The ToT model's drawbacks with respect to development and sustainability goals. Criticism of the ToT model began to emerge strongly in the late 1970s as evidence of negative socioeconomic and environmental impacts of the GR accumulated (UNRISD, 1975; Freebairn, 1995) leading to sharp controversies that are still alive today (Collinson, 2000). Sometimes a technology itself was implicated; in other cases the institutional and economic conditions for using a new technology effectively and safely were not in place or the services needed for small-scale producers to gain access to or realize the benefits were inadequate, especially for the resource- poor, the indigent and the marginalized and women (Hunter, 1970; Roling et al., 1976; Ladejinsky, 1977; Swanson, 1984; Jiggins, 1986). The loss of entitlements to subsistence brought about by changes in the agricultural sector itself and in societies as a whole; weather-related disasters; civil unrest; and war also left many millions still vulnerable to malnutrition, hunger, and starvation (Sen, 1981; Johnson, 1996). The evidence highlighted three areas of concern:

Empirical: The ToT model was shown to be unfit for organizing knowledge processes capable of impacting heterogeneous environments and farming populations (Hill, 1982) and did not serve the interests of resource-poor farmers in risky, diverse, drought prone environments (Chambers, 1983). In the absence of measures to address women's technology needs and social condition, technologies transferred through male-dominated extension services largely bypassed women farmers and women in farm and laboring households (Hanger and Moris, 1973; Leonard, 1977; Harriss, 1978; Buvinic and Youssef, 1978; Fortmann, 1979; Bettles, 1980; Dauber and Cain, 1981; Evans, 1981; Deere and de Leal, 1982; Safilios-Rothschild, 1982; Mungate, 1983; Carloni, 1983; IRRI, 1985; Gallin and Spring, 1985; Muzale with Leonard, 1985; Nash and Safa, 1985; Staudt, 1985; Gallin et al., 1989; Gallin and Ferguson, 1991; Samanta, 1995). In addition, the improved seeds rapidly displaced much of the genetic diversity in farmers' fields that sustained local (food) cultures (Howard, 2003) and which had allowed farmers to manage place-dependent risks (Richards, 1985); the higher use of pest control chemicals in irrigated rice in the tropics had detrimental effects on beneficial insects, soils and water (Kenmore et al., 1984; Georghiou, 1986; Gallagher, 1988; Litsinger, 1989) as well as on human health (Whorton et al., 1977; Barsky, 1984).The evidence of negative effects on equity was claimed by some to be a first generation effect. Analysis of data from the Northern Arcot region of Tamil Nadu, India, indicated that the differences in yield found between large and small-scale producers in the 1970s had disappeared by the 1980s (Hazell and Ramaswamy, 1991) but further empirical studies failed to resolve the extent to which the second generation effects were the result of "catch up" by later adopters or the result of smaller farmers having lost their land or migrated out of farming (Niazi, 2004).

Theoretical: A basic assumption of the ToT model that "knowledge" can be transferred was shown to be wrong. It is information and communications about others' knowledge and the products of knowledge that can be shared (Beal et al., 1986). No one is merely a

passive "receiver" of information and technology since every one engages in the full range of knowledge processes as a condition of human survival (Seligman and Hagar, 1972; Maturana and Varela, 1992; Varela et al., 1993). Information about people's existing knowledge, attitudes and practices was found to be a poor predictor of their response to new ideas, messages, or technologies because knowledge processes and behaviors interact with the dynamic of people's immediate environment (Fishbein and Ajzen, 1975). The organization of processes for generating knowledge that is effective in action (Cook and Brown, 1999; Hatchuel, 2000; Snowden, 2005) was shown to take many forms. Where the rights of individuals and communities to be agents in their own development and considerations of equity, human health, and environmental sustainability were important policy goals, the comparative advantages of the ToT model also appeared less compelling (Jones and Rolls, 1982; de Janvry and Dethier, 1985; Swanson, 1984; Jones, 1986).

Practical: The mix of organizational support and services needed to gain maximum impact from the ToT model often were inadequate, imposed high transaction costs or were not accessible to the poor and to women (Howell, 1982; Korten and Alfonso, 1983; Ahmed and Ruttan, 1988; Jiggins, 1989). The positive role of local organizations as intermediaries in rural development was demonstrated but also the tendency for agricultural services organized along ToT lines to bypass these (Esman and Uphoff, 1984). The credit markets introduced to support technology adoption for instance typically were selective and biased in favor of resource rich regions and individuals (Howell, 1980; Freebairn, 1995) although pioneering initiatives such as the Grameen Bank in Bangladesh demonstrated that alternative approaches to the provision of microcredit to poor producers, women and farm laborers were possible (Yunus, 1982). Institutional analyses demonstrated how and why ToT arrangements that worked well in one context might fail to perform as well when introduced into other contexts. A recent authoritative assessment concludes that after "twenty-five years in which agricultural extension received the highest level of attention it ever attracted on the rural development agenda" political support for ToT in the form of "relatively uniform packages of investments and extension practices in large state and national programs" had disappeared (Anderson et al., 2006).

2.1.2.2 Other models of knowledge generation and diffusion processes

By the early 1970s, empirical studies and better theoretical understanding indicated that better mental models of knowledge processes were needed to guide practice if broader development goals were to be reached (Hunter, 1970). The first wave of institutional innovation in the organization of knowledge processes in noncommunist states sought to make more effective the process of moving science "down the pipeline" and technologies "off the shelf" by creating mechanisms and incentives for obtaining feedback from producers so that their local knowledge and priorities could