July 23, 2010
Dakar, Senegal

Classical observation schema of the lifecycle of a Development policy in the African countries is generally based on an alternative: either it is adopted and not enforced; or enforcement effort is undertaken and it is a linear design organized around three main steps: the implementation agenda, implementation and assessment. The printed by this summa divisio trajectory is inherently gradual; a step may only start if the previous step is completely refined. Similarly, each level is a mirror and a point of control of the one that follows. The implementation is intended to reflect the updated agenda and allows checking its reality; assessment is intended to certify for the good implementation and ensures compliance of the implementation with objectives arising from the implementation agenda. Also, it appears that the relationship between the different phases of the life of a development policy is transitive (placing on agenda in relation to implementation, implementation in connection with assessment, thus placing on agenda in relation to assessment). In this logic, the link between these different times is one-dimensional, unbalanced, and is always led by the dynamics of the placing on the agenda. The stream is still down.

This rigid vision of the lifecycle of public action, strongly inspired by the currents of rationalistic thought and planning development, appears somewhat questionable in terms of reality. It implies, total control of parameters of uncertainty associated with the interaction with environment, and minimization or elimination of endogenous and exogenous limits specific to a policy.

At the contrary, the empirical observation reveals that relations between these steps are more complex and asymmetrical than in the previous model, the latter is defining and influencing each other. So is it that implementation is considered as the operational and concrete level of the existence of a development policy. In the classical schema implementation is intended to implement objectives fixed at the conception level. It only occurs when the latter have already been fixed. Limited in its objectives, it is also limited in the ways, predefined in the policy. Makgetlaneng (2004) and Darbon (2006) have a completely different perspective by supporting that the implementation must be seen as a process of interaction between defining goals and implementation of actions to achieve them. Far from being rigid and following a predetermined path, implementation is characterized by an ability to adapt and continuous learning. It is a scalable enterprise that resets constantly by informing the other phases of the process.

Makgetlaneng and Darbon’s position can be seen as pessimistic as it argues that implementation is rarely completed and successful because of the complexity of the links that it covers. However, it remains fundamental and was constitutive of a new approach to implementation, more flexible and “open”. Particular, gaze was moved to concrete elements influencing the phenomenon (participants, the process, goals). Nevertheless, despite its relevance, it conceals significant theoretical and practical limitations.

In this second session of the ’ AGI’s Fridays’, we will address the difficulties of implementation of development policies in African countries. We will see that they carry many inherent barriers not only politics, but, often, depend on many exogenous, statutory, or institutional conditions. Then, we will endeavor to present, from a field perspective, some examples where ’it works’ and which may generate common practices for Developmental Governance.