Most organizations working on agricultural research and rural development in the Third World now acknowledge that any proposals for technological innovation must take into consideration the reasons that farmers use in making their decisions. A form of systems analysis has been developed as a main tool for this purpose. In this approach, differences in the definitions and concepts revolve mainly around the scale of activities observed at the plot, farm, village or regional level. In the search for a method that can satisfy all users, there has been a tendency to contrast the approaches of the anglophone Farming Systems Research with those of the francophone "Recherche - Développement". The analysis of some field cases, however, shows that there may be more similarity between these approaches than is often acknowledged. That being said, some differences can be identified and they are not just a matter of linguistics. Such differences are due first of all to the institutional priorities of the organizations sponsoring the research. Those institutions whose goal is to promote techniques developed by agricultural research, generally use "top-down" approaches where the emphasis is on diffusing techniques developed in research stations. Those institutions whose mandate is to promote the development of agrarian societies without reference to any particular discipline, generally use a "bottom-up" approach wherein the emphasis is placed on working with the existing agricultural system. Finally, those operations concerned primarily with macro - economic factors take a more "political" approach, often called Farming Systems Perspective. Such methodological distinctions cannot be used to differentiate between anglophone and francophone approaches because they exist within programmes belonging to both of these approaches. The question of the rapidity of the diagnosis and the attention paid to agronomic analysis, however, appear as the main variables of these methods.