Harm minimisation (HM) is a key aim of many actual or proposed public health policies (e.g. supply of clean needles/syringes for injecting drug users; drug testing at music festivals; promoting e-cigarettes over burning tobacco etc). In this talk I try and do two things. First, there is conceptual work to do. Can HM be clearly distinguished from related concepts such as harm reduction, harm prevention and harm avoidance etc? I will argue that it is important to be clear about what we are talking about. Second, assuming we have a clear concept of HM, where, if at all, is it justifiable to use it as a policy aim? I will argue that it would make no sense to have HM as an overall policy aim, as in at least some cases it makes sense to weigh the chance of harms arising from a policy against other important considerations (such as benefits). It looks as though cases where HM is most plausible as a policy aim are where people are held to be going to act in a harmful way anyway, and we seek to minimise the chances of (preventable) harm arising from such acts. I will explore the nature of this justification in two ways. First, we should note that it takes the form of a conditional, where the antecedent involves an empirical claim which may be contested. Second, it might be argued that HM necessarily involves complicity in harms that it would be better to prevent.