Tackling corruption has proven to be one of the most intractable policy problems for governments and civil society worldwide. From the past 20 years, it is clear that there are few classes of crime which are so capable of diverting resources and perverting policy on grand and petty scales, yet so intrinsically linked with larger political and economic forces, intertwined with the self-interests of influential social actors, and so interwoven with other social behaviours, both positive and negative. As a consequence, there are few crime problems for which solutions are so systemic and elusive, at the same time.However, while a rich literature continues to develop on how corruption problems are best diagnosed and measured, 1 far less expert attention has been devoted to how best to evaluate the presumed answers to corruption. In particular, only limited attention has been placed on how to evaluate the policy systems, institutions and processes aimed at bolstering corruption's antithesis -integrity in the discharge of office, performance of institutions and exercise of public or other entrusted power. This is notwithstanding the obvious importance of understanding why so many anticorruption efforts seem ineffective, and of identifying what makes for relative success. Overall, anti-corruption research has given much less focus to the origins of 'high-