Flesh and bones: Working with the grain to improve community management of water

Despite cogent critiques and limited successes, community-based management (CBM) remains central to policies for natural resource management and service delivery. Various approaches have been suggested to strengthen CBM by ‘working with the grain’ of existing social arrangements and relationships. For advocates, such approaches ensure that management arrangements are rooted in local realities and are therefore more likely to be effective. Implementing this approach is, however, methodologically, empirically, and operationally challenging. In this paper, we centre these challenges through a study of community-managed water in rural Ethiopia, Malawi, and Uganda. We examine water management arrangements by undertaking an in-depth social survey of 150 communities in the three countries. We also undertake yearlong studies in 12 communities in Malawi and Uganda involving 30 diary keepers. This focus on the local is complemented by country-level political economy analyses and district-level sustainability assessments. Our multi-country extensive-intensive research design uncovers the flesh and bones of CBM, and provides explanations for our findings. In Ethiopia, water management arrangements are more likely to be fleshed out – fully formed committees often working in conjunction with other institutions. In Malawi and Uganda, water management arrangements tend to be skeleton crews of key individuals. The position we adopt is located between advocacy and critique. We recognise the potential of working with the grain. We also recognise the considerable challenges of operationalising this approach without reducing it to another standardised checklist or toolbox. In an attempt to reconcile this tension, we identify practical entry points and sketch out requirements for a more socially informed, reflexive, and effective approach to working with the grain. Whether this can be operationalised within the logics of mainstream development, and whether it can ‘save’ the CBM model, remain open questions