The management of staple crops as perennials is a historic legacy and a present-day strategy in some regions of Sub-Saharan Africa, yet perenniality is rarely an agronomic subject. Farmers in Malawi cut annual crops, such as pigeonpea and sorghum, to extend production for more than one growing season. Cassava, a perennial food crop, has a proven track record of abating hunger. Here we review ratooning, as well as the historic role of perennial staple crops in Malawi. Ratooning is a method of harvesting a crop which leaves the roots and the lower parts of the plant uncut to give the ratoon or the stubble crop. This review is completed with interviews with Malawian farmers. The major points follow. The management of staple crops as perennials is underserved by research. Indeed, we retrieved only 86 references on ratooning sorghum and pigeonpea. Of these, 9 % and 19 % respectively were from the African continent. The literature and interviews indicate that pigeonpea and sorghum have high productive potential when well managed in ratoon systems. Thirty-five percent of interviewee responses that supported ratooning mentioned saving seed. Other primary reasons to ratoon include stimulating regrowth (30 %) and saving labor (20 %). However, 31 % of responses that were against ratooning cited increased disease potential, as well as excessive vegetative regrowth (18 %).

Participatory scenario processes are associated with positive social learning outcomes, including consensusbuilding and shifts toward more systemic thinking. However, these claims have not been assessed quantitatively in diverse cultural and socio-ecological settings. We convened three stakeholder workshops around the future of agricultural development and rural livelihoods in Burkina Faso, Nigeria, and Malawi, using a participatory scenario generation process to examine proposed research and action priorities under conditions of uncertainty. We administered pre-and post-workshop surveys, and used a paired t-test to assess how stakeholders' rankings of research priorities changed after participating in the scenario visioning exercise. Workshop participants also listed their own priorities for research and implementation on both the pre-and post-survey forms. We found indications that the workshops promoted consensus-building around the research priorities, including a reduction in standard deviation of priority rankings post-workshop compared to pre-workshop; and a higher incidence of identical volunteered responses. We did not find evidence to support shifts in thinking to more systemic views of agricultural development. However, participants viewed themselves as having learned throughout the process. We conclude that scenario visioning does have the potential to foster consensusbuilding (one element of social learning) among diverse stakeholder groups. We urge researchers to continue to monitor and measure systems thinking outcomes from scenario visioning so that these processes may be designed to be more effective. 'game-changing' events which could alter the future state of a system (Swart et al., 2004). Scenarios which are generated by stakeholders themselves rather than by scientists or quantitative models are uniquely wellsuited to participatory planning and group exploration of the future of a complex system (Kok et al., 2011). Recent scenario planning exercises around agriculture in Africa include planning and development strategies in Tanzania (Enfors et al., 2008); and the Climate Change, Adaptation and Food Security scenario planning activities conducted by the Consultative Group on International Agricultural Research (CGIAR) Future Scenarios research team in East and West Africa (Vervoort et al., 2014).

No abstract