Turtles all the way down: bounded rationality in an evidence-based age

Evidence-based policy-making (EBPM) has become both a catch-cry and aspiration of governments. Drawing on ‘the evidence’, governments seek to focus on ‘what works’ and so avoid the pitfalls of policy driven by ideology or values. Critics of EBPM remind us that the policy process remains messy and uncertain and that while research may deliver the latest scientific evidence, it is not always translated effectively into policy. We argue that the problems with EBPM run much deeper. This is not only because of the way in which knowledge is employed by policy-makers but because of the way in which knowledge is collected and communicated to policy-makers. It is well understood that policy actors are boundedly rational and this impacts upon the way in which evidence is used in the policy process. What has not been clearly articulated and is the focus of this article is that bounded rationality applies equally to the production of evidence. Drawing on the work of Polish microbiologist Ludwick Fleck we discuss how the process of systematically collecting and communicating evidence can undermine the aspirations of EPBM. We illustrate this argument with reference to recent policy debates about obesity.