Skip to main content
    • Aa
    • Aa


  • Terry Horgan (a1) and Mark Timmons (a1)

In his 1958 seminal paper “Saints and Heroes”, J. O. Urmson argued that the then dominant tripartite deontic scheme of classifying actions as being exclusively either obligatory, or optional in the sense of being morally indifferent, or wrong, ought to be expanded to include the category of the supererogatory. Colloquially, this category includes actions that are “beyond the call of duty” (beyond what is obligatory) and hence actions that one has no duty or obligation to perform. But it is a controversial category. Some have argued that the concept of supererogation is paradoxical because on one hand, supererogatory actions are (by definition) supposed to be morally good, indeed morally best, actions. But then if they are morally best, why aren't they morally required, contrary to the assumption that they are morally optional? In short: how can an action that is morally best to perform fail to be what one is morally required to do? The source of this alleged paradox has been dubbed the ‘good-ought tie-up’. In our article, we address this alleged paradox by first making a phenomenological case for the reality of instances of genuine supererogatory actions, and then, by reflecting on the relevant phenomenology, explaining why there is no genuine paradox. Our explanation appeals to the idea that moral reasons can play what we call a merit conferring role. The basic idea is that moral reasons that favor supererogatory actions function to confer merit on the actions they favor—they play a merit conferring role—and can do without also requiring the actions in question. Hence, supererogatory actions can be both good and morally meritorious to perform yet still be morally optional. Recognition of a merit conferring role unties the good-ought tie up, and (as we further argue) there are good reasons, independent of helping to resolve the alleged paradox, for recognizing this sort of role that moral reasons may play.

Hide All

1 Urmson J. O., “Saints and Heroes,” in Melden A. I., ed., Essays in Moral Philosophy (Seattle and London: University of Washington Press, 1958), 198216.

2 See the works by Gert cited in note 37 below. Practical reasons concern all sorts of considerations that (as reasons) bear on choice and actions, and thus include nonmoral as well as moral reasons. Gert introduces the requiring/justifying distinction with respect to roles bearing on the rationality of action, and hence with regard to practical reasons generally. Moral reasons too, as a species of practical reasons, may play either a requiring or a justifying role. In Sections VI and VII, we explain and illustrate the idea of roles that practical reasons in general, and moral reasons in particular, may play.

3 Deontic concepts are used to morally evaluate actions and practices, and such concepts are expressed in English by such terms as ‘duty’, ‘obligation’, ‘right’, ‘wrong’, and ‘optional’. Evaluative concepts used with moral significance are expressed by such English terms as ‘good’ and ‘bad’ and are applied not only to actions but to persons and states of affairs. Reactive attitudes (of moral significance) have to do with such responses as praise and blame, guilt and indignation—attitudes that are appropriate responses to morally significant actions, practices, and persons in light of whatever moral significance they possess.

4 To say that an action is prima facie morally required is to say that there are reasons for performing the action which, if not outweighed by reasons favoring an alternative action, suffice to make the action in question all-things-considered (all-in) morally required.

5 Arguably, there can be cases of supererogation in which one must violate a prima facie duty (either to others or to oneself) in order to perform the action. If I have agreed to meet you for an appointment, I have a prima facie duty to keep it. But if, on the way to the appointment, I stop to help someone whose car has run out of gas, my action can still qualify as supererogatory even though (in the circumstances) my fulfilling my prima facie duty to you implies that I have a prima facie duty not to stop. In what follows, we set such cases aside and focus on “pure” cases of supererogation, in which the supererogatory action is completely optional in the sense that it is neither prima facie required nor prima facie wrong.

6 We are not proposing these elements as a hard and fast definition. For instance, some authors deny that genuine altruistic motivation is a necessary component of the supererogatory. We focus on cases of supererogation that fit our description because they seem to be the sorts of cases that Urmson and others writing on the topic have tended to focus upon. For an excellent discussion of the contentious nature of this concept, see Heyd David, Supererogation (Cambridge: Cambridge University Press, 1982); and Heyd, “Supererogation,” The Stanford Encyclopedia of Philosophy, 2006. See also Mellema Gregory, Beyond the Call of Duty: Supererogation, Obligation, and Offense (Albany: State University of New York Press, 1991), chap. 2.

7 Flescher Andrew M., Heroes, Saints, and Ordinary Morality (Washington, DC: Georgetown University Press, 2003), 172–91, characterizes heroes as those whose heroic actions are triggered by what he calls “reactive altruism,” while saints, who make helping others a vocation, engage in what he calls “proactive altruism.”

8 We take our list from Heyd, Supererogation, chap. 7. Cf. Schumaker Millard, Supererogation: An Analysis and Bibliography (Edmonton: St. Stephen's College, 1977), chap. 2.

9 Heyd introduced the unqualified/qualified terminology in his Supererogation; and Dancy Jonathan distinguishes strong from weak supererogation in “Supererogation and Moral Realism,” in Dancy J., Moravcsik J., and Taylor C. C. W., eds., Human Agency—Language, Duty, and Value: Philosophical Essays in Honor of J. O. Urmson (Stanford, CA: Stanford University Press, 1988); and Dancy, Moral Reasons (Oxford: Blackwell, 1993), 130–31. These pairs of terms (as they are used by those who have introduced them) are extensionally equivalent. According to unqualified/strong conceptions, (a) supererogatory actions have an intrinsic value, but (b) being “completely” or “purely” optional (in the sense that they are not even prima facie required, and hence failing to perform them is not prima facie wrong), they are not moral duties or obligations of any kind, nor are they requirements of rationality. Qualified/weak conceptions accept (a), reject (b), and then go on to claim that so-called supererogatory actions are either moral requirements which agents are excused from having to fulfill, or perhaps actions which, while being prima facie morally required, are not all-in morally required.

10 Mellema, Beyond the Call of Duty, chap. 5, uses the term ‘quasi-supererogation’ more narrowly than we do, to refer to nonobligatory actions whose performance is praiseworthy, but whose nonperformance is blameworthy. Our use of the term subsumes Mellema's.

11 For instance, the Oxford English Dictionary, compact edition (Oxford: Oxford University Press, 1991), 3159, entry for ‘supererogation’ includes the following two senses: 1. “The performance of good works beyond that which God commands or requires, which are held to constitute a store of merit which the church may dispense to others to make up for their deficiencies,” and 2. “Performance of more than duty or circumstances require; doing more than is needed.” The first sense reflects the Roman Catholic doctrine of Indulgences (instituted during the Crusades, ca. 1080–1300), according to which (roughly) sinners could withdraw merit for a fee from what was called the Spiritual Treasury of the Church (that had been built up by the good works of Jesus and the Saints) and then could apply that merit toward their own salvation. Unlike the first sense, the second sense makes no mention of supererogatory actions being meritorious. Other dictionaries—including The Cambridge Dictionary of Philosophy, 2d ed., ed. Audi Robert (Cambridge: Cambridge University Press, 1999), 890; the Merriam-Webster Online Dictionary,; and the American Heritage College Dictionary, 4th ed. (Boston and New York: Houghton Mifflin Co., 2002), 1384—feature definitions that are very similar to the second of the two senses just quoted. Heyd, Supererogation, chap. 6, defines ‘supererogation’ partly in terms of merit, thus preserving what he argues is an important element in the traditional Christian understanding of the term. Again, along with many other philosophers who write about supererogation, we are particularly interested in those instances of supererogatory actions that are praiseworthy because of how the agent was motivated. Hence, we focus on meritorious supererogation.

12 See, for example, Heyd, Supererogation, chap. 6. In any case, with respect to nonmeritorious supererogation, the paradox arises because such actions are still good in virtue of what they bring about (or are intended to bring about), yet they are morally optional. So the solution to the paradox we propose below in Section VI will apply to all cases of genuine supererogation—both meritorious and nonmeritorious.

13 We thank Holly Smith, Tom Hurka, and Doug Portmore for prompting this particular clarification.

14 Although there are some differences in the ordinary uses of the concepts of moral duty and moral obligation—the former having more to do with specific jobs, roles, and stations; the latter having more to do with agreements and benefactions—we will, following what is fairly standard usage in contemporary moral philosophy, use these terms interchangeably. However, for reasons noted in the next paragraph in the text, we depart from what we take to be common philosophical usage in not using these terms interchangeably with ‘ought’ (even in cases where ‘ought’ is used for moral evaluation). For a discussion of these concepts and their interrelations, see Brandt R. B., “The Concepts of Obligation and Duty,” Mind 73 (1964): 374–93.

15 On this point, see Heyd, Supererogation, 171, who distinguishes the broader “commendatory” sense of ‘ought’, which may be properly applied to supererogatory actions, from what he calls the “prescriptive, personal” sense of the term, which may not be so used.

16 There are other ways in which the perfect/imperfect obligation distinction is drawn, including one where the distinction hinges on whether, corresponding to an obligation to do or refrain from some action, others have corresponding rights that one perform or refrain from the action in question. See Campbell T. D., “Perfect and Imperfect Obligations,” The Modern Schoolman 52 (1975): 285–94, who finds five distinct contrasts that this pair of terms has been used to indicate. See also Schumaker Millard, Sharing without Reckoning (Waterloo, Canada: Wilfrid Laurier University Press, 1992), chap. 1, for further discussion.

17 Here, and throughout, we are referring to meritorious supererogation.

18 If there are morally best reasons to perform some action A, why does it not follow automatically that one is all-in required, and not just prima facie morally required, to perform that action? It does not follow automatically because, as argued by Portmore Douglas W., “Are Moral Reasons Morally Overriding?Ethical Theory and Moral Practice 11 (2008): 369–88, the inference in question assumes that (i) nonmoral reasons are not relevant in determining the overall deontic status of an action, and (ii) moral reasons always override competing nonmoral reasons. Portmore argues that these assumptions, while often taken for granted by moral philosophers, are false. So the premise in question is stated in a manner that allows for the possibility that (i) and (ii) are false.

19 A particularly clear presentation of the paradox in terms of reasons for action is to be found in Raz Joseph, “Permissions and Supererogation,” American Philosophical Quarterly 12 (1975): 164.

20 Railton Peter, “Normative Guidance,” in Shafer-Landau R., ed., Oxford Studies in Metaethics, 1 (Oxford: Oxford University Press, 2006), 3.

21 Another articulation of the general methodological point being embraced here is what Huemer Michael, Ethical Intuitionism (New York: Palgrave Macmillan, 2005) calls the Principle of Phenomenal Conservatism, according to which “it is reasonable to assume that things are the way they appear” (99).

In committing ourselves to the claim that, in some cases, people nonerroneously experience what they do as being beyond the call of duty, we take no stand on metaphysical issues concerning whether there is some property, being supererogatory, of the sort that a moral realist would countenance. We ourselves favor a version of metaethical expressivism, which denies the metaphysical claims of the moral realist, but allows for the idea that moral judgments (including those about the supererogatory) are subject to being true or false.

22 The expression ‘experiences as of supererogation’ is being used here (and elsewhere) instead of the expression ‘experiences of supererogation’ in order to remain neutral in our characterization of such experiences with respect to whether, on a particular occasion, there is a genuine act of supererogation that is the object of one's experience. Thus, to describe an experience being an experience as of supererogation allows that one's experience may not have as its object a genuine supererogatory action. That there are genuine instances of supererogatory actions is a claim we go on to defend against the anti-supererogationists. Similarly, the expression ‘experience as of obligation’ is used in contexts where it is important to remain neutral with respect to the question of whether the action being experienced is a genuine obligatory action.

23 This quotation is to be found in Monroe Kristin R., The Heart of Altruism (Princeton, NJ: Princeton University Press, 1996), 104. This sort of reaction among righteous gentiles is robust. For instance, Hallie Philip, Lest Innocent Blood Be Shed (New York: Harper, 1979) relates part of an interview with Magda Trocmé, one of the righteous gentiles in the southern French village of Le Chambon who protected Jews from Nazis:

Madame Trocmé was not the only citizen of Le Chambon who scoffed at words that express moral praise. In almost every interview I had with a Chambonnais or a Chambonnaise there came a moment when he or she pulled back from me but looked firmly into my eyes and said: “How can you call us ‘good’? We were doing what had to be done. Who else could help them?” (20).

24 Urmson, “Saints and Heroes,” 103–4, is among the pro-supererogationists who take the first tack, while Dancy Jonathan, Moral Reasons (Oxford and Cambridge, MA: Blackwell, 1993), 141–42, is among those who take the second tack.

25 See, for instance, Hale S. C., “Against Supererogation,” American Philosophical Quarterly 28 (1991): 273–85.

26 For a helpful introductory discussion of this third usage and its relation to the discipline/subject-matter usages, see Kriegel Uriah, “Moral Phenomenology: Foundational Issues,” Phenomenology and the Cognitive Sciences 7 (2008): 119. For a thorough treatment of the phenomenological tradition, see Spiegelberg Herbert, The Phenomenological Movement: A Historical Introduction, 2d ed., 1 and 2 (The Hague: Martinus Nijhoff, 1969). See especially chap. 14 in vol. 2, “The Essentials of the Phenomenological Movement.”

27 See, for instance, Horgan Terry and Timmons Mark, “What Does Moral Phenomenology Tell Us about Moral Objectivity?” in Paul Ellen Frankel, Miller Fred D. Jr., and Paul Jeffrey, eds., Objectivism, Subjectivism, and Relativism in Ethics (Cambridge: Cambridge University Press, 2008), 267300; Horgan and Timmons, “Prolegomena to a Future Phenomenology of Morals,” Phenomenology and the Cognitive Sciences 7 (2008): 115–31; Horgan and Timmons, “Morphological Rationalism: Making Room for Moral Principles,” Ethical Theory and Moral Practice 10 (2007): 279–95; Horgan and Timmons, “Moorean Moral Phenomenology,” in Nuccetelli S. and Seay G., eds., Moorean Themes in Epistemology and Ethics (Oxford: Oxford University Press, 2007), 203–26; Horgan and Timmons, “Moral Phenomenology and Moral Theory,” Philosophical Issues 15 (2005): 5677; and Horgan and Timmons, “Mandelbaum on Moral Phenomenology and Moral Realism,” in Verstegen Ian, ed., Maurice Mandelbaum and American Critical Realism (London: Routledge, 2010). See also Horgan and Timmons, “The Phenomenology of Virtue,” in Moral Twin Earth and Beyond (Oxford and New York: Oxford University Press, forthcoming).

28 Mandelbaum Maurice, The Phenomenology of Moral Experience (Glencoe, IL: The Free Press, 1955; reprinted, Baltimore, MD: The Johns Hopkins University Press, 1969), 127.

29 For further differences between direct and removed moral experiences, see Horgan and Timmons, “Mandelbaum on Moral Phenomenology and Moral Realism.”

30 The next two paragraphs are taken from our essay “Moral Phenomenology and Moral Theory.”

31 Mandelbaum, The Phenomenology of Moral Experience, 54.

32 Although experiences of both perfect and imperfect obligation involve a felt demand, in the case of imperfect obligations the felt demand attaches to rather unspecific act-types, in contrast with cases of perfect obligation, in which the felt demand attaches to a fairly specific act-type.

33 Two comments are in order here. First, it is certainly very common to have one's everyday experiences “colored” by an overall mood (e.g., depression) or by some more particular emotion (e.g., fear of missing a deadline). Such occurrent psychological features of one's experience at a time need not be the focus of one's attention, at least in those persons who seem to function quite normally under such conditions, though such features are part of one's overall experience. Second, Mandelbaum treats such emotions as disgust, indignation, admiration, and the like as by-products of removed moral judgments—moral judgments made from a spectator's point of view. If one sticks to Mandelbaum's direct/removed distinction, then what we are proposing is that first-person experiences of being morally obligated typically involve, as a part of the overall experience, viewing one's moral choice from a detached, spectator perspective. This seems plausible in light of the prevalence of the role in direct moral experiences of thoughts associated with such questions as: “What sort of person would I be if I did/didn't do such and such?” and “How would I feel if someone did/didn't do such and such for me?” Viewed this way, it makes sense to characterize different phases of the moral experience of the sort featured in our scenarios as including (a) an initial phase in which one experiences a felt demand grounded in one's apprehension of fittingness or unfittingness, (b) a reflective phase in which one mulls over one's choice (a phase in which one may imaginatively adopt a spectator's perspective), and (c) a final phase in which one decides what to do. Of course, such “phases” need not be temporally separate, but might instead be experientially superimposed.

34 Sinnott-Armstrong Walter, “You Ought to Be Ashamed of Yourself (When You Violate an Imperfect Obligation),” Philosophical Issues 15 (2005): 193208, argues that violations of perfect duty call for guilt on the part of the agent, while violations of imperfect duty call for shame on the part of the agent.

35 Hale, “Against Supererogation,” refers to the alleged phenomenon that failure to perform supererogatory actions calls for an excuse, and appeals to this phenomenon as evidence that supposed cases of supererogation are really cases of imperfect duty. But Olivia does not take her nonperformance to call for an excuse. If asked by her husband why she didn't go ahead and call Mary (in a scenario where she does not perform this particular act of supererogation), Olivia may say that she just didn't feel like it. Her not feeling like it explains why she didn't call, but it is not put forth as an excuse. Nor, from a spectator's point of view or her own, does she need an excuse. Like many other writers on the topic, Hale seems to focus on contentious cases of saintly and heroic actions, and she seems to assume that if an action is good then it is prima facie required.

36 For those with egoistic leanings who are skeptical of putative cases of pure altruism, we recommend the experimental work of social psychologist C. D. Batson as an antidote. See, for example, Batson, “How Social an Animal?American Psychologist 45 (1990): 336–46; and Batson, The Altruism Question: Toward a Social Psychological Answer (Hillsdale, NJ: Lawrence Erlbaum, 1991).

37 See Gert Joshua, Brute Rationality (Cambridge: Cambridge University Press, 2004); and Gert, “Normative Strength and the Balance of Reasons,” Philosophical Review 116 (2007): 533–62. Patricia Greenspan also defends the idea that practical reasons can function in ways that do not require. See, for example, Greenspan, “Asymmetrical Reasons,” in Reicher M. E. and Marek J. C., eds., Experience and Analysis: Proceedings of the Twenty-Seventh International Wittgenstein Symposium (Vienna: OEBV and HPT, 2005), 387–94; Greenspan, “Practical Reasons and Moral ‘Ought’,” in Shafer-Landau R., ed., Oxford Studies in Metaethics, 2 (Oxford: Oxford University Press, 2007), 172–94; and Greenspan, “Making Room for Options: Moral Reasons, Imperfect Duties, and Choice,” elsewhere in this volume.

38 See Gert, Brute Rationality, chap. 2.

39 In the following subsection, we say more about how the requiring/justifying distinction relates to supererogation.

40 In addition to the works mentioned in note 37, see also Gert Joshua, “Reply to Tenenbaum,” Canadian Journal of Philosophy 37 (2007): 463–76.

41 Here it is important to recall that our focus in this essay is on cases in which the supererogatory action in question is neither prima facie required nor prima facie forbidden. As explained above in Section II, cases in which a “supererogatory” action is prima facie morally required, but with respect to which there are competing reasons that make the action all-in morally optional, are what we call cases of “quasi-supererogation.”

42 That is, without playing the sort of role that Gert characterizes as the justifying role.

43 An example of this kind of rationale would be to argue (as Urmson does in “Saints and Heroes”) that a morality that does not put limits on what duty requires, and thus does not make room for supererogation, would not be livable given typical human psychology.

44 Heyd, Supererogation, 165.

45 Ibid., 175. In defending supererogation against the paradox, Heyd claims that “the model of reasons is inadequate to the explanation of supererogation” (ibid., 170). However, his claim is apparently based on the assumption that reasons for action can only play a requiring role in how they favor an action. Nevertheless, it seems to us that the sorts of considerations that Heyd brings forth in defense of supererogation could be cast in terms of the language of reasons if, as we are suggesting, one recognizes a moral-merit-conferring role that moral reasons can play.

46 We distinguish between actions guided by a sense of duty and those guarded by a sense of duty. Actions guided by duty include those performed because the agent believes them to be required (perfect duties), and those performed because they are believed to fulfill an imperfect duty. By contrast, actions are guarded by duty when the agent is careful about not doing something that violates duty. Actions guarded by duty in this way may nevertheless be done for reasons that serve to confer merit.

47 Portmore Douglas, “Are Moral Reasons Morally Overriding?Ethical Theory and Moral Practice 11 (2008): 369–88; and Portmore, Commonsense Consequentialism: Wherein Morality Meets Rationality (work in progress).

48 Dreier James, “Why Ethical Satisficing Makes Sense and Rational Satisficing Doesn't,” in Byron Michael, ed., Satisficing and Maximizing (Cambridge: Cambridge University Press, 2004), 131–54.

49 Portmore, “Are Moral Reasons Morally Overriding?” 381.

50 Portmore does not deny this claim.

* An earlier version of this paper was presented at the following venues: “The Varieties of Moral Experience: A Phenomenological Investigation,” Durham University, August 27–28, 2008; the Brackenridge Philosophy Symposium, “The Ethical and Epistemic Dimensions of Robert Audi's Intuitionism,” University of Texas, San Antonio, February 7–8, 2009; and the Department of Philosophy Colloquium Series, University of Nevada, Las Vegas. We wish to thank audiences at these conferences for very useful discussion of this paper. We also wish to thank Robert Audi, Matt Bedke, Paul Bloomfield, Michael Bukoski, Ginger Clausen, Josh Gert, Michael Gill, David Heyd, Uriah Kriegel, Victor Kumar, Ellen Frankel Paul, Stefan Sciaraffa, and especially Doug Portmore for very helpful comments on earlier versions of this paper.

Recommend this journal

Email your librarian or administrator to recommend adding this journal to your organisation's collection.

Social Philosophy and Policy
  • ISSN: 0265-0525
  • EISSN: 1471-6437
  • URL: /core/journals/social-philosophy-and-policy
Please enter your name
Please enter a valid email address
Who would you like to send this to? *


Full text views

Total number of HTML views: 1
Total number of PDF views: 21 *
Loading metrics...

Abstract views

Total abstract views: 242 *
Loading metrics...

* Views captured on Cambridge Core between September 2016 - 21st October 2017. This data will be updated every 24 hours.