Abstract
Theoretical writings on policy analysis have often dealt with the problem of adjudicating value issues as well as the problem of complexity. Since both these problems are ultimately epistemological in character, this paper explores their structure from a less familiar non‐foundational, coherentist epistemological framework, within the context of the synoptic/anti‐synoptic debate in policy analaysis. The paper argues against a sharp distinction between fact and value in policy choice, and recommends the gradual replacement of ubiquitous folk psychological simplifying assumptions in favour of scientific ones, where predictions systematically lapse, as a step towards dealing with complexity.

This publication has 11 references indexed in Scilit: