Footnotes

• Correction notice This article has been corrected since it was first published online first. The references to the primary editorial have been updated to reflect the correct authorship.

• Competing interests None declared.

• Provenance and peer review Commissioned; internally peer reviewed.

• ↵i Non-coercive policies could remain justified alongside a ban. Warnings that highlight the dangers of drunk driving are one example of where governments seek to further deter behaviour that they have already prohibited. Tobacco policy could easily do the same.

• ↵ii In making these claims, I am not targeting Grill and Voigt's assumption that a smoking ban would be ‘effective’, which they make to set aside discussion of negative side effects and externalities, such as the stimulation of black markets, as well as fallibilities of the criminal justice system, such as racial bias in policing. In raising questions about the impact of a ban on autonomy, I am not simply raising questions about effectiveness, but about the moral cost of a ban even if it were effective.

• ↵iii I add the qualifier ‘paradigmatically’ to note that these remarks merely sketch the form of many familiar cases of coercion. Providing an entire theory of coercion (even legal coercion) is much more difficult.

• ↵iv See, for example, Husak.2

• ↵v (1986: 27).

• ↵vi Here, I leave aside the question of whether infringing autonomy by way of coercive threats might also reduce well-being, for example, through the threat itself or perhaps through the culture of fear that highly coercive public policy might encourage; nothing in Grill and Voigt's discussion suggests otherwise.

• ↵vii The classic discussion here is Feinberg.3

• ↵viii For a fuller exploration of these themes, see Duff.4

• ↵ix Husak (2013: 42) distinguishes taxes from ‘punitive measures’ characteristic of criminal law.

• ↵x For more discussion, see Halliday.5