If policy-makers are to foster growth in most developing countries today, where economies remain heavily focused on agriculture, they must choose which of two ongoing trends to support. The choice is especially clear in sub-Saharan Africa, where suitable land and water for cultivation are scarce and where climate change is underway, compounding the shortages caused by population expansion. They can endorse the neocolonial trend of allowing foreign corporations to acquire the best land and water in order to work large holdings entirely for export. Or they can nurture the capacity of small farmers and irrigators in existing communities to grow crops for household subsistence while also producing a surplus, or even other special cash crops, for sale in domestic and international markets. The proposed research would attempt to do the latter by encouraging innovation, in a novel effort to enhance local food security while increasing cash incomes and fostering market growth. The aim is to increase the capacity of households and communities to produce for both purposes while also adapting to climate change, by promoting technologies that improve the efficiency with which basic resources are utilized, particularly water. These technologies-including both 'hard' and 'soft' types, new ones as well as old-have been adopted by farmers in certain parts of Tanzania, Malawi, and Bangladesh, in dynamic irrigation communities that would be sites of the proposed research. The study would first explore these local adaptations ethnographically in three selected villages, and then sponsor a programme of reciprocal knowledge exchange between their water-user groups. By means of the latter, the project would seek to expand the options available to the farmers residing in each place, thereby influencing--in ways that cannot be predicted but can be carefully observed--the direction and pace of change. Of great value in its own right, as a comparative study of adaptations to emerging rural markets and to climate change, the research would also break new ground in the field of participatory development.

Rice is one of the worlds most important crops, and it has a long history of supporting dense populations and civilizations throughout East, South and Southeast Asia. This project will reveal the history of rice cultivation comparatively across the region using cutting age archaeological science. One major aim is to reconstruct how rice was grown across the region at different times. Rice may be grown in wet cultivation systems (irrigated or flooded) and dry cultivation (based only on rainfall, often in upland areas), and in intermediate lowland, rainfed conditions. These different systems have important implications in terms of how productive rice is, and therefore how much human population it can support, as well as how labour-intensive it was. Dry systems yielded less but also cost less in terms of labour. How rice was grown has important implications for the impact that humans and rice had on environmental change. Intensive systems tend to require greater landscape modification and by supporting higher populations have knock-on effects on other resources, for example through deforestation. Another very important impact is the production of methane, a greenhouse gas that contributes to global warming. Dry rice cultivation systems produce little methane whereas the more productive wet systems produce a lot. It has been hypothesized by some climate scientists that methane from rice contributed to an anomalous rise in methane over the past 5000 years which is not explained by natural sources. If so, then this has contributed to global warming even before the industrial era and will need to be factored into models that hope to predict where global climate change is going. One of the aims of this project is to ground truth this hypothesis by modelling up from the empirical archaeological evidence for rice cultivation over time to assess whether this fits with explaining at least part of the methane anomaly. In order to do this we need better evidence not just for where and when rice was cultivated but also whether it was grown in wet or dry systems. Through systematic study of archaeologically preserved seeds, we can identify the weed flora associated with past rice and whether it fits with a wet or dry system. In addition we have developed methods for classifying the assemblages of phytoliths (microscopic silica from the decomposition of plants) from archaeological sites as indicating wetter or drier rice cultivation regimes. We are now hoping to apply these methods over a larger number of sites and regions, especially regions for which archaeobotanical evidence for early rice is limited or lacking, including parts of India (western and northeastern), Bangladesh, Myanmar, Cambodia, Vietnam, and southern China (Yunnan, Sichuan, Guangdong). By combining these new results in a GIS modelling system, together with data from other parts of the region, mostly collected by us and colleagues over the past few years, we will be better able to produce realistic spatial models of the spread of rice, the extent of wet rice, and likely methane emissions over time. We will also be able to improve our understanding of how the development of rice agriculture relates to the long-term history of human societies in this region.