Virtues and Reasons: Shafer-Landau's Euthyphro Dilemma for Virtue Ethics
Co-authored with Davis Smith
To be presented at the 74th Annual Northwest Philosophy Conference, Portland, Oregon

Abstract:          Russ Shafter-Landau in his introductory textbook, The Fundamentals of Ethics, poses a Euthyphro-like dilemma for virtue ethics. We grasp the second horn of the dilemma firmly, arguing that Shafer-Landau overestimates the cost of this alternative to virtue ethics because he fails to recognize the constitutive relation reason has to virtue: the dilemma only has force against exemplarist versions of virtue ethics that purport to offer an unequivocal theory of right action. We then briefly offer two accounts of the constitutive relation between virtues and reasons. 

1. The Dilemma

A standard taxonomy of ethics found in many if not most introductory textbooks divides the field into three divisions: Metaethics, Normative Ethics, and Applied Ethics. Given this taxonomy, it is unsurprising that virtue ethics is presented as a normative theory; indeed, the Stanford Encyclopedia of Philosophy’s entry on Virtue Ethics begins with, “Virtue ethics is currently one of three major approaches in normative ethics.” (The other two being consequentialism and deontology.) [1] Unlike Consequentialism and Deontology,[2] Virtue Ethics does not start with the question “what should I do and why?” Rather, it starts with questions such as “who should I be?”, “what kind of character should I seek to have?”, and “what does it take to be a good person?”.  These questions are fundamentally different from the questions found in Normative Ethics, so much so that Virtue Ethics might be deemed a whole separate field of study.

Nevertheless, some in this field have deemed it worthwhile to have a theory bridge this gap and attempt to provide an answer to the question “what should I do?” through answering the question “who should I be?”. Agent based and exemplarist virtue ethicists, most notably Michael Slote and Linda Zagzebski, have proposed accounts in which all other concepts germane to virtue ethics, including eudaimonia are to be analyzed in terms of the emotions and dispositions of virtuous agents. According to Zagzebski, “A wrong act = an act that the phronimos characteristically would not do, and he would feel guilty if he did = an act such that it is not the case that he might do it = an act that expresses a vice = an act that is against a requirement of virtue (the virtuous self)”[3]

It is not surprising then, that introductory texts tend to treat virtue ethics as a normative theory, concerning norms of behavior, as offering a theory of what makes actions right, rather than as a completely different branch of study.  For instance, Mark Timmons, citing Hursthouse, offers the following virtue-ethical principle: 

VE       An action is right if and only if (and because) it is what a virtuous person (acting in character) would not avoid doing in the circumstances under consideration.[4]

As it may beg a question about the nature of virtue ethical theory, let us rename this principle SRA (Standard of Right Action).   Russ Shafer-Landau’s characterization of this principle is almost identical:

SRA    An act is morally right just because it is one that a virtuous person, acting in character, would do in that situation.[5]

There is an issue with this in that there may be multiple moral exemplars whom one might seek to emulate, and these exemplars may disagree about what to do in a particular circumstance. To accommodate this, Shafer-Landau offers a tripartite version of this standard which qualifies the action according to the responses of this congress of virtuous people (hereafter referred to as the Aristoi[6]). Let us call this standard SRA2:

1. An act in a given situation is morally required just because all the Aristoi, acting in character, would perform it.

2. An act in a given situation is morally permitted just because some but not all of the Aristoi, acting in character, would perform it.

3. An act in a given situation is morally forbidden just because none of the Aristoi would perform it.[7]

Take notice of the “just because,” which appears in all three of these. This excludes background reasons for why the Aristoi might act in such a way, the act is right solely because they would act in that way.   Virtue ethics is presented here purely as a theory of moral exemplars, and as such, Shafer-Landau points out its vulnerability to a dilemma, formally similar to the Euthyphro Dilemma:[8]

1. Either virtuous people have good reasons for their actions, or they don’t.

2. If they lack good reasons for their actions, their actions are arbitrary, and thus cannot serve as a standard of morality.

3. If they have good reasons for their actions, then these reasons, rather than the choices of good people, determine what is right and wrong.[9]

So then either the actions of virtuous people are arbitrary (and thus cannot serve as a foundation for morality) or the reasons used by virtuous people are the standards for right action, not their actions. Good people, that is, choose their actions for good reasons, the actions are not good just because they choose them. Shafer-Landau rightly says that grasping this horn of the dilemma is the better option, but argues that it comes at a steep cost:

The cost is that the virtue ethicist’s account of right action is directly threatened. That account tells us that acts are morally right just because all virtuous people would perform them in the circumstances, and wrong just because such people would refrain. But as we have seen, the choices of virtuous people do not make actions right or wrong.[10]

2. Virtues and Reasons

We think this argument commits a fallacy, or at best applies to some (but not all) exemplarist theories of virtue. Most classical and contemporary versions of virtue ethics simply do not draw such an exclusive dichotomy between virtues and reasons, even when prioritizing the former in explaining the latter. For example, Hursthouse and Pettigrove write, “To possess a virtue is to be a certain sort of person with a certain complex mindset. A significant aspect of this mindset is the wholehearted acceptance of a distinctive range of considerations as reasons for action.”[11] 

Shafer-Landau’s dilemma thus rests on a mistaken account of what virtue ethics is and overly thin notion of what constitutes a virtue. Insofar as one can aspire to them, virtues are acquired character traits held to be desirable for certain reasons and inculcated for those reasons. Thus internalized, virtues shape frameworks for reasoning about and evaluating situations, which leads to consistently acting in certain ways. Fully virtuous agents may occasionally act reflexively, no longer needing to reference their reasons in every situation, but even then, they would still be acting as they did for those reasons.

Aristotle tells us a virtue is a hexis, which translates (not badly) into Latin as habitus, but frequently (and less fortunately) from there to “habit” in English.[12] While a habit connotes an acquired reflex that one might passively perform without thought or reason, a hexis is instead an actively exercised power like Aristotle’s account of perception in De Anima or Plato’s account of recollection in the Meno.[13] Like habits, virtues are reinforced through repetition, but unlike habits, virtues are essentially both acquired through reason and act upon the exercise of reasoning. Virtues shape not only one’s actions, but how one thinks, one’s practical reasons for action.

The virtues give one a more accurate scale to compare the different actions they might take. The virtuously generous person gives prudently, concerned with the effectiveness of their altruism. By contrast, the merely habitually generous person, lacking the intentionality virtue brings to action, gives impulsively, and tends to slide into inefficient prodigality.[14] The virtuously witty person is far less likely than the habitually witty person to fall into buffoonery or make a poorly timed joke which derails the conversation. Similarly, the habitually solemn person is far more likely to suffer through a conversation when a well placed joke by them would break the tension and allow creativity to flow more freely. They joke when appropriate, not impulsively, or if their mature impulses occasionally require no contemplation, those impulses have still been shaped by reason through long practice, they joke as reason would approve.  

We shall now consider two virtue-ethical approaches that stress the inherent interdependence of virtue and reason.

2.1 Eudaimonism as a rationalism

The role of reason (and reasons) is central to classical virtue theory.[15]  Reason guides one’s identification of the character traits to be inculcated and emulated. There is always a reason to be virtuous according to Aristotle, the ultimate reason being eudaimonia, the good for the sake of which everything else is undertaken, which itself Aristotle describes as “activity in accordance with arete,” ‘arete’ being translated as either virtue or excellence. Often misleadingly translated as “Happiness,” eudaimonia signifies the full flourishing of one’s human potential, which, according to Aristotle, cannot be achieved apart from the essential human ergon, which is reasoning. Thus, virtuous people not only act as they do for good reasons, they act as they do for the best of reasons.

Eudaimonia is not subjective but objective, not an emotion but a state of being.[16] The “Good Life,” one’s objectively best life, is achieved by fulfilling the essential function of a human being, to be an animal governed by reason. This is very different from leading a good life in the hedonistic sense. To see how these come apart, we can look to Robert Nozick’s thought-experiment called The Experience Machine.[17] People voluntarily agree to be ‘plugged into’ a machine, in which they enter into a permanent simulation of a perfect life. All their dreams seem to come true. The agreement stipulates they will not recall entering the machine and will not ever have to leave the machine. Persons living in such a simulation might genuinely believe they’d achieved the pinnacle of all their aspirations, and experience intense subjective happiness. But this wouldn’t be Happiness in the eudemonistic sense because they hadn’t really achieved anything. Far from approaching their full potential, they are little more than deluded corpses curled in a fetal position inside the machine. Nothing experienced there could count as the full flourishing of one’s potential, at least assuming one’s highest potential is something better than that of a deluded corpse.

The assumption that the Good Man is the man governed by reason is the metaphysical ground of Aristotelian ethics.[18] In the Politics, discussing whether there is a difference between a Good Man and a Good Citizen, Aristotle says a man qua man is always governed by practical reason, whereas man qua citizen sometimes has to put his own reason in abeyance to obey the state.[19] Similarly, temperance is not only a mean between too much appetite and too little, but “to desire only what reason approves.”[20]  Temperance transcends self-control because by the time one has achieved the virtue of temperance, there is no conflicting impulse to struggle against. One’s first impulse is (has become) the temperate one, which is to say, a rational one. The moral virtues, in general, are essential to the maintenance of the well-ordered polis, which in turn constitutes the (teleologically) natural human habit because it enables the existence of specialists dedicated to the exercise of the human ergon.

2.2 Virtue Consequentialism

Alternatively, Julia Driver in Uneasy Virtue takes the notions from Consequentialism and uses them to construct a theory of character. [21] Virtue Consequentialism lacks essentialist teleological assumptions,[22] instead defining a virtue as a character trait acting upon which would systematically produce the best results, where a number of differing metrics might be employed in calculating the best outcomes. Bradley, building off of Driver, starts a series of different ways to define a virtue using consequentialism with “V is a virtue (in the actual world) iff V is a character trait that systematically produces more good than not in the actual world.”[23] Bradley concludes his series of ever improving definitions with “It is a virtue for people S1–Sn to have character trait V1 rather than character trait V2 at world w iff (i) V2 is a member of the contrast class of V1, and (ii) the expected intrinsic value of a closest world to w where S1–Sn exercise V1 is greater than the expected intrinsic value of a closest world to w where S1–Sn exercise V2.”[24] In less mathematically precise phrasing, we would say that a trait is a virtue IFF, when compared to the other traits which one could have in its place, were a person to act on that trait, it would systematically produce more good and less evil than the alternatives. Here we use counterfactuals to define virtues so as to avoid cases of moral luck and the use of contrastivism to compare traits and have the virtue be the optimal one. It is an open question whether one ought to measure the outcome of an individual acting on the trait, a group of people, or all people.[25]

While it is possible to use this definition of virtue to make a claim that a virtuous person is one such that all of their traits are virtues and then have them stand as the people for SRA or SRA2, this is far from tenable. There are clear and obvious reasons why a virtuous person would act in a certain way in a certain situation, namely, they are acting to promote the good. The virtuous traits they have illuminate the correct course of action by giving a disposition to weigh the outcomes in certain ways and notice aspects others may miss.

Take, for example, the character trait caring. This trait is very often expressed in action. When one acts from a place of care, they act to promote the well-being of the individual they care for.[26] As a result, the reason the virtuous person would act as they do is clear, doing otherwise would harm another’s well-being or fail to promote it. Another example is being honest. Telling the truth always and without exception can cause more harm than good, so a virtuous person might lie in some cases because doing so would promote the good while they might be truthful in another context even if it causes a small amount of suffering immediately to save a person a greater amount later down the road. In both of these cases, the virtuous person has reason for their (dis)honesty, the virtues they have give them an accurate scale to measure the outcomes.

For a third and final example, Consequentialism is often accused of being overly impartial, to the point that one is expected to ignore special obligations.[27] Simple forms of Consequentialism have it that I have no greater duty towards the well-being of a friend or family member than I do a stranger. It may seem, however, that I have a greater duty towards my friend’s well-being than I do the stranger’s. Simple Consequentialism can and does provide reasons for discounting these special obligations but it could be argued that this is a mis-weighing of the factors in the case. It could be argued that acting consistently impartial does not maximize the good but rather having a degree of bias, the kind illuminated by the special obligations, promotes well-being far better.[28] The character traits involved in this kind of bias (virtues) adjust how one measures the outcomes. In these kinds of cases, as before, the virtuous person has reasons and the traits illuminated them.    

3. Conclusion

In this paper, we have offered a reply to Shafer-Landau’s criticism that virtuous people have reasons for their actions and it is those reasons, not their behavior, which determines rightness and wrongness of their actions. We fully accept this because, fundamentally, Virtue Ethics is concerned with the question of character and not the question of action. While one could claim that the virtuous person will, when acting on their character traits, always (or almost always) perform the right actions, it was never, or never should have been, the claim of Virtue Ethics that character, instead of reasons, determines the rightness of action. The claim should be that having virtues gives one the correct means of reasoning about one’s actions, that one might more clearly see the correct path. The connection between Virtue Ethics and norms of behavior, if there is one at all, would be “make a good person and good actions will follow,” but that claim simply isn’t tantamount to claiming that actions are right because they are what a good person would do.  

[1] Hursthouse, Rosalind and Glen Pettigrove, "Virtue Ethics", The Stanford Encyclopedia of Philosophy (Winter 2022 Edition), Edward N. Zalta & Uri Nodelman (eds.), URL = <https://plato.stanford.edu/archives/win2022/entries/ethics-virtue/

[2] Also called ‘Non-Consequentialism’ to make the excluded middle explicit. That is so say, a theory for right action must be either Consequentialist or Non-Consequentialist, there is no third option.

[3] Hursthouse and Pettgrove, ibid, quoting  Zagzebski).

[4] Timmons, Mark, ed. Disputed Moral Issues: A Reader, 5th ed. (Oxford, 2020), 29.

[5] Shafer-Landau, Russ. The Fundamentals of Ethics, 5th ed.  Oxford, 2021; 272-294, but see especially 289-290.Ibid.

[6] Meaning “the Best,” it was the term by which ancient Athenian nobility referred to themselves.

[7] Ibid.288. This principle would still presuppose a single list of virtues.

[8] We could call this the “Virtue Euthyphro Dilemma.”

[9] Ibid., 290.

[10] Ibid., the italics are in the original. It could also be pointed out that this is different from what he says two pages back. SRA2 was, “An act in a given situation is morally required just because all virtuous people, acting in character, would perform it.” “Required” is more demanding than “right,” the latter being closer to “permitted.” Although, of course, If an act is morally required, it must be right.

[11] Op. cit.

[12]See Joe Sachs, “Aristotle: Ethics,” Internet Encyclopedia of Philosophy. https://iep.utm.edu/aristotle-ethics/

[13] Ibid.

[14] For example, if an agent has some change, they might put it in the tin with the Santa Claus ringing the bell. But, in an extraordinary case where the agent is aware of the particular charity’s anti-LGBTQ+ policies, the virtuously generous agent may well refrain from such donations and seek a better, more impartial charity.

[15] Eudaimonism has sometimes been criticized on the grounds that eudaimonia is vague, ill-defined concept. We follow Hursthouse in replying that eudaimonia is no more problematic than the core concepts of happiness and rationality employed in other major normative theories. See Hursthouse, Rosalind, “Virtue Theory and Abortion,” Philosophy and Public Affairs 20 (1991).

[16] Though, we will admit that this state of being is often paired with an emotion, such as satisfaction.

[17] The original version of this thought-experiment appeared in Anarchy, State, and Utopia in 1974. In that version, one is only temporarily plugged into the machine. The version where one never leaves the machine is found in The Examined Life (Simon & Schuster, 1989), 104ff.

[18] Of course, a virtuous person may not always act in character. It is possible for a good person to do bad things, just as it is possible for a person of vicious character to behave well (for example, for fear of legal sanctions). But while good behavior does not necessarily follow from goodness of character, virtues are normally expressed behaviorally, or we wouldn't be able to identify them as character-traits we ought to emulate.

[19] Politics 3:4. Only when the good citizen is also the ruler is it possible that the virtue of the good man and the good citizen are one.  See especially 1277a12-24.

[20] The phrase is actually Roger Scruton’s, but it nicely captures the rationalism behind Aristotle’s Doctrine of the Mean--it is reason that finds that mean between cowardice and foolhardiness. Courage amounts to fearing only what reason permits one to fear. It is worth noting, however, that neither Justice nor Wisdom fit with the Doctrine of the Mean. There is no emotion, impulse, or desire of which either Justice or Wisdom is a moderate amount. 

[21] Julia Driver, Uneasy Virtue (CUP, 2001).

[22] Such as that a good person is one who fulfills their function and those functions are exemplifications of virtues as defined as a mean between two vices.

[23] Ben Bradley, “Virtue consequentialism.” Utilitas 17 (3):282-298.)

[24] Ibid 295

[25]This division can be seen if we take it that certain character traits are better had by people in certain positions in society than had by people in other positions. For example, obedience is a great trait to have if one is, for example, a child or a soldier, but not a great trait if one is a leader or a parent. The pre-colonial Nahuas claimed that philosophers should have a disposition to expose the hypocrisy in others. This is a great trait for philosophers but not a great trait for a jeweler.

[26] Or, at least, act in such a way as to not harm the well-being of the individual cared about.

[27] Special obligations are duties one has to a small group of individuals rather than those one naturally has towards all people. It is natural to think that I have a greater duty towards the well-being of my child than I do towards the well-being of a random child with whom I have no relationship. Jeske, Diane, "Special Obligations", The Stanford Encyclopedia of Philosophy (Winter 2021 Edition), Edward N. Zalta (ed.), URL = <https://plato.stanford.edu/archives/win2021/entries/special-obligations/>.

An example case is as follows: Your child and their friend are swimming in a lake. Your child is average while their friend is a super genius with great potential. Both begin to drown and you can only save one. Naïve Consequentialism would say that you ought to save the friend rather than your child, but this ignores the special obligation which you have to your own child.

[28] If all people acted to benefit those with whom they had a special obligation, the world as a whole would be better than the impartial one. See Driver, Julia, 2005. “Consequentialism and Feminist Ethics,” Hypatia, 20: 183–99