Any taxonomy risks distortion. But taxonomies can also focus our attention on elements of genuine importance in a thinker’s thought.
Nussbaum, Reference Nussbaum1999: 180.Footnote 1
So, I shall avoid taxonomy to avoid distortion; I shall focus on the thought of a single thinker: Julia Elizabeth Annas—a distinguished exponent of ancient ethics and a resilient defender of her version of virtue ethics; and, I shall focus on one element in that thinker’s thought which, I believe, is of cardinal importance to moral philosophy in general, and to virtue ethics in particular.
Annas has offered a powerful, wide-ranging theory of virtue ethics. It is a theory that accords virtues, the virtuous person, and the virtuous community primary roles. A virtue is a state or disposition of a person. But, the disposition, says Annas, is not just a mindless habit; it is “a disposition to act for reasons.”Footnote 2 Thus, virtue is a disposition to act appropriately, in the right way for the right reasons. Virtues, reasons, and right acts are indissolubly linked in a virtuous person. An act isn’t right, according to Annas’ theory, because it is in conformity with consequentialist rules or universalizable Kantian maxims. An isolated act isn’t right because it just happens to be in accord with what a virtuous person would have done. An act is right because, ultimately, it is an act that proceeds from a virtuous person; that is, a person with a settled virtuous disposition.Footnote 3
It is a mark of a virtuous person that he does the right act without any internal conflict, a conflict that would be a clear indicator that the agent has some way to go before becoming fully virtuous. Of necessity, virtue must begin with mimicking a role model or blindly following the rules; but, ultimately, “it requires that you develop a disposition to decide and act that involves the kind of understanding that only you can achieve in your own case.” Understanding and reasons play indispensable roles in decision-making as well as in the acquisition of virtues; and, in turn, the virtues have an inalienable connection with the final end: “virtue is devoted to our achieving the final end, which … is not in this way an end we can just cease to want.”Footnote 4 We can no more give up the final end than we can give up our humanity.
What is that final end? The end is flourishing—‘happiness’ may be more appropriate—but a wide variety of consequentialist doctrines already have a proprietary right over that term. What is distinctive about flourishing in Annas’ version of virtue ethics is that it cannot be specified as something independent of the virtues which the virtues then enable us to reach or obtain.Footnote 5 This would be to take an instrumentalist view of virtues—a view Annas staunchly (and, to my mind, rightly) opposes: “virtue ethics tells us that a life lived in accordance with the virtues is the best specification of what flourishing is.”Footnote 6 Perhaps, as in Kant, no one can be wholly virtuous; and, therefore, no one can be wholly flourishing. Yet, we must constantly aspire to be fully virtuous. But, if my aspiration to be fully virtuous is to be a constant feature of my life, it is quite essential, says Annas, that I must think of my life as a whole.Footnote 7 Then I will understand why the notion of flourishing is not a quality of this or that particular part of my life, in isolation from the rest, but, rather it is something that characterizes my life as a whole, a life in which I have steadily aimed to be a morally better person.Footnote 8
Let us, then, use as a working definition of a ‘virtuous person’ what Annas offers: “A person is virtuous only when his due actions are ‘complete’—when they are not only what he ought to do, and done because he recognizes this, but done from a fully virtuous disposition, one, that is, in which the reasons offered are unified and based on understanding and the emotional and other responses are fully harmonized with the reasons held.”Footnote 9 In other words, a virtuous person is not merely one with a settled trait; it is a trait that is nourished in, or oriented towards, ethical value.Footnote 10 Thus, a virtuous person is not only possessed of a fully virtuous disposition: he is disposed to do what he understands he ought to do and his reasons, emotions, and natural responses are in mutual harmony. The point of view of the virtuous agent—depicting his reason, labeled ‘virtuous reason’—is quite central; in terms of the traditional classification, Annas’ is a quintessential agent-centered theory.
Finally, a virtuous person is not made from nothing. A virtuous person is fashioned in a community of the virtuous,Footnote 11 in which the person learns what the right action is by having learnt—in the day-to-day lived world, in an “embedded context”Footnote 12—the right examples and the right judgments from the right models—the virtuous—in the community in which he or she lives.Footnote 13
This is barely a schema of Annas’ theory of virtue ethics; nevertheless, every element in the foregoing account is a significant part of that theory and will play an important role in the ensuing. Now, Annas’ intriguing theory of virtue ethics faces a pretty powerful objection—an objection that Annas herself has brought to our attention, signaled its importance to virtue ethics, and, indeed, given it a name: the Circularity Objection. This objection is stated and developed in Section I. The next four sections scrutinize in detail Annas’ various attempts to show how her theory of virtue ethics can avoid the Circularity Objection. Therein I argue that the sheer depth, inventiveness, and fertility of her proposed solutions notwithstanding, the Circularity Objection remains tenaciously unyielding; and, in one solitary case (Section IV), I argue that her solution to the Circularity Objection, if adequate, threatens to lead to relativism. In Section VI, I offer a brief conclusion.
At the outset, let me clearly state my aim in this paper and its limits. While I discuss some of Annas’ central works, such as, An Introduction to Plato’s Republic, The Morality of Happiness, as well as her latest book, Intelligent Virtue, and several of her classic papers, some markedly historical, my aim is not with her exegesis on Plato, the Stoics, or any of the other ancient ethicists—that is far beyond my competence. My aim in this paper is strictly philosophical.
I. The Circularity Objection
Even in one of her earliest works, An Introduction to Plato’s Republic, Annas had perspicuously and crisply delineated the Circularity Objection and what needed to be done to avert it:
We have to have some independent way of saying what the good person is like. It is clearly unhelpful to say that just acts are those the just person would do, if all you can say then about the just person is that he or she is the kind of person who would be likely to do just acts. You have to be able to say independently what kind of person the just person is.Footnote 14
Let me lightly reconstruct this as follows:
We have to have some independent way of saying what the virtuous person is like. It is clearly unhelpful to say that virtuous acts are those the virtuous person would do, if all you can say then about the virtuous person is that he or she is the kind of person who would be likely to do virtuous acts. You have to be able to say independently what kind of person the virtuous person is.
First, this reconstruction has the advantage of being more general than Annas’ version; it speaks of a virtuous person and virtuous acts rather than of a just person and just acts (and it does not speak of a good person who, arguably, is meant to be a just person in Annas’ version of the Circularity Objection).Footnote 15 It is more general in that a virtuous person includes the just person; the converse is not obviously true (except under the hypothesis of the unity of virtues). It is this version of the problem that is the highlight of this paper and Annas’ proposed solutions will be measured solely in its light.Footnote 16
Second, it retains the all-important idea in Annas’ version, namely, that we need some independent way of saying what makes for a virtuous person—independently of being one who performs virtuous acts. I shall call this the Independence Test. Annas maintains, and I endorse, that any adequate solution to the Circularity Objection must pass the Independence Test.
Let us explore the Circularity Objection a bit to establish some background. In An Introduction to Plato’s Republic Annas argues that Plato was trying to avert the Circularity Objection. Recall that the challenge posed by Thrasymachus was that it paid to be unjust, while Glaucon and Adeimantus demanded that Socrates demonstrate not only that it did not pay to be unjust, but also that it did not pay to be unjust even if one could get away with it. Now, on Plato’s view, a just soul is an ordered soul; ergo, it is in psychic harmony. Such an ordered soul is essentially disposed to doing just acts. But, one cannot draw up a set of consequentialist rules or Kantian maxims that will tell that soul in advance of being in concrete moral dilemma or circumstance what it ought to do. Therefore, if a just soul is defined neither in terms of right acts nor rules (which, in turn, define right acts), the Circularity Objection is averted.Footnote 17
My task is not to take up whether Plato successfully met the challenges he had these three men pose to Socrates or whether his solution ran afoul of the fallacy of equivocation or whether his philosophical method of treating this particular problem was adequate.Footnote 18 My task is to ask, given that Plato’s answer (whatever it is) is satisfactory, whether it circumvents the Circularity Objection (as formulated above). And, if it does, does it still retain the centrality of the virtuous person? Or, is the virtuous person dispensed with in favour of something else? More expansively, if Plato’s answer—whatever it is—has successfully shown what a virtuous person is like independently of the rightness of an act (that does not appeal to rules, maxims, or to a list of moral dos and don’ts; I don’t know how that can be done), then it is clear that Plato will have satisfied the Independence Test and escaped the Circularity Objection. And then right acts could be defined, without circularity, as the acts performed by a virtuous person, a person with an ordered soul. Let us call this the First Alternative. This alternative is exemplified in Sections II and III. As we shall see (Section III), one contemporary philosopher has outlined a solution along Platonic lines. But, that solution is not without its own difficulties; at any rate, I shall argue, it is a solution no contemporary virtue ethicist, akin to Annas, could accept.
On the other hand, if Plato’s answer—whatever it is—has successfully shown what the rightness of an act consists in, without invoking the notion of a virtuous person, and then defined a virtuous person as one who performs such acts,Footnote 19 then again it is clear that Plato will have satisfied the Independence Test and escaped the Circularity Objection. Let us call this the Second Alternative. This alternative is exemplified in Sections IV and V. But, then, it is equally clear that the Second Alternative will have relinquished making the virtuous person central in its theory—a conclusion no contemporary virtue ethicist, akin to Annas, can accept. To such a virtue ethicist, the Second Alternative, arguably, would be the least attractive.
With this as a backdrop, let us turn to Annas’ four ways of circumventing the Circularity Objection.
II. The First Solution: The Just Soul (or, the Virtuous Person)
To better understand and evaluate the First Solution, in which the just soul or virtuous person plays the principal role, we should begin with two important remarks allied to the First Alternative. Here is the first one. The “just person, we are told, is the person who is integrated and morally healthy. This account makes no appeal at all to just acts,” says Annas. “It tells us in some detail what a just person is without reference to the way he or she acts.”Footnote 20 A just soul or virtuous person is in a state of ‘psychic harmony’—a singularly important notion. (We shall turn to it in the next section.) All the elements of the soul are well-ordered in him, nothing out of place, nothing out of tune. Such a person has become virtuous by being brought up well, not by attending lectures on morality; he has had an education that builds character, not a way of stacking academic accomplishments.Footnote 21 This education is indispensable for the agent’s own well-being; and, so, it is an answer to “one’s human nature; it is the fulfillment of one’s basic needs.”Footnote 22
Here is the second crucial remark wedded to the First Alternative: “if just acts are to be specified by what the just agent will do, then we cannot lay down in advance what acts are just acts. … If we take seriously the primacy of being a certain kind of person, we cannot draw up lists of rules or maxims which will settle the question of which actions are the just ones—for this will depend to a great extent on the circumstances.”Footnote 23 This seems plausible to Annas since circumstances a virtuous person—say, a Guardian or Philosopher King—will usually face in life are enormously rich in variety, force, and complexity, not the sort of thing one can effectively respond to by mechanically appealing to an antecedently established scroll of rules and maxims, or schedule of duties and obligations, laid down by some “external source.”Footnote 24 If there are no mechanical rules to produce a work of art, then think of morality as art. An original work of art is a product of an artistic mind (which is not to say that no artistic gift or labour is involved); likewise, to produce a genuine moral deed, it, too, must be a product or an off-shoot of a virtuous agent (which is not to say that no proper upbringing, learning, and willing is involved).
If the First Alternative is satisfactory, it will have satisfied the Independence Test, circumvented the Circularity Objection, assigned the virtuous person a cardinal role in ethical theory, and succeeded in defining right acts in terms of what the virtuous person does or would have done.
Consider an objection that springs from the uniqueness of a circumstance. Let us say that there are several Guardians or Philosopher Kings with a right to lie to those beneath them in the social hierarchy, but not to one another. When they lie, they do so by keeping in view the overall good of the city. This is how Annas captures the point: “it is because [the Guardians or Philosopher Kings] are so keen on the truth, because they have natures that love it, that they can lie on occasion—for then it would be the right thing to do in the light of the whole truth. … We are not to worry about their breaking the rules, because, when just people do this, the rules ought to be broken.”Footnote 25
Now, imagine the Guardians or Philosopher Kings finding themselves in a fairly unique circumstance, something they had not encountered before (let alone ruled or decided on), and they are to determine what should be the right thing to do in that unique circumstance. (A) One might hasten to think that in that unique circumstance whatever a Guardian or Philosopher King (the virtuous person) did would be the right thing to do. This has the odd consequence that, if one Guardian or Philosopher King did X and the other did not-X, in the same unique circumstance, then both would be right.Footnote 26
(B) Since (A)’s consequence is quite unacceptable, we might be reminded that we have overlooked an important codicil, namely, that what is the right thing to do is “in the light of the whole truth.” If the whole truth is perceived in one way by one Guardian or Philosopher King, whereas a different Guardian or Philosopher King views that whole truth differently—in view of their distinctive sensibilities, different upbringings, particular experiences—then we simply fall hapless victims of the problem outlined in (A). For, given that each Guardian or Philosopher King views the whole truth differently, one decides to do X and the other not-X, each act is right only from different perspectives. Even if this is not a flat out contradiction, we need to be told a lot more about these unique perspectives that render opposed actions morally right. But, let us not lose sight of this important fact: no amount of story-telling about the virtuous person will do, if it simply succeeds in supplanting the person. The work that is being done is by the perspective (that view of the whole truth), not by the just soul or virtuous person (except incidentally in that the perspective happens to be his).
(C) On the other hand, if the whole truth is perceived uniformly by all the Guardians or Philosopher Kings, then quite clearly it is the agent-independent moral truth—the whole truth—that determines what ought to be done, not the agent: a particular Guardian or Philosopher King.
Let us turn to Ordinary Justice.Footnote 27 Ordinary Justice—packed with conventional rules and maxims—is not sufficient to acquire Platonic justice because, lacking a critical stance, an agent might become too shallow, rigid, even inapt by mechanically, if conscientiously, following the social norms. A virtuous Platonic agent has a kind of flexibility, is nuanced in his moral discrimination, which enables him to better and more effectively respond in morally appropriate ways than would someone else imbued merely with Ordinary Justice. Nevertheless, Ordinary Justice cannot simply be cast aside. It serves as a good starting point; it also serves as a provisional constraint on ethical theory—to be discarded as and when deemed later. Not only that but since Ordinary Justice holds that certain deeds—murder, incest, adultery, parricide—are so foul that no one should commit them, then even Platonic theory must be constrained by—better yet, explain—these norms; if it violates those norms, it will be unacceptable as a moral doctrine.
Impeccable though the point is, it must give us pause. If—one might argue—certain deeds, like murder and mayhem, should not be performed, it is surely not enough, if we are to conduct a proper moral inquiry, to simply state that such and such deeds are prohibited; that no ordinary person, let alone a virtuous person, should perform them. We must ask, why should these deeds not be performed. Why do such deeds, in particular, act as constraints on an adequate moral theory? Either our answers to such questions would take us into the realm of theory—say, Kant’s Formula of Humanity as an End Itself—which will explain why we morally ought to refrain from those gross deedsFootnote 28 or, our answers, to be consistent—as in Stoic theory—would have to hold the virtuous agents in such high esteem that, if they performed these deeds, then these otherwise-heinous deeds would be transmuted into moral deeds.Footnote 29 What is significant in this solution—inadequate though it is—is the just soul or virtuous person. This person has a significant property; it is the property of his soul being in “the state of psychic harmony.”Footnote 30 And, it is precisely this state of psychic harmony “which non-trivially explains its manifestations in just behaviour” or right actions.Footnote 31
III. The Second Solution: Psychic Harmony
The notion of psychic harmony, then, is worth careful examination. Either the notion of psychic harmony is a moral notion, a notion that is defined as that which engenders right acts or it is not. If it is not a moral notion, then it would be astonishingly serendipitous that non-moral psychic harmony just happens to issue in moral actions; that alternative will simply not do. But, couldn’t psychic harmony—wherein all the elements in the soul are well-ordered (as they are in a just soul or virtuous person)—be a moral notion that does not invoke, explicitly or implicitly, any notion of a right act, and yet issues in right acts from a well-ordered soul? And, if so, would that not be a satisfactory—remarkable, in fact—exit out of the Circularity Objection?
There is just such a solution. In Self-Constitution: Agency, Identity, and Integrity, Christine Korsgaard, professedly deeply indebted to Plato (and Aristotle and Kant as well), argues that as moral agents our fundamental task is self-constitution. Each of us has a disordered self; and, we are under an imperative to set it in order. Indeed, there is no self prior to having made choices and performed actions, for it is choices and actions that constitute one’s identity.Footnote 32 Unless one has identity, there is no agency; and, there is no agency where there is no unity.Footnote 33 To the extent that an agent is self-constituted—his soul is well-ordered—what he performs are actions, in sharp contrast to being merely acts. Acts can be random, mechanical, reflexive things; actions, by contrast, are performed self-consciously, self-reflectively, under self-guiding principles.Footnote 34 Since self-constitution comes in degrees, actions come in degrees.Footnote 35 Given that Socrates’ soul is so marvellously ordered, when, for example, he drinks the hemlock or refrains from sleeping with Alcibiades, what he performs are actions par excellence. Lesser men, having selves that are generally in shambles, perform deeds that are less than full-blooded actions.
Now comes the key point. Ultimately, what makes for self-constitution? There is exactly one principle or imperative that can order the self, no others play that key role—for example, neither the hypothetical imperatives, the principle of prudence, nor the principle of self-love could be cast in that principal role.Footnote 36 That principle, says Korsgaard, is Kant’s Categorical Imperative. So: imagine a person—say, Socrates—whose soul or self is well-ordered by the Categorical Imperative, and as such his soul enjoys psychic harmony. In other words, that self is self-constituted as it should be constituted and the natural off-shoot of that self-constitution is that whatever action Socrates performs is indeed an action; his words (promises, and such) and deeds are in harmony.Footnote 37 This is the kind of person all humans should be aiming to become. Kant and Korsgaard, I suspect, would happily resonate with the following from Plato: “Whenever I hear a man discussing virtue or some kind of wisdom, then, if he really is a man and worthy of the words he utters, I am completely delighted to see the appropriateness and harmony existing between the speaker and his words …; but the man who acts in the opposite way distresses me, and the better he speaks, the worse I feel.”Footnote 38 This, in a pithy form, is the essence of Korsgaard’s view.Footnote 39 Neither Annas nor the reader need agree with it. We are simply going to use it to raise a question or two for Annas’ view.
If, then, the notion of psychic harmony is a moral one, but not one that is defined by right acts, then it seems as if such a Platonic view satisfies the Independence Test and escapes the Circularity Objection. Now, either the notion of psychic harmony is left undefined or it is defined. To leave the notion of psychic harmony undefined is simply to abandon the philosophical task, so I will not pursue that alternative.
But, if we try to define psychic harmony—say, as Korsgaard does, in Kantian terms (and it has its own attractions)—then, of course, that will not do because Annas’ version of virtue ethics will become indistinguishable from a deontological doctrine; or, at the very least its foundation must be deontological in nature. It will equally not do, of course, to appeal to a fundamentally consequentialist principle to define psychic harmonyFootnote 40 since then Annas’ version of virtue ethics will be indistinguishable from consequentialism. Therefore, Annas’ task must be to define psychic harmony, without using the Categorical Imperative, or any other deontological or consequentialist moral imperative, which issues in moral rules and maxims and which, in their turn, define what it is for an act to be right. But, the prospect for doing so seems dim—particularly since in that view the virtuous agent has lost his supremacy in favour of whatever fundamental principle(s) are invoked to define psychic harmony. After all, psychic harmony is neither merely psychic, nor merely harmony, it is moral psychic harmony. And Annas’ task is to show how that kind of harmony can be understood or defined without invoking moral rules or principles which determine right or virtuous actions.
And so we get this conclusion: either psychic harmony is defined in terms of some ultimate moral rule(s), in which case Annas’ version of virtue ethics loses what fundamentally and uniquely distinguishes it from any rule-governed morality (deontological or consequentialist). Or, psychic harmony is defined only in terms of the notion of a virtuous agent (without explicitly or implicitly using the notion of a rule or right act in that definition), who either somehow mysteriously produces right actions, thus leaving philosophical matters unexplained or, who produces right actions as a result of being in psychic harmony, thus bringing us right back to the Circularity Objection: an act is right if and only if it is produced by the state of psychic harmony; and, a state is in psychic harmony if and only if it likely produces right actions. This is a specific instance of the general schema of the Circularity Objection.Footnote 41
IV. The Third Solution: The Community of the Virtuous
Since much rides, for Annas’ version of virtue ethics, on the Circularity Objection, it is worth scrutinizing Annas’ different and more comprehensive attempts, delineated in The Morality of Happiness as well as Intelligent Virtue, at escaping that circularity. Admittedly, “the virtuous person, the sophos or wise person, is as rare as the phoenix.”Footnote 42 Even so, to strengthen the argument, let us assume that we have a substantial community of virtuous persons.Footnote 43 Whatever philosophical arguments are to be provided, they should be made against the background of this assumption. This will enable us to circumvent the obvious rejoinder that it is no part of Annas’ theory of virtue ethics that it must attempt, let alone succeed in attempting, to convert the unvirtuous, or someone who is ethically neutral, to becoming virtuous.Footnote 44
One might ask, in such a community of virtuous persons: “what here is the standard of rightness?”Footnote 45 There are two options: either the standard of rightness is derived from virtuous agency or it is derived from something—anything—that is independent of virtuous agency. The second option is clearly unacceptable because it would make the appeal to virtues or a virtuous person secondary since what matters on that option is something—anything—that is independent of virtuous agency.Footnote 46 Then, the latter would be explained in terms of the former. This would make Annas abnegate the privileged position of the virtuous person in her theory in favour of that which is independent of it. Clearly, then, this will not do.
Consequently, it is imperative that the focus is on the first option which will immediately lead to the question: “how do we know which are the virtuous people?”Footnote 47 Significantly, and quite rightly, Annas dismisses the answer that we must resort to social consensus. Instead, she says, we should “appeal to virtuous reasoning itself.”Footnote 48 (There is this potential Circularity Objection: a person is virtuous if and only if he has the property of virtuous reasoning; and, reasoning is virtuous if and only if it is the property of a virtuous person. If Annas is to avoid this objection, clearly she must specify virtuous reasoning independently of the virtuous person. Given the indissoluble connection between virtues, reasons, and right acts in Annas’ version of virtue ethics,Footnote 49 this is not an innocent objection. Let us bypass this objection.)
Now, if the beginner is lucky enough, he will have been presented with the right examples (these fix the content of particular virtues),Footnote 50 by the right models (minimally, these are approximately the most virtuous in the community whom he must mimic at the start), which will lead him to make the right judgments (these will not only prompt right actions, but will also enable him to explain them).Footnote 51 And in time such a learner will take his rightful place in the community of virtuous persons. “As he becomes more virtuous,” says Annas, “he understands the basis of and reasons for bravery; so he can explain to others why one should do this rather than that. So the criterion for right action is indeed what the virtuous person would do. But this is not trivial. The beginner can appeal to it for guidance; and the virtuous person cannot merely come out with a judgment, but explain it.”Footnote 52
If one does not see the way out of the Circularity Objection, Annas protests, it is because one is hemmed in by a notion that an ethical theory must be hierarchical and complete.Footnote 53 A theory is hierarchical if it has certain basic terms in the light of which other non-basic terms are reduced or defined; thus, in virtue ethics, character and virtue should be deemed basic notions in terms of which right action is to be reduced or defined. And if virtue ethics is unable to execute this directive, then it is deemed insufficient. The alternative, of course, is to treat something other than character and virtue as a basic notion—say, a state of affairs—in which case virtue ethics would relinquish its right to be regarded as a distinctive, non-consequentialist ethical theory. Clearly, Annas must deny that her version of virtue ethics is hierarchical and complete; and so, indeed, she does.Footnote 54
Annas’ argument is that there is a different notion that must be appealed to: the notion of what is primary. “Most of us,” Annas rightly contends, “want not merely to do the right thing but to understand what it is about it that makes it the right thing … Our natural first step is to examine the particular reasons we have for doing one thing rather than another. But … the particular reasons and intuitions of particular people contain a complex mixture of prejudice and convention which is hard to filter out. The next step, therefore, is to move from particular judgments to find the general procedure we follow in coming to those judgments.”Footnote 55 And here, at last, we arrive at a critical juncture in the argument.
Diverted by concerns about hierarchy and completeness, says Annas, traditional ethical theory attempts to erase convention and prejudice. Then, having done so, it asks whether the ethical theory is consistent, and whether it possesses other theoretical virtues, such as, simplicity, unity, non-ad hocness, and so on—much as a physicist or biologist would ask those very questions with respect to his scientific theory. It is not that virtue ethics is uninterested in those questions, but they are not the questions it asks first; they get asked later.
Her version of virtue ethics, declares Annas, is distinguished by its markedly different approach:
[I]t takes that we first have to see the significance of these procedures from the agent’s point of view before we go on to ask these further questions. If the agent aims to produce good consequences, why is this so from the agent’s standpoint? Suppose the agent is following a rule, such as, ‘Don’t do cowardly things.’ Why does the agent accept such a rule—what point does he see in it? For virtue ethics, agents are primary for understanding, because if we want to know what the point is of avoiding cowardly acts, we shall understand this best if we look closely at the agent, and what it would be for him to be brave or cowardly. … Thus in an ethics of virtue initial interest in what we ought to do moves to an interest in the kind of people we are and hope to be, because the latter is taken to be the best way of understanding the former.Footnote 56
This passage captures the heart of Annas’ view; she could not be clearer that the virtuous agent is the ultimate arbitrator. Clearly, the agent’s point of view is indispensable if we wish to understand why she did what she did: she did not want to dishonour her country; she wanted to partake in the revolution to overthrow the dictatorship; or, she could not sit idly by watching homeless and underfed veterans straddling the streets. Now, we need an explanation of one sort, if we wish to understand; we need an explanation of quite a different sort, if we are seeking ethical justification. The original question is, “what here is the standard of rightness?”Footnote 57 So, if someone—let us call her ‘Fatima’—were to willingly risk her life, having joined the revolution to displace a Middle Eastern dictator,Footnote 58 thus forcing her already poor children to eke out a living on miserably meager resources, her action can be explained, without that explanation necessarily justifying her action as being morally right.Footnote 59
One might argue on Annas’ behalf that not to see Fatima’s action as right is to neglect three significant claims. First, it might be claimed, our working assumption was that there is a community of virtuous persons in place; so, Fatima is a virtuous person among virtuous persons. Second, virtuous agents in this community are reflective agents; so, Fatima is not “content with mindless performance”Footnote 60 of her action. When Fatima “parts company with the initial consensus”—consensus of the community of virtuous persons—“she can give reasons for so doing.”Footnote 61 Third, right action is a joint product of the “relation between the virtuous person’s viewpoint and society’s consensus as to what is right.”Footnote 62 This, of course, is just reflective equilibrium.
We can bring this all together, so the argument might continue, to demonstrate that this avoids the Circularity Objection. Neither the community, nor the act, nor the agent is basic: all three are primary.Footnote 63 The rightness of Fatima’s act—let us assume that it is right—takes seriously Fatima’s own point of view; but, that point of view is not isolated from the point of view of the community of virtuous agents to which she belongs. Furthermore, Fatima is as reflective, not uncritical, of her own actions as is her community of its own customs and practices. What results—Fatima’s heroic deed—then is right action: a joint product of the relation between her viewpoint and her society’s consensus about what is right. Fatima is both able to explain and justify why she did what she did.
Why, then, isn’t all this eminently plausible? Consider some cases by way of initiating the counterargument. Case 1: Let us assume that there is communal consensus. Fatima—from her own personal point of view—had particular reasons for doing what she did and believed her action to be right (even her children, say, were in support of it). Fatima’s community of virtuous persons also believed that Fatima’s action was right. But, it is simply not true that from Annas’ perspective it is the communal belief—thoughtfully and critically held—which makes Fatima’s action right. For one thing, consensus is hardly a guarantee that an act is right. It is difficult to improve upon Annas: “We can appeal to society’s consensus on this: the brave person learns what the right thing to do is by noting what is counted in her society as the right thing to do. But this provides no justification. Society’s consensus is all we need to determine what the polite thing to do is, but with the virtues there is the further question, is this consensus itself right.”Footnote 64 Thus, bravery is a virtue, to be sure; but, if Fatima’s act is virtuous because it is brave, it is not virtuous because of the consensus in her community of the virtuous.Footnote 65 There is the further question: is the communal consensus on Fatima’s action itself right?
Case 2: Let us, next, assume that Fatima’s action is not underwritten by communal consensus. After all, Fatima can part from the company of her peers, initial consensus notwithstanding and, what is more, she can actually give reasons why she departs from the communal ways. Note: Fatima has developed “a disposition to decide and act that involves the kind of understanding that only [she] can achieve in [her] own case”Footnote 66 and, furthermore, she has developed her disposition by contemplating the virtues and keeping in view her life as a whole. If the virtuous agent’s point of view—Fatima’s view—is to be given the kind of credence and significance Annas’ version of virtue ethics gives it, and which the more traditional ethical theories emphatically do not, then it is distinctly possible for Fatima’s action to be right even though it lacks the consensus of the community of the virtuous (or, Fatima’s action may be wrong even though it has the consensus of that community).
This essentially guarantees what was argued in Case 1, namely, the consensus of the community of the virtuous does not make the act right (or wrong). But, then, it would be extremely puzzling to claim that Fatima’s act is right because Fatima—via her particular reasons, encased in her point of view, in this case—deemed it to be right. Fatima has her own particular reasons for deeming her act to be right (and offers them by way of explanation); and, the community has its own particular reasons for deeming her act to be wrong (and offers them by way of explanation). Each is offering an explanation, but no justification: consequently, it is anything but obvious why one should lean toward one explanation rather than the other to determine whether Fatima’s act is right or justified.
Case 3: Finally, assume, a bit more realistically, that there are separate, but intricately connected, islands of consensus in the community of the virtuous. Recall how the moral apprentice initially learns from being presented with the right examples, examples that ultimately provide the particular virtues with content. Now, imagine two distinct islands of consensus; there may be strong family resemblances between them regarding, say, the virtues of bravery and prudence. Yet, another woman—let us call her ‘Zahra’—has learnt from her moral exemplars in her community that in extreme circumstances prudence should trump bravery; Fatima, on the other hand, has learnt from her role models in her community the opposite, namely, that one should be brave in the face of adversity (in these circumstances prudence is just another word for cowardice).Footnote 67 Fatima and Zahra, as well as their circumstances, are remarkably similar in all morally relevant respects and now face a common problem, namely, what should one do in the emerging uprising against a relentless dictator? Fatima does the brave thing; Zahra is duly cautious. We have a case in which, if the agent’s point of view is to be given pre-eminence, both Fatima and Zahra are right (both are virtuous), despite the fact that in the morally similar situation each performed an action that is exactly the opposite of what the other performed.Footnote 68 The threat of relativism looms.
This is not a threat that can be easily brushed aside. Annas has criticized—and quite rightly, too—other theories of virtue ethics that over-emphasize the development of virtues embedded within specific traditions of virtue, while failing to give enough emphasis to the aspirational side of virtue, on the ground that they entailed relativism. “Over-emphasis on this, however,” Annas says, “risks falling into relativism, the view that different virtues are developed within the context that cannot be meaningfully compared and thus are removed from mutual discussion and criticism.”Footnote 69 Thus, it would follow that, on Annas’ view, Fatima’s act and Zahra’s act can be meaningfully compared and evaluated, their distinct aspirations notwithstanding; there can be mutual discussion and criticism about their acts; and, this can lead to a determination as to which act—Fatima’s or Zahra’s—was the right one. Objectively right—not right relative to a particular tradition of virtue. There is no incommensurability; but, then, this undermines the agent’s point of view (at least, as far as justification is concerned) and displaces the virtuous person from his central position in Annas’ version of virtue ethics.
Virtues, therefore, in different and distinct traditions are commensurable and they can, and ought, to be compared with each other and that, in time, an open and critical process of choosing and creating different patterns of living—indeed, this process is “a fact about our species”Footnote 70—will, perhaps, lead to a single ideal tradition of virtuesFootnote 71—say, a tradition that flowers in the universal community of rational beings.Footnote 72
V. The Fourth Solution: The Universal Community of Rational Beings
Might looking at all this from the perspective of the Stoics—who placed virtue in a position of pre-eminence in their ethical theory “virtue is the only good”Footnote 73—help? The Stoics would have argued that Fatima and Zahra are not merely denizens of particular communities (say, Egyptians and Syrians, respectively), but also citizens of “the universal community of rational beings.”Footnote 74 Thus, morally speaking, they have “dual citizenship.”Footnote 75 To be a member of a particular community is to heed a communitarian call; to be a member of the other community is to hearken to a cosmopolitan cry. “Yet for the Stoics these are two viewpoints which are clearly taken to coexist in the same person’s moral consciousness.”Footnote 76 Clearly, the two viewpoints can potentially conflict. From Fatima’s and Zahra’s perspectives, aiding in the revolution potentially conflicts with their commitments as mothers (and any other long-term commitments they might have had); but, Fatima solves the conflict in one way, Zahra in the opposite way.
Now, as good Stoics, Fatima and Zahra are not wedded to a moral theory, scratching all their present commitments, and letting the moral theory—“a moral theory is thought of as a theory to live by”Footnote 77—dictate to them what they ought to do and how and why they ought to remodel their commitments. For both are mindful of the following: “When you become a Stoic, you don’t cease to think and act as a son, brother, town councillor; rather, you do your best to be a Stoic son, brother, town councillor. You do this by taking on the ideal of membership in the moral community of rational beings as an ideal that you aspire to live up to as you live your life.”Footnote 78 This is because—and this is crucial—“Stoicism demands that we act within our roles in a way guided by the principles of Stoicism, principles which in the fully virtuous person lead to the cosmopolitan perspective. This requires becoming virtuous, and this in turn requires a specific realignment of our values and priorities.”Footnote 79
This leads to a paradox. By our assumption, both Fatima and Zahra are virtuous persons (ex hypothesi, they are not beginnersFootnote 80); their emotions are duly attuned to the demands of Stoicism.Footnote 81 So, if each of them is guided by virtueFootnote 82 and the same principles of Stoicism,Footnote 83 then—given that they are required by these very principles to reorient their values and priorities—they can scarcely choose to do different things. If it is morally right for Fatima to risk her life for the revolution,Footnote 84 given the agent-independent principles of Stoicism, then it cannot be right for Zahra to shelter hers, and vice versa. But, then, their points of view—the agents’ points of view—are no longer playing the central role in their respective lives; that role is assigned to the principles of Stoicism. And just as significantly, the rightness of the action will no longer be dependent on the virtuous agent; it will be dependent on the principles of Stoicism. And it is those principles that will be telling the virtuous agent what to do. The Independence Test is satisfied; but, the Circularity Objection is averted at the cost of the virtuous agent losing his key status.
On the other hand, if the point of view of the agent is given pre-eminence, then the principles of Stoicism must take a back seat and the Circularity Objection will once again threaten (as in the First Solution).Footnote 85
In her latest book, Intelligent Virtue, Annas writes:
What has emerged is that being fully virtuous does seem to be an ideal that we aspire towards but can never achieve. At best we can be virtuous in a less than full way, one marred not only by our own deficiencies but by the point that the structures of our societies preclude us from being fully virtuous.
This would be troubling if we insisted on a rigorist approach, such that a person either is virtuous or is not virtuous at all. This would have the result that only the fully virtuous person is virtuous, while none of us are virtuous at all. This is in fact the Stoic position, but it is a very awkward one, since it strictly allows for no difference between the mediocre non-virtuous and the horrendously vicious non-virtuous, and that difference has to be brought in again by allowing that there are degrees of progress, though no degrees of virtue.Footnote 86
There are two dominating elements in this passage, one by implication: the fully virtuous person and the ideal structure of our societies. The non-ideal structure of our societies (these can take a vast variety of forms) prevents us from being a fully virtuous person while, arguably, it is only in the ideal structure of our society (can this take a vast variety of forms?) that we can develop into the fully virtuous person (can there be more than one way of being a fully virtuous person?).
Here, then, is why this significant passage is so very enigmatic. (1) If we assume that if the fully virtuous person, reared in a just society—the universal community of rational beings—would have done what Fatima did (and that is what makes Fatima’s act right), then despite the fact that Zahra was a virtuous person (ex hypothesi), what she did was wrong. Then, in Zahra, the twin notions of a virtuous person and the right action have come apart; and, indeed, so have they in Fatima: because the rightness of Zahra’s or Fatima’s actions is not dependent on either of them being virtuous in their respective communities; rather, it is dependent on what the fully virtuous person in the ideal universal community of rational beings would have done. The problem simply shifts to determining why the fully virtuous person’s action in the ideal community is right, and that returns us to the Circularity Objection faced by the First Solution.
(2) If we assume that, because what Zahra did was wrong, she was not really virtuous—she was still at the learner stage—then, there is no need to assume that Fatima is fully virtuous either; being human, fallible, she must have performed some act that was not right. Both Fatima and Zahra are beginners. But, then, this abolishes the distinction between the expert (the truly virtuous person) and the beginner, a distinction Annas is keen to make and retain: “When I use the phrase ‘truly virtuous person’ I imply only someone who contrasts with the beginner, someone of whom we truly say that they are generous, brave, or honest.”Footnote 87 If even an elderly and kindly Roman slave-owner is regarded as a learner, although he “is certainly virtuous by the standards of his society,”Footnote 88 then surely Fatima must be regarded as a learner too. We are all beginners. But, Fatima is no beginner and nor is Zahra. Therefore, each of their acts—mutually contradictory—was morally right.
(3) If we assume the notion of degrees of moral progress, then Fatima has made a greater degree of progress than has Zahra; nevertheless, she, Fatima, does not possess a higher degree of virtue than Zahra (since the notion of the degree of virtue is banned). Oddly enough—as Annas correctly remarks—Fatima has made progress towards being fully virtuous, but for all that she does not possess a higher degree of virtue than Zahra who has markedly failed to make similar progress.Footnote 89
(4) If we assume that there are degrees of progress (but not degrees of virtue), then we must postulate some ideal towards which we inch our way as we increase the degree of our ethical progress. Fatima, for example, has made a greater degree of progress towards that ideal than has Zahra. If we postulate that ideal of a fully virtuous person—whatever ideal Annas deems fit from her view of virtue ethics—then what is a right or virtuous action, and what is not, is defined by that ideal and not by any imperfect person raised by imperfect parents in an imperfect society (which is all of humanity). Once again, a particular agent’s view, ensconced in a particular time, place, and community is replaced by the view of the fully virtuous person in the universal community of rational beings. This leaves us with the unanswered question, what makes a fully virtuous person virtuous? Whatever the answer to that question, it must not default in the way in which the First Solution did.
What, then, is the upshot of the claims (1) through (4)? Annas’ view of virtue ethics pulls us towards two ideals: the ideal of the fully just society—the universal community of rational beings—and the ideal of the fully virtuous person. Now, if Annas could show how the notion of the fully virtuous person can be defined in terms of the fully just society (this is the heart of the Fourth Solution), and then explain right actions in terms of the fully virtuous person, then her theory of virtue ethics would satisfy the Independence Test and escape the Circularity Objection. But, this would displace the notion of the virtuous person from its cardinal role (as did the Third Solution); that role would be assigned to the fully just society. More significantly, the notion of the fully virtuous person (in the guise of the rational being in the universal community) is ineluctably involved—and rightly so—in the notion of the fully just society: the latter notion cannot therefore be defined without the former notion; so, such an escape route is not open to her.
Consequently, Annas is left only with the notion of the fully virtuous person. Now, if she can define the notion of the fully virtuous person in terms of right action (where the latter is defined, for example, in terms of Kantian maxims or consequentialist states of affairs), then she would satisfy the Independence Test and the Circularity Objection would be averted. But, then, she would end up relinquishing the notion of the fully virtuous person as a primary notion—the trade-mark of her theory of virtue ethics. Finally, if she defines the notion of the fully virtuous person in terms of right action and defines right action in terms of the fully virtuous person, then she ineluctably returns us to the Circularity Objection she had so clearly and forcefully warned us against.
VI. Conclusion
Virtue ethics, in Annas’ version, and Kant’s moral theory differ in remarkably deep and significant ways, even though they have some commonality. For example, “don’t-tell-me-what-to-do” is Annas’ shibboleth;Footnote 90 “think for yourself” is the Enlightenment credo, in almost Kant’s words.Footnote 91 Each applauds reason over desire.Footnote 92 Each exalts the virtuous person.Footnote 93 Each praises human wisdom. (Neither needs any reference.) Perhaps, they even agree that what morality demands is unconditional.Footnote 94 And, then, there is a parting of ways. One of these ways is what this paper is about: Annas’ version of virtue ethics highlights the virtuous person making him the ultimate, supreme arbitrator in moral matters; by contrast, in Kant, that agent is subservient to the demands of the moral law. At the end, what one can conjecture—quite dogmatically, I know—is that some ultimate version of Annas’ virtue ethics will show what a virtuous person must be, “the kind of people we … hope to be”Footnote 95 and Kant’s moral theory will show why just such a virtuous person can be understood only in terms of the moral law. Were this convergence to occur, it would be a fantastic coincidence.
Annas’ version of virtue ethics focusing as it does on the virtuous person faces the Circularity Objection (as she defines it). We explored four possible solutions Annas might offer to avert that objection, but none of them work. They either displace the virtuous person from his preeminent role in virtue ethics or they leave the notion of the virtuous person crucially unexplained or they lead right back to the Circularity Objection. Or, finally, where in one case it escapes the Circularity Objection, it leads to relativism: Fatima takes her route; Zahra moves along hers—and neither is better nor worse (even if they had switched roles). When Susan Wolf averred that “it would not be a moral tragedy if it turned out”Footnote 96 that there was no single, objective unifying principle of morality—perhaps there are as many moralities as there are communities of the virtuous—Derek Parfit responded that “it would be a tragedy if there was no single true morality.”Footnote 97 This paper has tried, in its own small way, to show how certain philosophical assumptions lead to that tragedy.
Acknowledgements:
I am grateful to my institution—Louisiana State University—for granting me sabbatical leave in Spring 2012 and to the College of Humanities and Social Sciences for the Manship Summer 2011 Research Fellowship (Humanities Projects) award which enabled me to write and do research on this paper. I am equally grateful to the two anonymous reviewers of this journal.
