Abstract

An offshoot of decision analysis, risk-cost-benefit analysis (RCBA) dominates US policymaking regarding science and technology. In this paper a central normative presupposition of RCBA, called "the linearity assumption" is argued against. This is that there is a linear relationship between the actual probability of fatality and the value of avoiding a social risk or the cost of a social risk. The main object of this essay is to show that the presuppositions underlying the linearity assumption are highly questionable. It is maintained that assessors ought to give more consideration to broadening their interpretations of "unit cost" and "societal risk" and to abandoning their claims about linearity.