Managing Values in Science: A Return to Decision Theory.

There are many proposals in the literature on how to "manage values." Many of these proposals have in common the assumption that the relevant values in science can be "packaged for transfer": they can be put in an envelope for scientists to hand to stakeholders or policymakers, or for members of the public or ethical experts to hand to scientists. The central aim of this paper is to argue that packaging values for transfer is a practical impossibility. The central argument of the paper concerns the best stance to take on how values in science should be conceptualized. Specifically, I argue that we need to return to a decision-theoretic definition of values (as, I argue, is strongly suggested by a close reading of Rudner, Jeffrey, and Hempel.) Further, I argue for a picture of these values that is nonpsychologistic, stance relative, and always and everywhere entangled with credences. I call my account of the nature of values in science the Putnam-Hempel account. Unfortunately, the Putnam-Hempel account forces us to see that any proposal that depends on packaging values for transfer will ultimately suffer from great difficulties.