Epistemic and non-Epistemic Values in Science

I contend that science's value dependence has at least four different forms, which give us four different roles that values play in the scientific endeavor: (1) values functioning as selectors among different fields of investigation; (2) values functioning as selectors among alternative, empirically equivalent theories or hypotheses (these are often referred to as epistemic values); (3) values functioning as "regulative" ideas of science, that is, as indicators of the place and meaning that the scientific enterprise as a whole should have in our culture, in society, and in our life in general; (4) values functioning as guides to the application of our scientific knowledge and technology to practical decision making. Under the fourth role I classify cases in which the piece of knowledge to be applied is highly reliable as well as cases in which our knowledge is instead partial and limited, generating nonnegligible "inductive risks" in the sense introduced by Carl Hempel (1965) and further developed by Heather Douglas (2000)