Until recently, participatory and conventional approaches to agricultural research have been regarded as more or less antagonistic. This article presents evidence from three sub-projects of a Thai–Vietnamese–German collaborative research programme on ‘Sustainable Land Use and Rural Development in Mountainous Regions of Southeast Asia’, in which participatory elements were successfully integrated into conventional agricultural research as add-on activities. In all three sub-projects the costs of studying local knowledge or enhancing farmers' experimentation consisted of additional local personnel, opportunity costs of participating farmers' time, and travel costs. However, these participatory elements of the research projects constituted only a small fraction of the total costs. It may be concluded that conventional agricultural research can be complemented by participatory components in a cost-effective way, while producing meaningful benefits in terms of creating synergies by blending scientific and local knowledge, scaling up micro-level data, and highlighting farmers' constraints affecting technology adoption.

Co-operation between researchers in the global North and South is critical to the production of new knowledge to inform development policies. However, the agenda-setting process is a formidable obstacle in many development research partnerships. The first section of this article examines how bilateral donor strategies affect collaborative agenda-setting processes. The second section explores researchers' motivations for entering into North–South partnerships; the obstacles that Southern researchers encounter in agenda-setting processes; and the strategies that they employ to ensure that research partnerships respond to their concerns. This analysis suggests that while strong Southern research organisations are best placed to maximise the benefits of collaboration, donors and researchers alike are well advised to recognise the limitations of this approach and use it prudently, because North–South partnerships are not necessarily the best way to advance research agendas rooted in Southern priorities.

For the purposes of accountability and uniformity, and as a way of giving insight into their intellectual capital regarding development practices, NGOs in Southern Africa are required by donor agencies to describe their intended activities in very clear, unambiguous terms. These requirements may include the expression of theoretical approaches, the development of logical frameworks, clear objectives, indicators for success, criteria for sustainable development, and relationships to government policies. However, the interface between reality and these planning measures and tools, most often completed without the input and contributions of the communities whom they are to serve/service, produces a much more messy, dynamic, and involved picture of the development process. None the less, the NGOs are still required to be accountable on the basis of their original proposal and planning. The author presents examples of this phenomenon and discusses the challenges facing an evaluator when dealing with competing principles of accountability, autonomy, and authenticity.

News about Norway's plans to establish a ‘doomsday vault’ for seeds in the permafrost of the Artic archipelago of Svalbard as a back-up for conventional gene banks reached the world press in 2006. The idea of a Global Seed Vault, which today is considered a ‘Noah's Ark’ for seeds, was previously regarded with suspicion and considered to be unrealistic. In 1989 the Norwegian government offered to construct an international depository for seeds in permafrost, but the initiative was sidelined in the agitated debates between developed and developing countries over access to and control of plant genetic resources. The realisation of the FAO International Treaty on Plant Genetic Resources for Food and Agriculture (2004) resolved some of the most difficult issues and made possible the launching of a new Norwegian initiative to safeguard some of the world's most important plant genetic resources for the future.

This article undertakes a critical re-evaluation of a DFID-funded project in South Africa which ran between 1998 and 2001. The evaluation sought to test whether the development of community-led indicators would improve governance. Since the project ended, a series of papers have been published that are critical of such participatory methods, arguing particularly that they are apolitical and adopt a technocratic approach. In the light of these criticisms, this article re-assesses the DFID project, following on from the initial evaluation carried out by the author in 2001. Sobantu, a black township in Pietermaritzburg, was one of the original project sites. It was chosen as the subject for research because the local implementing agency was a politically astute, well-connected institution that understood the political nature of the process required to develop the indicators. Although the project achieved some positive outcomes, the long-term commitment to the indicators has since been compromised. This was in large part due to the inability of community members to engage meaningfully with key municipal service providers. However, recent changes to the South African planning regime might provide opportunities for the indicators to become more useful again.

This article explores efforts to bridge multi-disciplinary research and policy engagement to tackle child poverty in the contexts of developing countries, based on the experiences of Young Lives, an international longitudinal policy-research project. It focuses on a case study involving the application of research evidence on child poverty to shape policy debates concerning Ethiopia's second-generation Poverty Reduction Strategy Paper (2006–2010). The discussion is situated within theoretical literature on the interface between knowledge, policy, and practice, which supports the conceptualisation of policy making as a non-linear dynamic process. It pays particular attention to the importance of understanding the political and policy contexts of Southern countries, rather than assuming that they should simply import Northern-derived models of advocacy. It concludes by identifying general lessons for translating research into social-policy change.

International Wheat and Maize Improvement Center (CIMMYT) projects on new resource-conservation technologies (RCTs) in the Indo-Gangetic Plains of Nepal aimed to strengthen equity of access, poverty reduction, and gender orientation in current rural mechanisation processes – more specifically, to promote machine-based resource conservation and drudgery-reduction technologies among smallholder farmers. These projects, together with other projects and other actors, gave rise to an informal ‘coalition’ project, which used participatory technology development (PTD) approaches, where farmers, engineers, scientists, and other partners worked towards equitable access to new RCTs. This experience showed that PTD projects need to be flexible, making use of learning and change approaches. Once successful adoption is occurring, then what? Such projects need to ensure that everyone is benefiting in terms of social inclusion and equity; this might necessitate new unforeseen work.