The present era can be called the 'Age of Scientific Assessment'. Governments on both sides of the Atlantic, intergovernmental organisations, NGOs, and private firms have all increasingly resorted to a variety of techniques, such as probabilistic risk analysis, pollution dispersion models, urban planning models, traffic-flow models, dose-response curves, and so on, ostensibly to guide the prudent use of resources to generate social and welfare and, increasingly, the natural environment. At the same time that the span of technocratic assessment has expanded, there has been a disconcerting decline in electoral participation in many industrialised countries. This paper suggests that there is a direct link between these two phenomena. Critics suggest that the science informing such assessments should be subjected to effective democratic participation and control. Social scientists have responded to this situation by designing ingenious ways to reconcile the conflicting demands of technical competence in making scientific judgments with popular participation in assessment and decision-making processes. Such techniques individualise values and represent the challenge of democracy as that of aggregating individual preferences. The paper opens the question whether these techniques are really the solution or, perhaps, might be part of the problem? It suggests that an answer to this question requires a radical rethink of our ideas about the institutions of science, democracy, and resource management.