Reformulating Mill’s Harm Principle

Mill’s harm principle is commonly supposed to rest on a distinction between self-regarding conduct, which is not liable to interference, and other-regarding conduct, which is. As critics have noted, this distinction is difficult to draw. Furthermore, some of Mill’s own applications of the principle, such as his forbidding of slavery contracts, do not appear to fit with it. This article proposes that the self-regarding/other-regarding distinction is not in fact fundamental to Mill’s harm principle. The sphere of protected liberty includes not only self-regarding conduct, but also actions that affect only consenting others. On the other hand, the occasional permissibility of interfering with self-regarding conduct can plausibly be explained by reference to the agent’s consent. Thus, the more important distinction appears to be that between consensual and non-consensual harm, rather than that between the self-regarding and non-self-regarding action. That is, interference can be justified in order to prevent non-consensual harms, but not to prevent consensual harms. It is argued that the harm principle, thus reformulated, both captures Mill’s intentions and is a substantively plausible position.