Abstract

This paper defends a probability-raising theory of what it is to be a means to an end, and how much reason transmits from an end to its means. In short, an action is a means to an end insofar as it raises the probability of the end relative to the worst one could do. The paper also considers and criticizes several alternative probability-raising theories as well as non-probability-raising conditions on being a means and being supported by means-based reason.