At lunch the other day I overheard a conversation between a guest who, I gathered, had worked for the World Health Organisation (WHO) at some time in his career, and a couple of other people. The talk was of government attitudes to cigarette advertising. The visitor claimed that in 1950 after Richard Doll published his study in the British Medical Journal showing the causal connection between smoking and lung cancer, a discussion took place in HM Treasury about how the government should respond. Should it move to ban smoking, or at any rate to restrain tobacco advertising? In the end (so the speaker claimed) the decision was made to do nothing — on two grounds: (a) the tax revenues generated by the tobacco industry; and (b) the fact that higher mortality among smokers reduced the state’s pension burden. I’ve no idea if this is true, but if it were it would be a perfect example of realpolitik in action. The conversation then moved on to an analysis of why, in the end, the anti-cigarette campaign succeeded. The consensus was that the key decision was to focus on passive smoking, because this would, in due course, create a politically-unstoppable groundswell. It worked by neatly undermining the libertarian argument that smokers should be free to put their health at risk. They should not, however, be free to take risks with the health of others. I can imagine an A-level philosophy class studying JS Mill having great fun with this.