¹ Gerald Dworkin, ‘Paternalism’, The Stanford Encyclopedia of Philosophy, <http://plato.stanford.edu/archives/sum2010/entries/paternalism/> (2010).

² I focus on Dworkin's account of the paternalist act rather than other accounts because the form of his account is something that I seek to reproduce in my own account and by examining his account I can move neatly to an account of the paternalist act that I think is right (Dworkin, ‘Paternalism’).

³ Solely for presentational purposes, I’ve altered the symbols marking the identity of the agents, the labelling of each of the conditions, and the spacing.

⁴ I say ‘improve the welfare’ for shorthand. My intention is not to challenge Dworkin's fuller description.

⁵ Nevertheless I haven't found clear textual evidence that he does have this in mind. Dworkin writes that he is providing an analysis of paternalist action but he does not write that a paternalist act occurs if and only if the conditions he lists are met. If the conditions he provides are intended only to be of paradigm cases of paternalist acts, then let my comments on his analysis be seen as indicating what I believe a more basic analysis would look like.

⁶ As well as S having some epistemic standing in favour of Z, it might be thought that S should also not have any epistemic standing of equal or greater strength against Z: so S does Z just because S is motivated by a positive epistemic standing that doing Z may improve R's welfare and does not also have an epistemic standing of the same or greater strength that doing Z may or will disimprove R's welfare. (If S just had say a strong belief that Z wouldn't improve R's welfare then that wouldn't be a reason against acting on the basis of a weaker epistemic standing to do Z.) Requiring that the standing for the paternalist act be greater than or equal to the standing against seems to presuppose that the paternalist actor must be exhibiting rationality in a particular way, but this seems unnecessary.

⁷ I shift in this paragraph from discussing the paternalist act to the paternalist for stylistic reasons. For the sake of clarity, it should be born in mind that the intended object of my analysis is the paternalist act. The need for clarity here arises from the fact that someone who defends the theory of paternalism, even if they have never acted in a paternalistic way, may be described as a paternalist. This implies that a paternalist need not just be one who acts in a paternalistic way.

⁸ I couch this condition in terms of what S believes because we can conceive of a world in which S is actually acting irrespective of the wishes of R but has the false belief that he is acting in accordance with the wishes of R.

⁹ Shiffrin, Seana, ‘Paternalism, Unconscionability Doctrine, and Accommodation’, Philosophy and Public Affairs 29 (2000), pp. 205–50, at 213CrossRefGoogle Scholar.

¹⁰ The argued-for rejection of C2 in favour of C1* plausibly further contributes to the category of the paternalist act being broadened.

¹¹ In a paper currently in preparation I argue that both the more permissive conditions of the paternalist act and the adjustment to the motivation condition are such that the libertarian paternalist act also satisfies the requirements of the paternalist act. Indeed the adjustment to the motivation condition, specifically that the paternalist actor acts because of a positive epistemic standing that the action will or may improve the welfare of the targets of the action, allows us to make sense of how libertarian paternalist social policies, with their impact on large numbers of people who are unknown to libertarian paternalist actors, can be understood as being properly motivated so as to qualify as paternalist acts.

¹² Obviously the acts are only briefly described and further details might swing a person's evaluation one way or another. It seems plausible though that details could be filled in about the acts such that the acts remain relatively commonplace while a general evaluation of the cases as morally permissible holds.

¹³ It should be noted that paternalism is taken to involve not only interference in or the violation of autonomy, but a lack of respect for the autonomy of the agent who is the object of the paternalist act. For an example, see Joel Feinberg, Harm to Self (New York, 1986), p. 25. An alternative objection to the one considered here would be to claim that paternalist acts of the sort described are wrong in that they manifest a lack of respect for the autonomy of the relevant agents. I don't believe, however, that such acts need imply a lack of respect for the autonomy of the relevant agents.

¹⁴ I address the relevance of shifting to knowing later.

¹⁵ Independent of what has been said here, there is often a practical reason to defer to the wishes of an agent in matters of what is best for them; that is, that generally they’re in a better position than others to know what is best for them.

¹⁶ The approach of always deferring to another's wishes in matters of their good seems to invite not caring about what the good is of another person, regarding it as being their concern, and not looking beyond whatever wishes they express. While it seems plausible that this will be the effect of this view for some adherents, this needn't be the case. We could have the same headline view, that one should defer to the wishes of another in the sort of cases described above, but also hold that it's permissible or good if one looks out for another person – one tells them if one believes one can do something that might improve their welfare. On this approach it still wouldn't be permissible to act irrespective of their wishes.

¹⁷ Of course, we might think that a number of different mother–adult son relationships can be appropriate.

¹⁸ All this is not to say that paternalist acts don't raise concerns. These concerns vary from slippery slope concerns that there will be encroachment into our legitimate areas of autonomy, to concerns about the harm that might be caused by bungled and sham paternalist acts. By sham paternalist acts I mean acts done in the name of improving the welfare of another, irrespective of her wishes, that are in fact not done for the sake of the other's welfare.

¹⁹ Groll, Daniel, ‘Paternalism, Respect, and the Will’, Ethics 122 (2012), pp. 692–720, at 695–6CrossRefGoogle Scholar.

²⁰ A quote from Dworkin describing the paternalist act also makes Groll's list, but I’ve omitted it to avoid repetition. (The quote is not in the 2010 version of Dworkin's Stanford Encyclopedia article that Groll cites, though it is in Gerald Dworkin, ‘Paternalism’, The Stanford Encyclopedia of Philosophy, <http://plato.stanford.edu/archives/sum2009/entries/paternalism/> (2009).)

²¹ Arneson, Richard, ‘Mill versus Paternalism’, Ethics 90 (1980), pp. 470–89, at 471CrossRefGoogle Scholar.

²² Feinberg, Harm, p. 5.

²³ Groll, ‘Paternalism’, p. 696, n. 10.

²⁴ Archard, David, ‘Paternalism Defined’, Analysis 50 (1990), pp. 36–42, at 36CrossRefGoogle Scholar.

²⁵ Beauchamp, Tom, Standing on Principles: Collected Essays (Oxford, 2010), pp. 101–19, at 103Google Scholar.

²⁶ Feinberg, Harm, p. 7.

²⁷ Feinberg, Harm, p. 5.

²⁸ Shiffrin, ‘Paternalism’, p. 218.

²⁹ Groll, ‘Paternalism’.

³⁰ Raz, Joseph, Practical Reason and Norms (Oxford, 1999)CrossRefGoogle Scholar and Darwall, Stephen, The Second-Person Standpoint: Morality, Respect, and Accountability (Cambridge, MA, 2006)Google Scholar.

³¹ Groll, ‘Paternalism’, p. 707, n. 32.

³² A referee from Utilitas provided me with very helpful comments on an earlier version of this article, as did a referee reviewing quite a different version of this article for another journal. I thank both. I also thank Elizabeth Ellis and Duncan Pritchard for their feedback.