by Stephen Whitfield
Outlook on Agriculture, Volume 41, Number 4, December 2012 , pp. 249-256(8)

Evidence-based policy represents an emergent discourse in African agriculture and is welcomed by many for the emphasis it places on the legitimization of policies and strategies through reference to observed realities. Its intuitive premise places realized results, as opposed to theory or bias, at the foundation of policy making. However, the universal appeal of evidence-based policy, as demonstrated by the geographical and inter-sector spread of the discourse, belies the fact that its legitimacy relies on a set of prerequisites that are by no means universally established. This paper highlights some of the current incompatibilities between a leaning towards evidence-based policy in African agriculture and various issues that currently compromise the quality of national agricultural statistics across the African continent. The case of NERICA rice is used to highlight how ‘success stories’ – which may become an evidence base of their own, justifying scaled-up investments and technology delivery – may be successfully constructed on the basis of weak or incomplete evidence. It is argued that the virtues of evidence-based policy rely critically on the quality of evidence and transparency in the way evidence speaks to policy, such that weaknesses do not become lost in a process that distorts data into policy truths.