What have we learned about policy learning in the EU? Theory and meta-theory compared

The European Union may well be a "learning organization", yet there is still confusion about the nature of learning, its causal structure and its implications for public policy. In this article we select four perspectives that address complexity, governance, the agency-structure nexus, and how learning occurs or may be blocked by institutional features. They are the political community of Deutsch and his collaborators, Cram's purposeful opportunism, experimental governance (Sabel and Zeitlin), and Scharpf’s joint decision trap. We appraise these four major contributions by considering three research questions. How do they investigate EU learning via their diagnostic and prescriptive modes of analysis? How do they vary in relation to meta-theoretical criteria? What is their theoretical leverage beyond EU studies? We first show how the four perspectives identify different learning types, and then explain how conclusions about learning depend on various causal arguments. Further, we move to a meta-theoretical mode to compare and appraise the four perspectives at a deeper level. Finally, we discuss the pros and cons of our approach, answer the research questions, and suggest how these learning contributions can be of use to the next generation of European integration scholars.