Anthony Mugenyi’s scientific contributions

Farmers, local extension workers and NGO field staff play increasingly more important roles in 'community based' agriculture and natural resource management projects or programs. Ideas about priority problems and how they might be solved are expected to come from the community. A bottom-up, participatory approach to project design and implementation is the operational hallmark of these projects. Suddenly, communities rife with conflicts over the exploitation of farmland and natural resources are expected to work together to conserve nature. Suddenly, farming systems that degrade the soil and pollute water resources are expected to become ecologically sound. Suddenly, government and non-government institutions that have had little experience working together are asked to form partnerships. While no one doubts the desirability of these changes, little time or resources are given to bring them about. Project participants are given no time to understand the perspectives of different communities about agroecosystems and their management. Indeed, little effort is given to finding out who the stakeholders are, let alone time for negotiating concerted action in the management of agroecosystems. No one should really be surprised when project evaluators report that most farmers are not participating in the project and that little will continue after the project ends. Equally we should not be surprised when the expected farming or conservation improvements have not been realized. Farmers are often not impressed with the impact of so-called 'improved' technologies. They complain that funds are attached to technological fixes that are inappropriate. Little effort is given to the development of knowledge systems for ecologically sound agriculture. Traditional knowledge about ecologically sound practices is rarely documented in a manner that is useful to other farmers. Project participants are rarely plugged into the growing international knowledge system of organic, ecological or alternative agriculture. There is little room in projects to learn and change. External monitoring and evaluation, the main opportunities for adjustments, usually provoke defensive attitudes in which mistakes are hidden rather than learned from. As local people do not have the capability or responsibility for evaluation, valuable lessons go unlearned. As local people remain isolated from external knowledge networks more appropriate technologies go unused. To complicate matters further, many African nations are decentralizing and privatizing much of their public sector agriculture support services. District level staff are now asked to respond to farmer demands and form partnerships with other service providers to meet those demands. The logic behind these policies is not only to save government money through sharing tasks and narrowing responsibilities; but also to provide better targeting of services and more efficient services. Building viable inter-institutional partnerships is hard, slow work. Moreover, responding to farmer demands requires considerable flexibility and dynamism of those organizations. How to create farmer demand for services and how to form viable partnerships is rarely a subject of study. There are few success stories and best practices to build on. In these circumstances the challenge confronting donors, central and local government officers, and project participants is to invent their own ways of working. This is partly because local conditions and complexities require a level of on-site innovation that cannot be satisfied by emulation of 'best practices'. This is not to say that the proverbial wheel must be reinvented everywhere, but best practices do need to be adapted to local conditions or better still, re-invented by local people. A capacity to re-invent and innovate is essential to partnership building and community progress in agroecosystem management. In this chapter we describe how a learning approach provides communities, local government and agriculture service providers with opportunities to learn their own way through to better partnerships and better management of agroecosystems. We describe an organizational structure for learning, a process for learning and several key instruments to facilitate learning. We conclude with a look at what progress has been made in the areas of community development of agroecosystem management strategies, farmer demand for agriculture support services, formation of partnerships, and capacity for local innovation. Our conclusion ends with a brief look at the constraints to further development of a learning approach to community-level agroecosystem management = In C. Flora (ed) Interactions between