IWRA Proceedings

AbstractPolicy-oriented learning is critical to water governance in a context of constant change. The complexity of water systems and the unpredictable changes require adaptive water governance. Policies should be based on the best available knowledge and leave room for flexibility in their implementation. Their implementation should be monitored and evaluated at regular intervals, to enable learning about a policy’s impacts as its implementation unfolds. This should ensure that adjustments are made when necessary and that emerging insights feed into subsequent policy cycles. Learning is required about the physical water system and about the human system through which policy measures affect the physical resources base. Unfortunately, such learning is complicated by numerous factors, including a lack of monitoring data, time-lag effects between the implementation of policy measures and their impacts, as well as thresholds and leaps in nonlinear ecological processes. Also, water governance involves various actors with different roles and responsibilities. It is difficult to understand how institutional and socio-economic policy measures work in these multi-actor networks, as well as to organize collaborative learning among those actors. Especially for this multi-actor dimension, there seems to be a need for new tools and approaches that facilitate learning for water governance. Model-based actor analysis offers a promising approach to facilitate such learning that addresses the multi-actor complexities. Model- based actor analysis approaches cover several methods that help to understand the perceptions, resources and values of the actors involved, and/or the structural characteristics of actor networks. This helps to gain insight into the factors that influence the outcomes of actors’ interactions. Past application of these approaches for policy development suggest that they yield interesting new insights and have the potential to contribute to the interaction and learning processes among actors. Especially the models that analyze actors’ perceptions, such as dynamic actor network analysis, are promising for policy learning. These models are compatible with the idea of theory-driven evaluations, using actors’ inputs to reconstruct critical assumptions behind policy mechanisms and to identify different success criteria. The paper will illustrate the use of dynamic actor network analysis to evaluate early experiences with efforts to support the implementation of the EU Water Framework Directive in Turkey. The results show some important strengths and limitations of the application of the Water Framework Directive to support water governance in Turkey. More generally, the findings confirm that model-based approaches for actor analysis can make an important contribution to policy-oriented learning for better water governance.