Abstract

The paper critically examines how evaluation gains visibility as a «soft» policy instrument in the current policy making climate shaped by the rhetorical appeals for neutrality and sound evidence, and which consequences can be observed on the kind of legitimate knowledge devices and actors that are called to use them, admitted into the policy arena. In this framework we analyse specifically the experimental methods and performance analysis systems, often recognized as legitimate devices to provide value free accounts able to prompt reassuring effects on the complexity of decision making processes. Our intention is not prescriptive, we do not seek to identify the better instruments of evaluation nor to deny the potential worthiness of the mentioned devices, rather we aim at removing the «taken-for granted» character of their use, stressing their problematisation and insisting on the need of surfacing the assumptions underpinning their choice, that are often doomed to remain in the dark. For this purpose, we use the technical dimension as a preferable point of view to surface the logic of reasoning underpinning their adoption, outlining how they construct and legitimize interpretations of truth and causal links between behaviour and events, and vice versa how they are effected by them.