Aristotle’s Attainable and Attributable Phronimos

Abstract

Unlike many philosophers in the last quarter of the 20th century, many current scholars of Aristotle are less critical of the doctrine of the reciprocity of the virtues. They recognize that the reciprocity of the virtues is grounded in Aristotle’s accounts of eudaimonia, moral virtue, and phronēsis. However, there remains a concern that if the phronimos must have all the moral virtues in addition to phronēsis, then the phronimos is, for actual humans, unattainable. In what follows, I will argue that Aristotle’s phronimos is intended to be attainable and attributable to actual humans. In Part 2 of the paper, I will set out some preliminary points about the “methodology” Aristotle employs in his ethics, with special attention to the need to reconcile as many of the phenomena as one can and the recognition of the inescapable imprecision of ethics. In Part 3 of the paper, I will discuss passages where Aristotle insists upon the attainability of both eudaimonia and the moral virtues. In Part 4, I will explore how moral virtue and phronēsis admit of degrees and that his distinction between heroic virtue and moral virtue supports the attainability and attributability of the latter. Finally, in Part 5, I will propose a plausible reading of the reciprocity of the virtues that supports the thesis that the phronimos is attainable.

1. Introduction

Elizabeth Anscombe [1]1, Peter Geach [3], Philippa Foot [4], and Alasdair MacIntyre [5] helped renew interest in and appreciation for Aristotle’s ethical theory among 20th-century Anglo-American philosophers. To them, his Nicomachean Ethics offers insights and resources that seem to be lacking in consequentialism and deontology. Nevertheless, Geach, Foot, and MacIntyre all dismiss Aristotle’s doctrine of the reciprocity of the virtues2. This doctrine maintains that someone cannot have practical wisdom (phronēsis) without all the moral virtues (NE VI.12.1144a12–36), nor can someone have full moral virtue (kuriōs aretē) without phronēsis (VI.13.1144b1–17)3. Geach thinks the doctrine is “odious and preposterous” because you can have a virtue in one area (e.g., courage) but a vice in another area (e.g., intemperance) ([3], pp. 162–168). Foot and MacIntyre reject it because they think virtues can be used to achieve wicked ends ([4], pp. 15–18; [5], pp. 155, 179–80).

Later philosophers add to their critique. Richard Kraut and Elizabeth Telfer object that a person could have moral virtues in some areas along with mere continence in other areas ([8], pp. 83–84; [9], pp. 38–40)4. Neera Badhwar agrees that the reciprocity of the virtues doctrine is false, for “the virtues are disunited across different domains (areas of practical concern)”; however, she concedes they are “united within domains”. She calls this the Limited Unity of Virtue (LUV) ([10], p. 307)5. T.H. Irwin objects that the doctrine of the reciprocity of the virtues is inconsistent with other claims in the Nicomachean Ethics. In short, if Aristotle thinks (1) the large-scale virtues of magnificence and magnanimity are genuine and distinct virtues; (2) a person can be liberal without being magnificent or have the right disposition towards small and medium honors without being magnanimous; and (3) one cannot fully possess one moral virtue without possessing all of them, including phronēsis, then Aristotle’s ethical theory is inconsistent [13]6.

Subsequent philosophers are less critical of the doctrine of the reciprocity of the virtues. Inter alia, Julia Annas defends aspects of it in [18], Daniel Russell defends aspects of it in [6,19], and Christopher Healow defends aspects of it in [20]. They recognize that the doctrine of the reciprocity of the virtues is grounded in Aristotle’s accounts of happiness (eudaimonia), moral virtue, and phronēsis. Russell aptly remarks,

Some have found the reciprocity of the virtues so implausible that they have denied the very idea that the virtues might require phronesis. Yet doing so would throw out Aristotle’s entire way of thinking about the virtues, when in fact there seems to be a lot of merit in the idea that the virtues require phronesis.

([6], p. 204)

Phronēsis is crucial for good deliberation concerning what is conducive to the end (cf. VI.5.1140a24-b5). Russell explains that this “includes thinking about not only means to an end but also the very specification of that end in more concrete terms” ([6], p. 205). He identifies three parts to the decision-making process: (1) the indeterminate end from which deliberation begins (which is often desirable due to the moral virtues); (2) making the end determinate in the specific situation; and (3) working out the effective means to that determinate end (both of which, when carried out well “with regard to the things that are good or bad for man” (VI.5.1140b5), are the work of phronēsis) ([6], p. 205). The process of making the indeterminate end determinate, and working out the means to that determinate end, can involve salient features of more than one moral virtue, for instance, kindness and fairness (see [21], pp. 142–144). Hence, the full moral virtues require phronēsis and phronēsis requires all the other moral virtues.

Consequently, in this paper, I will assume that the doctrine of the reciprocity of the virtues is integral to Aristotle’s ethical theory and is not, as Kraut claims, “a hastily drawn conclusion” ([8], p. 83). However, there remains a concern that, as Healow puts it, “[T]he standards one must reach in order to be a truly virtuous person are so lofty as to be almost unattainable” ([20], p. 13). Annas puts the objection this way:

Every account of virtue has to make some appeal to an ideal (such an account would scarcely be convincing if it concluded that we are all fine just the way we are). But in some accounts the ideal can appear too far from the reality of ordinary people’s lives—too hopelessly ideal—and thus too demanding when we judge the characters of ourselves and others. How are we to deal with the point that however hard we and others try, none of us will ever actually be generous, brave, or fair, or have any virtue?

([18], p. 89)

One response to this objection, which Annas herself endorses, is to distinguish between the ideal of the phronimos, and the actual character of individual people. She writes,

The claim that the virtues are unified doesn’t, then, prevent us from recognizing virtue at the everyday level, and respecting people for it, but it does require us not to be complacent or easily satisfied about our own or other’s virtue, to expect our role models to have flaws and to respond to this maturely and without prematurely giving up on the idea of progress in virtue…The ideal of unification is what we aim at, not what we have actually achieved, and so we can happily admit that the world contains many people, including ourselves, who display a mixture of virtue, vice, and mediocrity.

([18], pp. 90–91)7

Annas’ distinction between the reciprocity of the virtues as an ideal and real-life virtuous people is similar to Russell’s distinction between the reciprocity of the virtues as a model thesis, which he defends, and the reciprocity of the virtues as an attributive thesis (i.e., attributed to individual people), which he rejects ([19], pp. 367–373). Indeed, he calls the attributive thesis “wildly implausible” ([19], p. 370). Later, he adds,

It is an interesting question whether Aristotle himself understood UV [the Unity (reciprocity) of the Virtues] as a model thesis… Charity would certainly recommend such a reading of Aristotle, unless we are to suppose that somehow he had simply overlooked the fact that there are no people who are virtuous in every way. Nonetheless, Aristotle’s text on this point is too scant to determine the issue with any certainty.

([19], p. 372, n. 30)

I think, properly understood, the attributive thesis is plausible, and that there is enough textual evidence to show that Aristotle defends the reciprocity of the virtues on both the model and the attributive level. In what follows, I will argue that phronimos—the person with full moral virtue and phronēsis—is not a standard so lofty as to be almost unattainable. Rather, Aristotle’s phronimos is intended to be attainable and attributable to real, flesh-and-blood humans8.

In Section 2 of the paper, I will set out some preliminary points about the “methodology” Aristotle employed in his ethics, with special attention to the need to reconcile as many of the phenomena as one can and the recognition of the inescapable imprecision of ethics. In Section 3 of the paper, I will discuss passages where Aristotle insisted upon the attainability of both eudaimonia and the moral virtues. In Section 4, I will explore how moral virtue and phronēsis admit of degrees, and I will note that the distinction between heroic virtue and moral virtue in NE VII.1 indicates the phronimos is attainable and attributable. Finally, in Section 5, I will propose a plausible reading of the reciprocity of the virtues that supports the thesis that the phronimos is attainable.

2. “Methodology” and Imprecision

Aristotle clearly thinks ethical theory or, more precisely, politikē, is a distinct intellectual discipline (cf. I.2. 1094a26-b11). Does he also think it has its own distinct methodology? At the outset of his discussion of weakness of will, Aristotle remarks,

As in the other cases we must set out the appearances (phainomena), and first of all go through the puzzles. In this way we must prove the common beliefs (endoxa)9 about these ways of being affected—ideally, all the common beliefs, but if not all, most of them, and the most important. For if the objections are solved, and the common beliefs are left, it will be an adequate proof.

(VII.1.1145b1–7; [25])

It is tempting to treat this as a general statement of Aristotle’s methodology in ethics. With some qualifications, this approach might apply to his discussion of eudaimonia (NE I.5, I.8–12), his discussion of phronēsis (NE VI.12–13), his discussion of pleasure (NE VII.12–14), and some of his remarks about friendship (NE VIII.1). But there are other topics in his Ethics, such as his “function argument” (NE I.7), his account of moral virtue (NE II.1–6), and his treatments of voluntary action, choice, deliberation, and wish (NE III.1–4), where he does not seem to employ this approach.

Nevertheless, Aristotle consistently maintains that a sound thesis in ethics must not conflict with the most common and authoritative beliefs. When these beliefs “have been held by many men and men of old” or “by a few eminent persons”, it is “not probable that either of these should be entirely mistaken, but rather that they should be right in at least some respect or even in most respects” (cf. I.8.1098b27–29). For instance, those who insist that the mere possession of virtue is sufficient for eudaimonia are “maintaining a thesis at all costs” (I.5.1096a2) and “those who say that the victim on the rack or the man who falls into great misfortunes is happy if he is good are, whether they mean it or not, talking nonsense” (VII.13.1153b18–20). Some people claim that pleasant objects have a compelling power, but, in contrast, noble actions are voluntary. Aristotle retorts, “It is absurd to make external circumstances responsible, and not oneself, as being easily caught by such attractions, and to make oneself responsible for noble acts but the pleasant objects responsible for base ones” (III.1.1110b12–15). Socrates is criticized for denying weakness of will exists because “this view plainly contradicts the observed facts” (VII.2.1145b28). Critics of pleasure might think they are doing most people, who are slaves of pleasure, a favor by leading them in the opposite direction. Aristotle disagrees, “For arguments about matters concerned with feelings and actions are less reliable than facts: and so when they clash with the facts of perception they are despised and discredit the truth as well” (X.1.1172a34–36). Consequently, if the reciprocity of the virtues entails something that strongly conflicts with the endoxa, such as that the phronimos is an unrealistic, unattainable ideal, despite the common practice of crediting some people with phronēsis (cf. VI.5.1140a25), Aristotle would consider that a serious and perhaps decisive objection.

Another important aspect of Aristotle’s ethical theory is its relative lack of precision, which he attributes to the subject matter of ethics. At the outset of NE I.3, he argues,

Our discussion will be adequate if it has as much clearness as the subject matter admits of, for precision is not to be sought for alike in all discussions, any more than in all the products of the crafts. Now fine and just actions, which political science investigates, admit of much variety and fluctuation of opinion, so that they may be thought to exist only by convention, and not by nature. And goods also give rise to a similar fluctuation because they bring harm to many people; for before now men have been undone by reason of their wealth, and others by reason of their courage. We must be content, then, in speaking of such subjects and with such premises to indicate the truth roughly and in outline, and in speaking about things which are only for the most part true and with premises of the same kind to reach conclusions that are no better. In the same spirit, therefore, should each type of statement be received; for it is the mark of an educated man to look for precision in each class of things just so far as the nature of the subject admits; it is evidently equally foolish to accept probable reasoning from a mathematician and to demand from a rhetorician scientific proofs.

(1094b11–27)

After his “function argument” concludes with his definition of eudaimonia as “activity of the soul in accordance with virtue, and if there are more than one virtue, in accordance with the best and most complete… in a complete life” (I.7.1098a16–18), he repeats this point (I.7.1098a25–26). So, if even his argument for eudaimonia is imprecise—it serves as “an outline of the good” that is first sketched roughly and requires later efforts to fill out the details (cf. I.7.1098a20–24)—then it is possible that other aspects of his ethical theory are also imprecise, including where the line between pre-virtue and virtue lies, how many moral virtues there are (cf. III.5.1115a3–5), and which ones are included in the reciprocity of the virtues doctrine.

3. Eudaimonia Is Attainable and Attributable

Despite the imprecision of Aristotle’s definition of eudaimonia, it clearly involves virtuous activities (cf. I.8.1098b30–1099a6; I.9.1099b25–26) being performed, as he later explains, by a durable character (cf. II.41105a32–33), which is formed by the frequent performance of virtuous activities (cf. II.1.1103a32-b2, 1103b13–25). Moreover, eudaimonia clearly is attainable. In NE I.9, Aristotle supports his account of eudaimonia by noting, “It will also on this view be very generally shared; for all who are not maimed as regards their potentiality for virtue may win it by a certain kind of study and care” (1099b19–20). In NE I.10, Aristotle confronts the puzzle that emerges from Solon’s proverb that “we must see the end”—hence, no one should be called happy while he is still alive (1100a10–11). He comments,

Now if we must see the end and only then call a man happy, not as being happy but as having been so before, surely this is a paradox, that when he is happy the attribute that belongs to him is not to be truly predicated of him because we do not wish to call living men happy, on account of the changes that may befall them, and because we have assumed happiness to be something permanent and by no means easily changed, while a single man may suffer many turns of fortune’s wheel.

(I.10.1100a32–b4)

Aristotle proceeds to turn the paradox into confirmation of his account of eudaimonia:

For no function of man has so much permanence as virtuous activities (these are thought to be more durable even than knowledge of the sciences), and of these themselves the most valuable are more durable because those who are happy spend their life most readily and most continuously in these; for this seems to be the reason why we do not forget them. The attribute in question [i.e., durability], then, will belong to the happy man, and he will be happy throughout his life; for always, or by preference to everything else, he will be engaged in virtuous action and contemplation, and he will bear the chances of life most nobly and altogether decorously, if he is ‘truly good’ and ‘foursquare beyond reproach’.

(I.10.1100b11–21)

At the end of that chapter, Aristotle summarizes his argument that, despite the uncertainty of the future, eudaimonia is attainable:

Why then should we not say that he is happy who is active in accordance with complete virtue and is sufficiently equipped with external goods, not for some chance period but throughout a complete life? Or must we add ‘and who is destined to live thus and die as befits his life’? Certainly, the future is obscure to us, while happiness, we claim, is an end and something in every way final. If so, we shall call happy those among living men in whom these conditions are, and are to be, fulfilled—but happy men. So much for these questions.

(I.10.1101a14–20)

Since eudaimonia is attainable, and eudaimonia is defined as virtuous activities performed over a complete life, then a complete life of virtuous activities is attainable. Of course, as Aristotle explains in NE II.1, virtuous activities performed regularly and repeatedly generate the moral virtues: “…we become just by doing just acts, temperate by doing temperate acts, brave by doing brave acts” (1103b1–2). Hence, the moral virtues, whereby one engages in virtuous actions over a complete lifetime, are attainable. Furthermore, when Aristotle defines “moral virtue” in II.6, he includes a role for the phronimos:

Virtue, then, is a state of character concerned with choice, lying in a mean, i.e., the mean relative to us, this being determined by a rational principle, and by that principle by which the man of practical wisdom (phronimos) would determine it.

(1106b36–1107a2; cf. VI.13.1144b22–26)

Well before he argued for the reciprocity of the virtues, Aristotle suggested that the moral virtues are tightly connected with phronēsis. Once he makes that connection clearer at the end of NE VI with the doctrine of the reciprocity of the virtues, it follows that the phronimos is attainable.

Perhaps at this point it is appropriate to ask, can someone with an erroneous view of eudaimonia count as a phronimos? If the erroneous view is that eudaimonia is pleasure, wealth, or honor, then, based on his criticisms of those views in NE I.5.1095b13–30 and 1096a6–10, I think Aristotle’s answer is no, such a person cannot be a phronimos. This negative answer is reinforced in Eudemian Ethics VII.15, where Aristotle’s criticizes “the man who thinks he ought to have the excellences for the sake of external goods does deeds that are noble only per accidens” (1249a15–16). What if, like Plato, someone thinks eudaimonia is virtue? Based on NE.I.8.1098b13–17 and 1098b25–1099a7, I think his answer is yes, such a person can be a phronimos. This is reinforced by Aristotle’s alleged elegy of Plato. According to Olympiodorus ([26], p. 268), Aristotle said the following:

To a man [Plato] whom it is not right for the wicked even to praise.

He alone or first of mortals clearly showed

From his own life and the manner of his arguments

That a man becomes good and happy at the same time.

These answers may prove important, for the objection that someone could have one moral virtue and phronēsis without some of the other moral virtues must, if it is to refute Aristotle, include in any counterexample an approximately correct view of eudaimonia.

4. Degrees of Virtue

Nevertheless, much more must be said to show that the phronimos is attainable and attributable to actual human beings (besides Plato). In this section, I will discuss two arguments in support of my thesis. The first, and more important reason, is that Aristotle acknowledges that there are degrees of moral virtue. Doug Reed identifies several passages where Aristotle suggests there are degrees of virtue: NE III.9.1117b16; IV.1.1120b6–10; IV.3.1124a2; VIII.6.1158a33–35; IX.12.1172a13; X.3.1173a19–22; cf. Categories 10b26 ([27], p. 92)10. Consider the passage from NE X. As part of his refutation of those who deny that pleasure is something good, Aristotle argues the following:

They say, however, that the good is determinate, while pleasure is indeterminate, because it admits of degrees. Now if it is from the feeling of pleasure that they judge thus, the same will be true of justice and the other virtues, in respect of which we plainly say that people of a certain character are so more or less, and act more or less in accordance with these virtues; for people may be more just or brave, and it is possible also to act justly or temperately more or less.

(1173a15–22)

Aristotle acknowledges that, in Christopher Toner’s words, “virtue is a degreed or continuous concept, so that it makes sense to say of three people that one is ‘quite generous’ (say), the second even more so, but the third less so” ([12], p. 213)11. Indeed, this seems built into Aristotle’s account of how the moral virtues are acquired: “For the things we have to learn before we can do them, we learn by doing them.” (1103a32). Just as one does not suddenly transition from the inability to play a musical instrument to proficiency at playing it and then remain at that level with no need for further improvement, there is not, in Toner’s helpful phrase, a “magic moment” in a person’s moral development in which he goes from non-virtuous to virtuous and then remains at that level of excellence ([12], p. 213).

Hence, the doctrine of the reciprocity of the virtues does not require all phronimoi to have the same degree of moral virtue; some might be further along than others. Nor must an individual phronimos have the same degree of courage as he has of temperance, etc. Given what Aristotle says about natural virtue—“For all men think that each type of character belongs to its possessors in some sense by nature; for from the very moment of birth we are just or fitted for self-control or brave or have the other moral qualities” (1144b3–5)—people who start out as naturally daring may be further along in courage, but further behind in another virtue, than those who are naturally cautious but further along in a different virtue, such as temperance. Furthermore, since it is full moral virtue (kuriōs aretē) that requires phronēsis, a person might acquire and possess a moral virtue that falls short of full moral virtue12, without possessing, so far, the other moral virtues and phronēsis.

The acknowledgement of degrees of virtue also fits with Aristotle’s insistence that ethics is imprecise. The vague boundaries in the acquisition and possession of virtue resemble the vague boundaries concerning the mean:

But this [finding and hitting the mean] is no doubt difficult, and especially in individual cases; for it is not easy to determine both how and with whom and on what provocation and how long one should be angry; for we too sometimes praise those who fall short and call them good-tempered, but sometimes we praise those who get angry and call them manly. The man, however, who deviates little from goodness is not blamed, whether he do so in the direction of the more or of the less, but only the man who deviates more widely; for he does not fail to be noticed. But up to what point and to what extent a man must deviate before he becomes blameworthy it is not easy to determine by reasoning, any more than anything else that is perceived by the senses; such things depend on particular facts, and the decision rests with perception. So much, then, is plain, that the intermediate state is in all things to be praised, but that we must incline sometimes towards the excess, sometimes towards the deficiency; for so shall we most easily hit the mean and what is right.

(II.9.1109b14–25; emphases mine)

Because it is not easy to determine where the mean lies, one of the ways in which people can make progress in virtue is by improving upon their ability to find and to hit the mean. Phronēsis involves a set of intellectual skills: excellence in deliberation (euboulia) (VI.1142b30–34), a good understanding (sunesis, eusunesis) about what should be done (VI.10.1143a1–15), good judgment (gnōmē) at determining what is equitable (VI.11.19–24), and nous concerning the particular facts of the situation (VI.11.1143a25–b6). Hence, it also seems that people can make progress in the acquisition and exercise of phronēsis13. One phronimos might be more perceptive and offer more helpful advice than another, because he has more and wider experience. “Therefore, we ought to attend to the undemonstrated sayings and opinions of experienced and older people or of people of practical wisdom not less than to demonstrations; for because experience has given them an eye they see aright” (VI.11.1143b11–14).

Nevertheless, someone might object that if moral virtue and phronēsis admit of degrees, this threatens to blur the distinction between moral virtue and continence (enkrateia). Continence involves acting appropriately (VII.1.1145b9), in accordance with one’s praiseworthy rational choice (VII.1.1145b11; I.13.1102b15), and either despite bad desires (VII.1.1145b14; VII.9.1152a1) or, if the non-rational part of the soul is obedient to reason, then it is less obedient in the continent than in the virtuous person (I.13.1102b26–28). If people with a lower degree of virtue presumably have emotions and desires that do not quite hit the mean (cf. II.9.1109b14–25), or their emotions and desires are not quite as obedient to reason as those with a higher degree of moral virtue and so can show improvement in these areas, then they appear to be much like those who are continent.

I concede that the line between moral virtue and continence is blurry, partly because Aristotle warns us that ethics is imprecise but primarily because there are several kinds of continence. The focal meaning of continence is the disposition (ēthos) (VII.1.1145a15) that falls short of temperance (cf. VII.4.1148a15, 1148b11; VII.5.1149a21–23). As a disposition, continence seems less stable than virtue. (When Aristotle refers to continence as a hexis (VII.2.1146a14), it is in the context of setting out a puzzle concerning continence and temperance. He also says that continence and endurance, which involves resisting pain, are not identical with virtue, but they are not in a different genus (VII.1.1145b1).) There also are qualified forms of continence, such as continence with respect to anger, honor, or gain (VII.4.1148b12–14; VII.5.1149a3). Both continence and endurance are thought to be good and praiseworthy (VII.1.1145b8–9). However, depending on how continence is framed, it might even be bad, such as if it restrains us from following good desires (VII.2.1146a14–15).

Since continence refers to a variety of dispositions, any boundary between it and a particular moral virtue will depend upon the kind of continence in question. The sort of “continence” that restrains us from following good desires is not at all related to the lowest level of a virtue. The qualified forms of continence, which concern desires for things generically noble and good, such as wealth, gain, victory, and honor (VII.4.1148a22–25), fall short of the mean by desiring these goods to excess (1148a26–27). People who gradually overcome these strong desires will transition to a lower level of moral virtue, but without a “magic moment” where they are clearly virtuous, rather than continent. As for the type of continence that concerns the strong desires for things that are base (cf. VII.5.1148b20–1149a1), which is also continence by analogy (VII.5.1149a2), perhaps the lowest level of moral virtue emerges when people begin to no longer desire these base things. Finally, as for continence simpliciter, the initial resistance of natural, bodily pleasures might be painful and so an indication the person is not yet at even the lowest level of temperance (II.3.1104b4–6). Yet, since temperance, like the other moral virtues, exists in degrees (X.3.1173a19–22), we need not saddle Aristotle with the view that people are truly temperate only when they no longer feel any desires contrary to reason.

The second, related reason for thinking that the phronimos is attainable and attributable to actual humans is Aristotle’s distinction between moral virtue and heroic virtue. At the outset of NE VII, Aristotle remarks,

[The contrary] to brutishness it would be most fitting to oppose superhuman virtue, a heroic and divine kind of virtue, as Homer has represented Priam saying of Hector that he was very good,

For he seemed not, he,

The child of a mortal man, but as one that of God’s seed came.

Therefore if, as they say, men become gods by excess of virtue, of this kind must evidently be the state opposed to the brutish state.

(1145a19–25)

Aristotle adds that heroic virtue is rare (VII.1.1145a27), just as the brutish type is rarely found and, when it occurs, is sometimes due to “disease or deformity” (VII.1.1145a30–31). But if heroic virtue exists, rare though it is, then a fortiori, the morally virtuous phronimos exists14.

5. The Cardinal Virtues Are the Basic Virtues

There is one final objection to Aristotle’s doctrine of the reciprocity of the virtues I wish to consider—it is unrealistic to suppose a person can possess all the moral virtues. Perhaps MacIntyre raises the objection most succinctly: “It is difficult to suppose that he seriously means ‘all’—it seems obvious that one can be genuinely brave without being socially agreeable, yet agreeableness is counted by Aristotle among the virtues, as of course is courage—but that is what he says” ([5], p. 155)15. If Aristotle’s doctrine entails that the decorated, valiant soldier who seems a bit cantankerous (IV.6.1127a11)—perhaps because of the trauma of war—is not genuinely courageous, then the doctrine seems to conflict with our most common beliefs, and so Aristotle could be charged with “defending a thesis at all costs” (cf. I.5.1096a2).

Part of what makes MacIntyre’s objection plausible is that it glides over the role phronēsis plays in linking one moral virtue to another. By themselves, courage and social agreeableness are very different moral virtues, dealing with very different subject matters (which in some situations might overlap). They are linked because both need phronēsis, which is a single intellectual virtue that understands the human good as a whole (VI.5.1140a24–31; VI.13.1145a2). Once the role of phronēsis is clarified, the connection between courage and social agreeableness is a little less implausible.

But only a little. Aristotle’s doctrine needs a way to distinguish between central moral virtues and secondary ones. Irwin’s influential essay on the “disunity” of the virtues of generosity and magnificence, and magnanimity and proper pride, shows this need. Perhaps the doctrine can be partly rescued by identifying a generic virtue that deals with wealth and another generic virtue that deals with honor16. Nevertheless, as Daniel Russell points out,

We could raise yet further problems for Aristotle here. For instance, consider the diversity of his list of the virtues: even if it is easy to see how generosity might require justice, it is much more difficult to see how it might require, say, a good sense of humor or wit (NE IV.8…). Another problem arises from the multiplicity of Aristotle’s list of virtues: although all ancient schools maintained that every virtue requires all the virtues, “all the virtues” takes on rather alarming proportions in Aristotle’s theory. Not only did Aristotle produce by far the longest catalog of virtues (i.e., basic or “cardinal” virtues), but even then it is not obvious either that Aristotle regarded that catalog as complete… or how one could determine when such a catalog is complete in the first place… It is likely that Aristotle thought that the virtues should cover all the “key areas” of life (e.g., one’s finances, emotions like fear and confidence, etc.), but it is far from obvious either that Aristotle’s catalog does cover all such areas or that he thought it did. It is even less obvious how to identify such “key areas,” how to individuate them, and how to know when they all have been identified.

([6], p. 217, n. 26)17

Russell makes some good points. While I am not as sure as he is that Aristotle’s catalog of virtues is the longest, its length suggests that, if he fulfills his promise to make “plain how many they are” (III.5.1115a5), he does so in a backhanded way. That is, his message seems to be that plainly there are lots of moral virtues. There are so many that there is little reason to think that the list is complete.

Indeed, there is a very good reason to think the list is incomplete. Aristotle famously omits piety (eusebeia). Several scholars have tried to account for this (for example, [29,30,31])18. Regardless of whether any of their arguments are successful, that they need to attempt this at all indicates Aristotle’s catalog of virtues is incomplete. Perhaps, just as Aristotle’s function argument ends with the concession that it is an outline of the good and “it would seem that any one is capable of carrying on and articulating what has once been well outlined, and that time is a good discoverer or partner in such a work” (I.7.1098a21–25), so too his account of the particular moral virtues is open to further advances.

Of course, when those further advances add moral virtues to Aristotle’s catalog, it might make the problems worse for the reciprocity of the virtues. What is needed is a defensible distinction between what Stephen Gardiner called “basic” and “non-basic” virtues [15]. That distinction might be grounded in the subject matters of the moral virtues19. But even if we can enumerate different subject matters, it will be challenging and perhaps appear to be somewhat arbitrary to determine which areas are “key” and under the domain of “basic” moral virtues and which areas are secondary and under the domain of “non-basic” moral virtues. Moreover, given that the moral virtues deal with passions and actions (cf. II.3.1104b13), if we could settle on a list of passions, perhaps we could settle on a list of virtues. But how many passions are there? Are any of them more basic than the others? This does not appear promising, especially since not every passion admits of a mean (II.6.1107a9–11).

Nonetheless, Annas suggests, “It is extremely plausible that there is a basis in human nature for at least some of the central virtues” ([18], p. 98). In what follows, I propose an interpretation that develops her suggestion. Namely, the reputable tradition, exemplified by Plato’s Republic IV [33], that there are four “cardinal”20 virtues—phronēsis, justice, courage, and temperance—is implicit in Aristotle’s account, and, if so, the doctrine of the reciprocity of the virtues applies first and foremost to those four21.

One indication of their prominence is the length of the discussion devoted to each of them. The treatments of justice and phronēsis each constitute a “book” of the Nicomachean Ethics, and the treatments of courage (four “chapters”) and temperance (three “chapters”) are longer than the discussions of any of the other moral virtues (which are each a single “chapter”).

Another sign of their importance occurs in Aristotle’s Politics VII, where he prefaces his discussion of the best form of the state with a cursory review of eudaimonia:

For no one would maintain that he is happy who has not in him a particle of courage or temperance or justice or prudence [phronēsis], who is afraid of every insect which flutters past him, and will commit any crime, however great, in order to gratify his lust of meat or drink, who will sacrifice his dearest friend for the sake of half-a-farthing, and is as feeble and false in mind as a child or a madman. These propositions are almost universally acknowledged as soon as they are uttered.

(1323a27–35)

This fits with Aristotle’s practice of typically (but not always; see II.9.1109b14–25) using the cardinal virtues to illustrate general points about moral virtue (e.g., NE I.13.1102b27–28; II.1.1103b1; II.2.1104a20; II.4.1105a17–20). In these contexts, he occasionally mentions other virtues, such as generosity (I.13.1103a6–10) and good-temper (I.13.1103b19–20), but when he does, it is always alongside one of the cardinal virtues.

Returning to Annas’ suggestion, the cardinal virtues have a uniquely prominent connection to Aristotle’s account of human nature. Aristotle advises the “student of politics” to study the soul to the extent sufficient for politkē (I.13.1102a14–1103a5). At the end of NE I, in order to ground the distinction between the intellectual and moral virtues in human nature, it suffices to distinguish between the rational part of the soul and the non-rational part of the soul that listens to reason (1103a2–10)22. But when Aristotle introduces temperance in III.10, he remarks, “After courage let us speak of temperance; for these seem to be the virtues of the irrational parts” (1117b24). Irwin notes, “A reader of Republic IV might take this view” ([25], p. 248; see Rep. IV, 436a ff.). So too might a reader of On the Soul III.9. Although Aristotle does not completely accept Plato’s tripartite soul, he nonetheless accepts the distinction between epithumia and thumos:

It is absurd to break up the last-mentioned faculty: as these thinkers do, for wish (boulēsis) is found in the calculative part and desire (epithumia) and passion (thumos) in the irrational; and if the soul is tripartite, appetite (orexis) will be found in all three parts.

(432b5–7; cf. NE III.2.1111b12)

Consequently, I propose that for Aristotle, courage is a cardinal virtue because it is the central virtue of thumos, and temperance is a cardinal virtue because it is the central virtue of epithumia23. A further indication of the significance of temperance (sōphrosunē) is that it supposedly preserves phronēsis (sōzousan ten phronēsin) by preserving the judgments about what is to be done (VI.5.1140b11–19).

Phronēsis, in turn, is clearly identified as the central virtue of practical (or calculative) reason and thus is a cardinal virtue. In NE VI.1, Aristotle sets up his discussion of phronēsis:

Previously, then, we said there are two parts of the soul, one that has reason, and one non-rational. Now we should divide in the same way the part that has reason. Let us assume there are two parts that have reason: with one we study beings whose principles do not admit of being otherwise than they are, and with the other we study beings whose principles admit of being otherwise… Let us call one of these the scientific part, and the other the calculating part; for deliberating is the same as rationally calculating, and no one deliberates about what cannot be otherwise. Hence the rationally calculating part is one part of the part of the soul that has reason. Hence we should find the best states of the scientific part and the best state of the rationally calculating part; for this state is the virtue of each of them.

(1139a3–17 [25]; cf. VI.11.1143b15)

The remaining cardinal virtue is justice. One approach, which St. Thomas Aquinas adopts, is to ground it in the will24. If what Aristotle refers to as the “wish” (boulēsis) is sufficiently similar to what Aquinas refers to as the “will” (voluntas), insofar as the “wish” is the rational appetite (cf. On the Soul III.9.432b5–7; III.10.433a24; Rhetoric I.10.1369a2–7)25, this move may be open to Aristotle (cf. NE V.1.1129a7–10)26. Another option is to ground justice in our nature as social and political animals (cf. NE I.7.9–11; Politics I.2.1253a1–20). Without getting bogged down in the many details of Aristotle’s account and the challenges of reconciling it with other passages in the Ethics, he insists general justice—which corresponds to the sense of justice as lawful “is complete virtue, but not absolutely, but in relation to our neighbor” (V.1.1129b27). Furthermore, the two particular forms of justice—distributive and rectificatory—which correspond to the sense of justice as fairness, concern relations to other people (cf. V.2.1130b30–1131a9). While this approach has the advantage of grounding it in an aspect of human nature that Aristotle indisputably stresses, it involves a different feature of human nature than the various parts of the soul.

Therefore, there is strong evidence that Aristotle takes the traditional list of cardinal virtues to be central, and the reciprocity of the virtues certainly applies to them27. Does he think there are other central or basic virtues alongside them? After all, if the central or basic virtues are the ones that are required to be unconditionally good, as Gardiner argues ([15], p. 265), it seems that generosity and good temper are also required. For the opportunities for exercising generosity and good temper are almost as frequent as the opportunities for temperance and more frequent than the opportunities for courage, especially if Aristotle’s account of courage is as Homeric in its martial focus as Andrei Zavaliy argues [39].

Nevertheless, as important as generosity and good temper are, I suspect Aristotle thinks they are secondary to the cardinal virtues, for it is courage and not good temper, and temperance and not generosity, that he calls “the virtues of the irrational parts” (III.10.1117b24). This is not to say that the phronimos can be courageous but irascible, or temperate but miserly. But he seems to think the virtues of courage and temperance provide a unique foundation for the acquisition of the other moral virtues.

The best evidence for this lies in his recommendations for the moral education of youth found in Politics VIII. He notes the following at the outset of VIII.3:

The customary branches of education are in number four; they are- (1) reading and writing, (2) gymnastic exercises (gumnastikē), (3) music (mousikē), to which is sometimes added (4) drawing. Of these, reading and writing and drawing are regarded as useful for the purposes of life in a variety of ways, and gymnastic exercises are thought to infuse courage.

(1337b23–26)

Although Aristotle is a bit skeptical about the effectiveness of gymnastic exercises for the development of courage, at least as it was practiced by the Spartans (Politics VIII.4.1338b8–38), he does grant its value, especially when it is aiming at what is noble (cf. Politics VIII.4.1338b29). He also recommends that the body should be trained before the mind (Politics VIII.3.1338b5–7; VIII.4.1339a6–10), which suggests that some degree of courage, short of full moral virtue, is attained before some of the other moral virtues are acquired.

As for music, Aristotle speculates that “originally it was included in education, because nature herself, as has been often said, requires that we should be able, not only to work well, but to use leisure well; for, as I must repeat once again, the first principle of all action is leisure” (Politics VIII.3.1337b29–32). He rejects amusement as the telos of life (1337b35–36), just as he did in NE X.6 (1176b9–35). Hence, the true value of music does not lie in amusing the audience. Rather, its value in education lies in the ways it contributes to the “intellectual enjoyment in leisure” (1338a21). When he returns to the subject in VIII.5, Aristotle argues that “music conduces to virtue” (1339a21), with particular attention to courage and temperance. He explains,

Since then music is a pleasure, and virtue consists in rejoicing and loving and hating aright, there is clearly nothing which we are so much concerned to acquire and to cultivate as the power of forming right judgments, and of taking delight in good dispositions and noble actions. Rhythm and melody supply imitations of anger and gentleness, and also of courage and temperance, and of all the qualities contrary to these, and of the other qualities of character, which hardly fall short of the actual affections, as we know from our own experience, for in listening to such strains our souls undergo a change.

(1340a15–23)

The chapter concludes, “Enough has been said to show that music has a power of forming the character and should therefore be introduced into the education of the young” (1340b10–14)28.

Aristotle is, unfortunately, silent on how the various moral virtues discussed in NE IV are connected to the cardinal virtues, and how the acquisition of courage, temperance, justice, and phronēsis can foster the acquisition of, e.g., generosity and good temper. He seems to think it is plausible that someone who has begun to acquire some level of courage, temperance, and justice, along with a rudimentary level of phronēsis that recognizes eudaimonia involves virtue, at some level will desire and take pleasure in the right ways concerning the class of things generically noble and good, such as wealth, gain, victory, and honor (cf. VII.1148a22–25). This includes, should the opportunity arise, great wealth and honors (cf. IV.2–3). Perhaps someone whose fear and confidence in the face of danger is, at some basic level, obedient to reason is apt to feel anger in a manner that is, at some basic level, obedient to reason too. Perhaps someone whose desires for food, drink, and sex are, at some basic level, obedient to reason is apt to desire wealth, gain, and honor in a manner that is, at some basic level, also obedient to reason. At the very least, such a person is law-abiding and fair, with enough phronēsis to know how to make some good decisions in different situations. Hence, that person would not need convincing to value what is noble concerning the goods of other moral virtues and can readily learn how to make good decisions in situations that involve those goods. Such a person is like the proper student of politikē (NE I.4.1095b4–9).

6. Conclusions

We have strong reasons for thinking that Aristotle’s phronimos is intended to be attainable and attributable to actual human beings. Considering the importance Aristotle’s ethical theory places on preserving as many of the common and authoritative opinions as possible, he likely would resist an account of the phronimos that is unattainable and hopelessly ideal. Considering his acknowledgement of the imprecision of ethics, he likely would resist both an attempt to precisely delineate the boundaries between pre-virtue, continence, and virtue and an attempt to catalog an exhaustive list of moral virtues. His clear defense of the attainability of eudaimonia, which he defines as virtuous activity over a complete lifetime, shows that the moral virtues and phronēsis should be attainable. He offers strong reasons for thinking that moral virtues and phronēsis exist in degrees, and not all phronimoi have the same virtues to the same degree. His distinction between moral virtue and heroic virtue implies that, since the latter exists, the former is attainable and attributable. His account of human nature grounds the four cardinal virtues of courage, temperance, justice, and phronēsis, and his account of moral education suggests that he takes these four virtues to be central and basic. Hence, it is reasonable to think the reciprocity of the virtues applies chiefly to them, and that the phronimos who has these virtues is well suited to acquire any other moral virtue.