¹ Luke 10:35.

² See, e.g., Augustine, De sancta virginitate (in Patrologiae cursus completus, ed. by J.-M. Migne, vol. 40, Paris: Garnier, 1841; De sancta virginitate written about 401), chap. 5.

³ Alexius Meinong, Psychologisch-ethische Untersuchung zur Wert-Theorie (Graz: Leuschner & Lubensky, 1894).

⁴ J. O. Urmson, ‘Saints and Heroes', in Essays on Moral Philosophy, ed. by A. I. Melden (Seattle: University of Washington Press, 1958); Chisholm, Roderick M., ‘Supererogation and Offence’, Ratio 5 (1963), 1–14 Google Scholar; David Heyd, Supererogation (Cambridge: Cambridge University Press, 1983); Gregory Mellema, Supererogation, Obligation, and Offence (New York: State University of New York Press, 1991); McNamara, Paul, ‘Making Room for Going Beyond the Call of Duty’, Mind 105 (1996), 415–450 CrossRefGoogle Scholar.

⁵ Nor with the other question the Fathers of the Church had on their mind: how to use the surplus of merits the saints gained by their supererogatory actions? As is generally known the answer to this question led to the selling of indulgences.

⁶ This question has been asked for all major types of ethics: for consequentialism or even utilitarianism, e.g., by Portmore, Douglas W., ‘Position-Relative Consequentialism, Agent-Centered Options, and Supererogation’, Ethics 113 (2003), 302–332 CrossRefGoogle Scholar, and Vessel, Jean-Paul, ‘Supererogation for Utilitarianism’, American Philosophical Quarterly 47 (2010), 299–319 Google Scholar; for Kantian ethics, e.g., by Timmermann, Jens, ‘Good but Not Required?’, Journal of Moral Philosophy 2 (2005), 9–27 CrossRefGoogle Scholar, and Marcia Baron, ‘The Supererogatory and Kant's Wide Duties’, in Reason, Value, and Respect, ed. by Robert Johnson and Mark Timmons (Oxford: Oxford University Press, 2015); and for virtue ethics, e.g., by Mellema, Gregory, ‘Moral Ideals and Virtue Ethics’, Journal of Ethics 14 (2010), 173–180 CrossRefGoogle Scholar, and Claire Michelle Brown, Supererogation for a Virtue Ethicist, PhD dissertation (Notre Dame: University of Notre Dame, 2011).

⁷ It is most clearly presented and defended by Michael Slote (Common-Sense Morality and Consequentialism (London: Routledge & Kegan, 1958), chap. 3, and ‘Rational Dilemmas and Rational Supererogation’, Philosophical Topics 14 (1986), 59–76 CrossRefGoogle Scholar) and James Dreier (‘Why Ethical Satisficing Makes Sense and Rational Satisficing Doesn't’, in Satisficing and Maximizing, ed. by Michael Byron (Cambridge: Cambridge University Press, 2004)).

⁸ Throughout this paper I use the expressions ‘going beyond the call of duty’ and ‘doing something that is supererogatory’ as if they were synonymous – although Paul McNamara (‘Supererogation, Inside and Out’, in Oxford Studies in Normative Ethics, vol. 1, ed. by Mark Timmons (Oxford: Oxford University Press, 2011)) has recently argued that they are not. If McNamara is right about this, the model I will suggest is a model for ‘going beyond the call of duty’ rather than for supererogation. For example, McNamara deems supererogation (unlike going beyond the call of duty) to depend on the agent's intentions in a certain way that is not covered by that model. I believe that the sensitivity to intentions that McNamara has in mind could be incorporated into it, but that, of course, is a different matter.

⁹ Supererogatory actions are good or even very good, but not obligatory. In doing them the agent goes beyond the call of duty, and to say that the agent goes beyond the call of duty, but does something that is impermissible would be odd. After all, if the agent goes beyond the call of duty, she at least fulfils all the obligations that apply – not matter what else she does. Hence, whatever supererogatory actions may be, they are permissible. This has been pointed out by many, e.g. by David Heyd, op. cit. in note 4, 120–5, and by Paul McNamara, op. cit. in note 8, 219.

¹⁰ Some think otherwise, among them Michael Slote (op. cit. in note 7, 47f): ‘A satisficing concern for good results […] [is] permissible in some cases […] where there is no issue of personal sacrifice on the part of the agent’. However, most authors agree that supererogation also depends on what is at stake for the agent – even if they disagree how much has to be at stake. While David Hyde (op. cit. in note 4, 2), e.g., thinks that sometimes even small favours suffice, Jackson, M. W. (‘The Nature of Supererogation’, The Journal of Value Inquiry 20 (1986), 289–296 CrossRefGoogle Scholar) and many others believe that much greater efforts like ‘repeated sacrifice of self-interest’ or even risk of losing one's life are required. Saying that the size of the agent's sacrifice matters does not commit those authors to departing from the threshold model, though. The threshold is one of goodness (when enough good is done, every further grain of good done is supererogatory), but nothing in the threshold model excludes (i) that the threshold differs from situation to situation and (ii) that the question where the threshold lies in a situation depends on the costs the agent would incur by performing the various actions open to her.

¹¹ For further objections see Ulla Wessels, Die gute Samariterin (Berlin: de Gruyter, 2002), chap. 1.5.2.

¹² Something along these lines has recently been suggested by Martin Petersen (The Dimensions of Consequentialism (Cambridge: Cambridge University Press, 2013)) and by others before him – e.g., by Eriksson, Björn (‘Utilitarianism for Sinners’, American Philosophical Quarterly 34 (1997), 213–228)Google Scholar.

¹³ Is the threshold condition plausible? One might be inclined to think that it is at the same time too weak and too strong – that, on the one hand, an action cannot be supererogatory unless something is morally gained by it and that, on the other hand, an action can also be supererogatory if it doesn't impose a burden on the agent. Think, however, of a situation in which a woman risks her life to save someone else's child from a house that is on fire – she does so instead of the owner who would have done the same so that nothing is morally gained by her action. Her action is supererogatory all the same. And, similarly, think of a situation in which a man donates one of his kidneys to a dialysis patient whom he does not know – because he has three of them and would be better off with just two. Since his action does not impose a burden on him, it is not supererogatory.

¹⁴ Here are two real-life examples for such an ‘obligational pull’: a person who works twenty hours a week for a charity that spends a considerable share of the donations it receives on its own administration – though she could change her mind and work the same number of hours for another, more efficient organization; or a company that provides a certain amount of money to fund a youth centre while knowing that the amount does not quite suffice to also fund a position for a social worker who would make an immense difference to the good that the youth centre can achieve.

Of course, the examples are bound to raise new questions. Perhaps the company deliberately donates an amount that does not quite suffice just in order to make the local council realize its own responsibility. And perhaps the person who spends twenty hours a week working for an ‘inefficient’ charity does so because this is the only charity towards which she feels a strong affinity. Such things occur. But it also occurs that, along the lines illustrated by the examples, the fact that someone is going to perform an action at least as good as a certain prima facie supererogatory action generates an obligation to further the good even more than she would by performing that action.