Document Type


Publication Date

March 2005


Natural resource regulation is heavily "scientized," by which we mean both that the current regulatory structure requires the use of science in a wide range of decisions, and that decisionmakers generally emphasize the role of science in those decisions. Nonetheless, critics on all sides of the political spectrum claim to believe that regulatory decisions remain too political and insufficiently scientific. Administration of the Endangered Species Act (ESA) in the Klamath Basin illustrates the challenges of scientifically managing nature. A series of science-based decisions are needed, from species listing to consultation on federal actions. Those decisions carry substantial costs for the people who share the landscape with protected species. Unless science can provide some level of confidence that management actions are both necessary and effective, those decisions will be widely perceived as unfair. The key question, not yet answered, is just how much confidence should be expected. There may well be points in the decisionmaking process at which greater objectivity would be desirable. As we argue in more detail below, however, science can never provide the perfect rationality we have been conditioned to expect from it. Therefore, simplistic generalized demands for objective rationality are not a useful reform strategy. Typically, the disputes are fundamentally about how incomplete data are interpreted and applied, rather than about what the data are or how they have been gathered. Agency judgments, in other words, are the real issue. It is impossible to entirely prevent the exercise of judgment, influenced by the subjective values and biases of the decisionmaker, from creeping into decisions. A more useful inquiry would take a closer look at the role of judgment, asking at what stage and through what mechanisms it factors into resource management decisions, whether the effect of those judgment steps is to advance or retard the identification and achievement of societal goals, and, when correction is needed, how judgments might be more closely constrained.