensification of agriculture; product innovations and marketing by the agrochemical industries (FAO/WHO, 2001; Macha et al., 2001; Kroma and Flora, 2003; Touni et al., 2007); and direct and indirect policy supports such as tax or duty exemptions for pesticides (Mudimu et al., 1995; Jungbluth, 1996; Gerken et al., 2000; Williamson, 2005). In recent years, leading agrochemical companies have integrated seed ventures and biotechnology firms, enabling them to establish synergies among key segments of the agricultural market. This trend is expected to continue and lead to increasing convergence between the segments, with possible inhibition of public sector research and of start-up firms (UNCTAD, 2006). The history of chemical control illustrates a phenomenon in agricultural science and technology development, in which early success of a technical innovation (often measured by a single agronomic metric such as productivity gains), when accompanied by significant private sector investment in advertising and public relations (Perkins, 1982) and by direct and indirect policy supports from dominant institutional arrangements (Murray, 1994), translates into narrowing of organizational research and extension objectives, widespread if uncritical grower adoption and delayed recognition of the constraints and adverse effects of the technology (e.g., resistance, health hazards, etc.).

2.3.2.2 Integrated Pest Management (IPM)

Integrated Pest Management (Box 2-7) in its modern form was developed in the 1950s in direct response to the problems caused by use of synthetic insecticides in actual conditions of use (Perkins, 1982). IPM took many forms but in general emphasized cultural and biological controls (Box 2-8) and selective application of chemicals that do not harm populations of pest predators or parasitoids (Stern et al., 1959), based on scientific understanding of agroecosystems described as complex webs of interacting species that can be influenced to achieve crop protection. IPM adoption in industrialized countries was stimulated by growing concern for human health and the environment, consumer desire for low or no pesticide residues in food (Williamson and Buffin, 2005) and public sector recognition that regulatory interventions were needed to remove the most harmful chemicals from sale. The spread of IPM in the South was driven