Never pollute
our law with innovations. No, my citizens,
foul a clear well and you will suffer thirst.
Aeschylus 1977: 706-708Footnote 1
“For a wise man,” Edmund Burke declared, in his On American Taxation speech, “he seemed to me at that time, to be governed too much by general maxims.”Footnote 2 Burke was roundly chastising Lord Chatham (William Pitt, the elder), a member of the House of Lords, for being an inveterate subscriber to general rules. He should have been wiser, Burke thought, and learnt to discriminate—as a virtuous person might—where in human affairs general rules are applicable from where they are not. Burke’s denunciation could be read as a pithy expression of virtue ethics. Virtue ethics declares that, while moral rules or maxims are good to have on hand, in the end, they are not always available, useable, or helpful. An act is right, according to virtue ethics, not because it is in conformity with rules or maxims (of utilitarianism or Kant); it is not right because a person invoked, or was guided by, the correct rules or maxims. The act is, ultimately, right because it is an act that proceeded from a virtuous person.
Julia Elizabeth Annas has offered a rich variety of arguments—in remarkably wide-ranging works—that collectively produce a pretty powerful defense of virtue ethics. Here is a working definition of virtue: a virtue is not merely a settled disposition, or even an intrinsically desirable skill; it is a dynamic skill nourished in, or oriented towards, ethical value.Footnote 3 A virtuous person is not merely possessed of necessary virtues; he is also disposed to being and doing a lot more.Footnote 4 Annas takes the virtuous person to play a cardinal role as the supreme, ultimate arbitrator in moral matters.Footnote 5 Quintessential as virtues are, they do not pre-empt the role of the virtuous person because it is not as if once the virtues are in place the person simply follows what the rules of virtues dictate. If that were the case, then, as Rosalind Hursthouse pointed out, there would be little interesting difference between virtue ethics, on the one hand, and a consequentialist or deontological doctrine, on the other.Footnote 6 If we divide moral theories into those that are either act-centered, agent-centered, or principle-centered, then Annas’ theory can be deemed to be agent-centered: acts are to be ultimately explained in terms of the virtuous person. Virtues and virtue-rules are merely dispensable guides.
But, if the virtuous person is assigned to be the supreme, ultimate arbitrator in virtue ethics, then that theory faces a powerful objection, namely, the Circularity Objection. It would be difficult to improve on Annas’ version of it in An Introduction to Plato’s Republic:
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 7
The Circularity Objection, then, is this: we cannot explain a right act in terms of a virtuous person and then explain what makes a virtuous person in terms of right acts.Footnote 8 Annas has not only clearly formulated it; she has also explained what needs to be done to avert it. In this paper, I highlight the role of the virtuous person and how Annas attempts to circumvent the Circularity Objection. To my knowledge, Annas has treated the Circularity Objection with more careful analysis, and offered a variety of fresh perspectives from which to squarely confront it, than has any other virtue ethicist.Footnote 9 She has made it perspicuously clear that this objection is deserving of serious attention since it infiltrates and infects the very core of the doctrine of virtue ethics. No wonder Annas is deeply troubled by this objection and has offered several rebuttals.
Here is the trajectory of this paper. In Section I, I present Annas’ argument that no mature, responsible adult wants to be told what to do, morally speaking; the burden of moral decision must ultimately rest upon the shoulders of each moral agent and cannot be shifted to anyone or anything else (particularly including moral rules). Consequently, wisdom is what gets advertised. In this section I also state the Partiality Principle that lies at the core of Annas’ theory and sketch the Relativism Objection. In Section II, I argue in defense of the Impartiality Principle by detailing a remarkably interesting—and, yet, relatively-unknown—argument in Kant. This principle will serve to undermine the Partiality Principle that is central to Annas’ theory of virtue ethics; at the very least, my aim in this paper is to expose the rival core principles (Kant’s and Annas’) so as to compare and measure the viability of those principles. I should note at the outset that the Impartiality Principle enables us to circumvent both the Circularity Objection and the Relativism Objection while the Partiality Principle serves only to strengthen those objections. In this section, I also draw upon some of the works of Martha C. Nussbaum and G. E. M. Anscombe to underwrite my arguments. In Section III, I analyze Annas’ views on moral teaching and learning, and how, with it, she attempts to avert the Circularity Objection. In Section IV, I explain why Annas deems the ethical character of the moral apprentice indispensable in her theory of virtue ethics and I present two cases to critically evaluate Annas’ ‘developmental account.’ Distinguishing two separate stages—the learner or apprentice stage and the expert stage—it is shown that either the Circularity Objection or the Relativism Objection inevitably re-emerges even from her more nuanced account of virtue ethics. In the brief Section V, I conclude.
I. Don’t Tell Me What To Do
The trouble, as Annas sees it, with any rule-governed moral theory is that it is always telling us what to do when we are faced with a moral problem or a moral dilemma. But, Annas asks, “Do I really want to be told what to do?” (64)Footnote 10 One might interject that a moral theory does a lot more than merely issue blanket moral commandments. A moral theory will tell us not only what we should do, but also why we should do it.Footnote 11 But, even such a theory, Annas thinks, will not do. Why not? “Aren’t we losing an important sense,” she wonders, “in which we should be making our own decision?” (65) The point is not that the why part of the theory is unsatisfactory. No, her point is that, even when we have internalized the satisfactory reasons the theory offers, the final decision, ultimately, must flow from us. “So,” says Annas, “whether the theory is pictured as outside me, like a manual, or inside me, like a set of directions as to how to think, it is still telling me what to do”Footnote 12 (66).
Annas finds such a view overridingly objectionable; such a view leaves out of its picture one cardinal thing, namely, the virtuous person. We, as virtuous persons, are to be praised for doing the right thing; and, we—presumably when we are not virtuous—are at fault when we do something wrong and cannot blame the theory (as we might blame a computer manual for giving us the wrong information). So, from Annas’ point of view, what is missing in this picture is the virtuous person, the agent who is the supreme, ultimate arbitrator in moral matters that pertain to the agent’s own life and actions.Footnote 13
On Annas’ view, even our moral obligations are ultimately dependent on us.Footnote 14 We can espouse a moral theory, consent to be guided by it, acknowledge the reasons it offers for why a particular act, or a species of acts, is our moral obligation; but, in the end, such espousal, consent, or acknowledgement must flow from us.Footnote 15 Just what, exactly, is the force of this? What it says in effect is that we—this inexplicable quiddity each of us is—are the ultimate arbitrators over our morals and our obligations. Since it is we who accept a moral theory and the obligations it declares to be ours, we can—should our wisdom so dictate—abandon that theory and those obligations it has imposed on us. For, if we are not free to renounce a moral theory when we want, as per the dictate of our wisdom, then we will be bound by something external to us—in this instance, a moral theory—that will tell us what to do. And, the burden of Annas’ argument, of course, is that we do not want anyone or anything to tell us what to do.Footnote 16
This poses an obvious threat. If the quiddity of the moral agent, A, prompts him to decide that his moral obligation is X under any given set of circumstances, while the quiddity of the moral agent, B, prompts her to decide that her moral obligation is not-X under that very same set of circumstances, then both A and B are right in their assessments of their obligations.Footnote 17 Let us call this, the Relativism Objection. I shall explore the Relativism Objection more fully in Sections III and IV. For now, let us focus on Annas’ antipathy towards moral rules and the Circularity Objection.
In an earlier, unremittingly anti-Kantian paper, “Personal Love and Kantian Ethics in Effi Briest,” Annas wrote—with the full force of Bernard Williams’ view of why ethics should be first-personal behind her—that Kant’s rule-governed, unconditional ethics is untenable. Annas attempted to show this by drawing upon Theodor Fontane’s famous 1894 novel Effi Briest. For our purposes, the following account of the novel should suffice. Effi Briest is a 17-year-old married to a 38-year-old, successful civil servant, Baron Geert von Innstetten. Annas would have us believe that Innstetten typifies or exemplifies a person enamoured and guided by Kantian ethics.Footnote 18 Innstetten is a man of principles. Briest has committed adultery which Innstetten discovers only six years later through some letters carelessly left undestroyed. In the event, Innstetten (who had once unsuccessfully pursued Briest’s mother), being a man of honour, fights a duel with the man who slept with his wife, kills him, divorces Briest, and teaches his daughter to hate her mother—all in the name of principles. Being rule-driven, Innstetten would not have been able to live with himself, Annas tells us, if he had not done what he, in fact, did.
“Why,” asks Annas, “is Innstetten forced, unwillingly and against his better nature, to cut off all the feelings he has for Effi as his wife, companion, mother of his child?” “We find the answer,” says Annas, “in Kant’s sharp separation of the motive to be moral on the one hand from all inclinations on the other; and this is the core of the defect in Kantian morality which Innstetten’s tragedy shows up.”Footnote 19 Just a little over a decade before the publication of Annas’ paper, Bernard Williams had written:
Some thinkers have been disposed to regard as the essence of morality itself: a principle of impartiality.Footnote 20
As we now know, Williams has ever since been scathingly against the principle of impartiality. Keeping that in mind, we can surmise Annas’ view thus. If Innstetten had not tried to be impartial in the Kantian way, being rule-governed, rule-obsessed, rule-bound, and keeping his inclinations at bay—if he had accorded his inclinations the force and value that were rightfully theirs—then Innstetten would not have done those horrific things. His inclinations would have guided him—as rigid following of rules clearly did not—not only to be more humane in his dealings with others, they would also have fashioned him as he should have been, namely, a good, kind, decent, caring, faithful, responsible friend, father, lover, and husband. Without such inclinations, Innstetten’s life—or anyone’s life—borders on being either immoral or meaningless.
And, so, Annas:
And not only is the unconditional force Kant ascribes to morality not really available to us, we are also all too aware of the dangers implicit in regarding morality as essentially a matter of impartiality.Footnote 21
Several pages later, we have this:
Innstetten is certainly being completely impartial in that he is not smuggling in any factor which favors himself. The reverse is true. He could certainly universalize the maxim that all injured husbands should avenge their wrongs.Footnote 22
And, then, this is how her paper ends:
To live successfully by the Kantian ethic is to risk destroying one’s sources of love and concern for others, and that this not only hurts the others but leaves one’s own life bare and meaningless too. And it also suggests that to reject the Kantian demand by standing by one’s love and care for others is not weakness, but a realistic recognition that there are moral values that are not discerned from the impartial point of view.Footnote 23
Clearly, the stakes are high. Kant and Annas cannot both be right and they disagree on matters of ultimate importance in moral theory.
Let us call the principle implicitly operating in Annas’ argument, the Partiality Principle, to distinguish it from the Impartiality Principle, allegedly objectionable, that infects the Kantian principle (or, so goes Annas’ accusation). Annas’ principle, which dovetails nicely with her principle, ‘Don’t tell me what to do,’ can be stated thus:
The Partiality Principle: A moral principle must make a virtuous person the ultimate arbitrator in moral matters.
Morality, staying closely with Annas, is not a matter of impartiality. Indeed, it is anything but impartial; it is a matter directly, irreducibly, and ultimately connected to a—any—virtuous agent. To live successfully (as Innstetten did not) is to pay homage to one’s inclinations. These inclinations—one’s particular inclinations—are one’s sources of love and concern for others and these give one’s life substance and meaning. Standing by one’s love and care for others is a realistic way of recognizing that there are moral values that can be discerned, but only from the partial point of view of a virtuous agent. It is precisely from this point of view that the agent can rightly determine her moral obligations, what she ought to do, what she should refrain from doing—or anything else that is covered under the rubric of ‘moral matters.’
My concern in this paper is sharply delimited. I am not undertaking to defend Kant against Annas on, for example, the issues of the separation between inclinations and reason, moral motivation, moral judgment, moral values, whether these values are objective or subjective, or how a substantive, purposeful, meaningful life is related to Kantian moral principles.Footnote 24 Perhaps, several papers—if not a book—would be needed to treat these two philosophers adequately on all those interrelated issues. My sole purpose here is to focus on how Annas’ Partiality Principle is related to the Circularity Objection and the Relativism Objection. More specifically, I aim to show that there is a paradox facing Annas’ view of virtue ethics. Either the Partiality Principle is defensible—in which case Annas’ version of virtue ethics can circumvent neither the Circularity Objection nor the Relativism Objection—or the Relativism Objection as well as the Circularity Objection can be circumvented, but only if the Partiality Principle is abandoned.Footnote 25
In sum: Annas affirms the Partiality Principle; and, Kant, clearly, would deny it.
II. The Impartiality Principle
Let us, then, turn to Kant. In The Metaphysics of Morals, Kant claims:
If the I that imposes obligation is taken in the same sense as the I that is put under obligation, a duty to oneself is a self-contradictory concept. For the concept of duty contains the concept of being passively constrained (I am bound). But if the duty is a duty to myself, I think of myself as binding and so as actively constraining (I, the same subject, am imposing obligation). And the proposition that asserts a duty to myself (I ought to bind myself) would involve being bound to bind myself (a passive obligation that was still, in the same sense of the relation, also an active obligation), and hence a contradiction.—One can also bring this contradiction to light by pointing out that the one imposing an obligation (auctor obligationis) could always release the one put under obligation (subjectum obligationis) from the obligation (terminus obligationis), so that (if both are one and the same subject) he would not be bound at all to a duty he lays upon himself. This involves a contradiction.Footnote 26
This rare, significant passage is not exactly perspicuous; so, let us parse it. Kant is twice attempting, in this passage, to demonstrate that there is a contradiction in the notion that an agent can place himself under a moral obligation. In the first attempt, in the first half of this passage, Kant relies on a distinction between passive obligation and active obligation.Footnote 27 Kant claims that, when one is under an obligation, a duty, one is under a constraint, one is passive under the demand of duty; one is, as Kant puts it, bound. Here, the obligation of the I is a passive one. But, if I am to put myself under an obligation, be bound, then that I (my I) is simultaneously doing the binding. Here, then, the obligation of the I is an active one.Footnote 28 Kant contends that it is a contradiction to suppose that the I that binds (the I that is engaged in active obligation) is also the I that is bound (the I that is subjected in passive obligation). This is what Kant contends because he is assuming that nothing can, at the same time, possess two contradictory properties; and, that the I that is attempting to bind itself is attempting to possess two contradictory properties: the property of passive obligation and that of active obligation. Generalizing, Kant says that what is bound cannot also simultaneously bind and what binds cannot also simultaneously be bound: that, according to Kant, is contradictory.
Kant is saying: if the I that is imposing an obligation (let us call it, the ‘obligation-imposing I’) is identically the same I as the I on whom the obligation is imposed (let us call it, the ‘obligation-recipient I’), and the obligation-imposing I is not bound by any I-independent moral law that constrains that obligation-imposing I, then that obligation-imposing I could just as easily or readily annul that obligation; there is nothing to hinder it from doing so. (Kant rightly insists that in viewing this problem we must look at the individual as “a dual personality,” “a doubled self.”Footnote 29) The obligation—any genuine obligation is so of necessity—imposed is only an alleged obligation for it is “in itself contingent”; and the I can “give up” anything that is “in itself contingent.”Footnote 30 It is the wholly discretionary will of the obligation-imposing I (unconstrained by any I-independent moral law) which gives that obligation its force—a force which it can just as easily destroy as bring into existence. In short, the obligation is contingent because it is entirely at the obligation-imposing I’s arbitrary discretion (and, consequently, is not a genuine obligation).
Continuing to assume, as does Kant, that the obligation-imposing I is identical to the obligation-recipient I, consider Kant’s second attempt, in the second half of the foregoing passage: the obligation-imposing I says to the obligation-recipient I, ‘I impose upon you obligation X.’ Then, the obligation-recipient I says to the obligation-imposing I, ‘I annul the obligation X you impose upon me.’Footnote 31 There are no constraints on the obligation-imposing I as to what duties or obligations it can impose; likewise, there are no constraints on the obligation-recipient I as to what duties or obligations it can annul. And, so, we can see why Kant unassailably contends, “that (if both [the obligation-imposing I and the obligation-recipient I] are one and the same subject) he would not be bound at all to a duty he lays upon himself.”Footnote 32 Contrariwise, if there were an I-independent moral law, the obligation-imposing I would no more have the liberty to impose an obligation that did not issue from an I-independent moral law than the obligation-recipient I would have the liberty to annul an obligation if it did issue from an I-independent moral law.
Let us call this Kantian thesis, namely, that there is something objective, or I-independent, to tell the moral agent what she should, or should not, do (what her obligations are), the Impartiality Principle. It can be stated thus:
The Impartiality Principle: A moral principle must make an agent-independent law the ultimate arbitrator in moral matters.
It is impartial—Kant’s Categorical Imperative is plainly an instance of it—because it is a principle that applies unreservedly, uniformly, and universally to every rational, moral agent.Footnote 33 But, Kant and Annas cannot both be right. Kant maintains (in view of the Impartiality Principle) that obligations cannot be dependent on the moral agent (he can no more create obligations than he can destroy them). Annas maintains (in view of the Partiality Principle) that the virtuous person, or his wisdom, is the ultimate source of that agent’s moral obligations (he can create obligations as well as he can destroy them); in other words, it is the virtuous person who decides what his obligations or duties are, or what he should specifically do under given circumstances.
In sum: Kant affirms the Impartiality Principle, and, Annas, clearly, denies it.
How, then, should the virtuous person proceed, as Annas views things? There are only three alternatives. (1) The virtuous person appeals to reason; but these reasons cannot be of the universal type because then they would be indistinguishable from a rule-governed moral theory and it will be the theory that will tell the virtuous person what to do. Rather, when, in his particular circumstances, he is facing a particular moral problem, the virtuous person who heeds his particular natural feelings, emotions, and sentiments will better discover what he ought to do than if he were to rely on cold, impartial rules.Footnote 34 It is precisely because Innstetten is obsessed with rules, objectivity, impartiality, and their ilk—and his natural feelings, emotions, and sentiments are suppressed, distorted, or ill-informed—that he is led to such a callous life and such inhumane behaviour.
(2) The virtuous person appeals to individual-specific reasons; but, these individual-specific reasons will not do either since, at the very least, such reasons will tell the individual, albeit only that individual, what he should do in similar circumstances. Furthermore, such individual-specific reasons can be cobbled into an individual-specific moral theory and, once again, it will be the individual-specific moral theory that will tell the moral agent what to do. Note: in (2), as in (1), the virtuous person has internalized the reasons of the theory in question. But, as we saw earlier, Annas roundly condemns the internalization of a moral theory no less than the external, agent-independent moral rules. Not only has Innstetten accepted objective, external rules (or rules sponsored by individual-specific reasons), he has internalized them to such a degree that his naturally good, warm, and humane sentiments and feelings are left languishing.
To underscore my arguments against alternatives (1) and (2), let me invoke Nussbaum’s response to Annas’ “Personal Love and Kantian Ethics in Effi Briest.”Footnote 35 Nussbaum argues that Innstetten’s failure shows that there is something seriously wrong in bringing up people by rules alone, devoid of the accompanying love, imagination, and compassion.Footnote 36 So far, Annas and Nussbaum are aligned. However, “Imagination is of no use without a moral code of some sort,” says Nussbaum. “But it is also true that compassion guides us truly toward something that lies at the core of morality, and without which any moral judgment is a ghastly simulacrum. … In short: compassion does not supply a complete morality; far from it. But there is reason to trust it as a guide to something that is at the very heart of morality.”Footnote 37
First, the notion of compassion Nussbaum has so lauded here is neither irrational nor bad (as being predicated upon false beliefs)—as the critics of compassion (the Stoics, Seneca, Adam Smith (with minor qualification), Spinoza, and Nietzsche) would have us believe.Footnote 38 At its core, compassion is cognitive, rational, and good.Footnote 39 This is not to say that the invocation of compassion is infallible. Mistakes can be made in assessing whether a situation warrants compassion: “But the pro-compassion tradition is not prevented from judging that some occasions for compassion are illegitimate, and based upon false evaluations.”Footnote 40 Consequently, it is important to get compassion right; and, so, “compassion takes up not the actual point of view of any and every sufferer, but rather the point of view of an onlooker who appraises the seriousness of what has happened. The normative suitability of this emotion, as of fear and grief and anger, depends on whether the person gets the appraisal right, using a defensible theory of value.”Footnote 41 What I want to underscore is the objectivity of the assessment (getting the appraisal right) of the suitability of even compassion, let alone right action.Footnote 42
Second, while no moral judgment, in principle, can be made without the moral codes or moral rules, compassion will be needed to make the moral judgment in the manner it should be made—“even in a fully just society we need compassionate judges and juries.”Footnote 43 Let us refer to the objective ethical theory, and the moral codes or moral rules dependent or derived from that theory, Nussbaum prefers as M (this may be, as I believe it is, neo-Kantian, but it need not be). Innstetten needs not only compassion; he will also need M, if he is to make proper moral judgments. Only M and compassion will conjointly enable Innstetten to render a proper judgment about what he should do with his wife, his wife’s lover, and his daughter. Here is the point to focus upon: even if Annas has no difficulty embracing Nussbaum’s concerns about compassion,Footnote 44 she could not so readily embrace the insistence upon M, the moral theory. M (in conjunction with compassion) will tell Innstetten what he should do; but, Annas does not want a virtuous person to be told by a moral theory, moral codes or moral rules, what she should or shouldn’t do.
Third, for all of Nussbaum’s emphasis on compassion—and rightly so—her version of virtue ethics is markedly distinct from Annas’ version: Nussbaum’s version is avowedly non-relativistic.Footnote 45 Compassion is a cognitive emotion—something, says Nussbaum, Kant got wrong.Footnote 46 Let us suppose that Kant is wrong; does that help Annas? Compassion is not some non-rational psychological state devoid of any cognitive content. If it has cognitive content, it has a truth-value. If it has a truth-value, it cannot be wholly dependent on the virtuous person (even if he is a fully virtuous person). But, if compassion that has cognitive content is essential in rendering a moral judgment, or determining a moral obligation, that content can hardly vary from virtuous person to virtuous person (allowing for obvious and reasonable variations). In what way, then, is a virtuous person the ultimate arbitrator? How could the Partiality Principle be defended? Innstetten’s wrongful moral judgments can be translated in terms of his failing to have the true or right cognitive content and his wrongful acts can be translated in terms of transgressing the moral codes or rules enshrined in M.
Fourth, unlike Nussbaum, Annas thinks that Kant’s moral principle or, indeed, any rule-governed principle is a veritable obstruction to leading a good or meaningful life or performing a good or moral deed: witness Innstetten. By contrast, Nussbaum insists on a moral code—and by implication the Impartiality Principle—as a necessary condition (if not a sufficient one), not only for sound moral judgment and good moral deed, but also for the making of a good or just society. “We should not attempt to produce a good society through the motive of compassion alone, since it is only within the limits of reason, so to speak, that compassion proves worthwhile rather than quirky and unreliable.”Footnote 47 Thus, so deep are the connections of compassion and its ilk to reason and objectivity that they make Nussbaum’s version of virtue ethics immune not only to the Relativism Objection but also to the Circularity Objection in a way in which Annas’ version is not.Footnote 48
Finally, the third and last alternative as to how a virtuous person should proceed, as Annas views things: (3) the agent is a quintessentially virtuous person. Such a virtuous person, when immersed in a moral problem, is so keenly attuned and sensitive to the several factors present in the problem situation that whatever emerges from a deep reflection on it—reflection in a very broad sense (not only including reasons, but things other than reasons)—constitutes the answer to the moral problem. Such a virtuous person would know what to do under the circumstances. Had Innstetten been a virtuous person, and not enslaved to rules, he would not have led a hollow life, nor acted dastardly. What emerges, please note, is not reducible to reason, feeling, or sentiment (or, even all three); it is something sui generis—something that bears the unique signature of that moral agent. Perhaps, on Annas’ view, this is just what wisdom is.Footnote 49
But, then, there is the point about the content of what one has decided, aside from self-legislation. Here’s how Anscombe expressed the point: “That legislation can be ‘for oneself’ I reject as absurd; whatever you do ‘for yourself’ may be admirable; but is not legislating. Once one sees this, one may say: I have to frame my own rules, and these are the best ones I can frame, and I shall go by them until I know something better: as a man might say, ‘I shall go by the custom of my ancestors.’ Whether this leads to good or evil will depend on the content of the rules or of the customs of one’s ancestors.”Footnote 50 Thus, even if one can legislate for oneself (and only for oneself)—a view Anscombe, like Kant, rejects—there is the further question of whether the content of what you hold is good or evil;Footnote 51 this, too, Anscombe contends, does not depend on the virtuous person.
III. Teaching and Learning Virtue: The Developmental Account
Might one, then, simply take the virtuous person as the starting point for philosophical inquiry, especially to determine what constitutes a right action and let the moral codes or rules fall where they may? To such a proposal, Annas crisply anticipates the Circularity Objection:
What I take to be the common view holds that the virtuous person must be identified independently of their performance of right actions. Otherwise, we would not have an account of right action which was explanatory. If we define right action as what the virtuous person would do, but it turns out that the virtuous person is even in part defined in turn by the doing of right action, the claim goes, we have a circle, and so no explanation. (67)
This objection, as Annas reports, is itself susceptible to several objections. First, how is a virtuous person to be identified (without invoking the moral codes or rules)?Footnote 52 (The invocation of such rules would undermine the supremacy of the notion of a virtuous agent.) Merely pointing to examples restricts us by our cultural boundaries. Second, what if there are no actual virtuous persons around for us to point to?Footnote 53 If we are restricted to giving an account by referring to ideals, will not our account suffer from vagueness? Third, invoking the idea of a virtuous person—even a fully virtuous one—is of no use since that person would not have been in the moral quandary someone else is in, and thus cannot advise what the latter should do. Fourth, and finally, Annas offers the cardinal objection that what is wrong with the suggestion that we ultimately appeal to the virtuous person—or a fully virtuous one—is that we are relying upon a moral expert. (For one thing, if the expert himself appeals to a moral manual, or moral rules, we would be back to where we started.) “But even if the expert is supposed not to use a manual herself, but to have an understanding of right and wrong action which cannot be codified in a manual, we are still, on this model, using the expert to tell us what to do, in exactly the way we used the manual: namely, tell us what to do” (68). And the last thing we need, in Annas’ view, is for someone to tell us what to do.
But, even granting these four objections (which I will track in what follows), both the Circularity Objection and the Relativism Objection would still remain. Here’s why. First, we are not permitted to invoke rules or manuals; second, there are no actual virtuous persons; and, third, any appeal to fully virtuous persons is ruled out.Footnote 54 We are, therefore, left with particular, fallible, virtuous persons. Let one such person be X. If an act of type Z is morally right because it is performed by X, in circumstance C, and X is virtuous because she typically performs actions of type Z, in circumstances of type C, then this is simply a straightforward instance of the Circularity Objection. Let another such virtuous person be Y. If an act of type non-Z is morally right because it is performed by Y, in circumstance C, and Y is virtuous because he typically performs actions of type non-Z, in circumstances of type C, then act of type Z is morally right and act of type non-Z is morally right. This is simply a straightforward instance of the Relativism Objection.
What, then, is the alternative? How can the Circularity Objection or the Relativism Objection be escaped? Annas offers:
A developmental account, in which we give an account of being virtuous and doing the right thing in a way which involves a developmental process. If so, we would have a theory involving not two items but three: the virtuous person, right action and the relevant developmental process.Footnote 55 (68)
Let us begin by explicating Annas’ meticulously-argued developmental account; we can evaluate it at the end. Learning to be virtuous is like learning to be a builder, she says, following Aristotle; being practical skills is what the two types of learning have in common.Footnote 56 One who begins to learn how to be virtuous must pick a role model, and mimick the role model as closely as possible; presumably, the role model would correct the error of the learner’s ways when the latter makes mistakes. At this beginning stage, he is more dependent on the role model, less on himself, lacking resources and experiences as he does. Gradually, he becomes more skillful at being virtuous, less dependent on his role model, and “expresses more understanding of his own.”Footnote 57 (69) What does this involve? “He progresses from piecemeal and derivative understanding of building to a more unified and explanatory understanding of his own. His actions may at this point differ from those of his role model precisely because he is a better builder. This is because he is learning, and learning contains the notion of aspiration to improve” (69). With this Annas arrives at the crucial point in her argument where she delineates the fundamental difference between a rule-governed morality, which is irredeemably defective, in her estimation, and her own view of virtue ethics:
For instead of trying to force our everyday moral thoughts into a system of a one-size-fits-all kind, virtue ethics tells us to look elsewhere—at what happens when we try to become a builder or a pianist. The moral beliefs we have taken over from others are just the beginning stage. It is up to us to put in the work needed to develop into someone who has more understanding. (70)
Let us say that the role model—the moral teacher, the moral expert, the virtuous person—possesses more understanding of his own (non-fragmentary, unified, and explanatory) as will the apprentice in due course.Footnote 58 But, we must be extremely cautious. We are not to suggest that the moral apprentice is like an apprentice in physics so that in due course what he will understand will eventually match or converge with what his teacher, expert, or model knows: when each (the teacher and the apprentice) understands Newton’s physics or Maxwell’s theory of electricity and magnetism, both of them understand the same thing. This is nothing if not inaccurate as a description of Annas’ position.Footnote 59
What, then? Are we confined to saying that in ethics to each his own way of expressing his own moral understanding as we might say that Haydn’s Variations for Piano in F Minor (1793) has a variety of interpreters (such as, Alfred Brendel, Margaret Fangman, and Mikhail Pletnev)? We can then have as many unique forms of understanding Haydn’s Variations as there are major interpreters; and, if each major interpreter has her own apprentices, then these, too, in time will develop and express their own distinctive understanding of Haydn’s famous piece.Footnote 60 And so with morality: each role model, the moral teacher, will give advice which expresses his own understanding of morality in that particular case; and we shall have as many unique forms of expression as there are role models and apprentices who, in time, will become role models in turn, each expressing his own unique understanding. (Nor will the builder analogy suggest anything different. Builders among the Eskimos, Australian aborigines, residents of rain forests, or inhabitants of the hinterlands of Bangladesh and Zambia are unlikely to have “a [building] system of a one-size-fits-all kind.” Yet, assuredly, within each of these communities there will be a group of builders composed of experts and apprentices.) If Annas is right, we too must track the ethical systems of various communities in a similar fashion.
Annas’ analogy assumes that there is, at least, one thing upon which everyone is agreed: namely, it is Haydn’s Variations for Piano in F Minor which is the subject of different renditions or interpretations. This cannot quite track the moral analogy we want: for there it is not as if there is one rule or one unique set of rules—using the word ‘rule’ very broadly here—to which different virtuous persons (role models) are proposing different interpretations; for example, it is not as if there is only one moral law, namely, the Categorical Imperative or the utilitarian rule of maximizing happiness, and there is merely a disagreement whether we should be emphasizing the Formula of Universal Law or Formula of Humanity as an End in Itself in the first case or, in the second case, we merely disagree over whether we should be an act utilitarian rather than a rule utilitarian. In any event, such an analogy would fly in the face of Annas’ considered view. It is rather that different virtuous persons are coming up with different (indeed, conflicting) rules: so that one virtuous person advocates that lying is permissible in one context and gives his reasons for it; another virtuous person insists that justice, in the very same context, is required and gives other reasons for her stand on the matter; a third virtuous person, in that context, demands maximization of welfare; and so on.
Assuming ethical propositions have a truth-value, one can say that in morality one moral expert, teacher, or model can give advice (expressing his understanding) that is contradictory to the advice given by another moral expert, teacher, or model (expressing her understanding); but, at the very least, it is not obvious that different interpretations of a musical piece are mutually contradictory (in the strict logical sense).Footnote 61 The unsettling conclusion one reaches from this paragraph and the two before it is obvious: the Relativism Objection is inescapable.
And, yet, Annas gives the indelible impression that in her story relativism of the kind feared has little or no role to play.Footnote 62 She gives that impression because she talks of “the fully virtuous person,” “the truly brave person,” “the completely virtuous brave person,” and “the ideally virtuous person”Footnote 63 (70-74). This would lead her readers to believe that it is false that there are as many ideals of a virtuous person as there are moral experts, teachers, or role models, but rather that there is one single objective ideal towards which even sundry experts, teachers, or models are aspiring. Such an idea of the ideal Kant would assuredly have applauded.Footnote 64 But, it is anything but obvious, given her theory, what resources Annas has left to defend the idea of the ideally virtuous person. And even if such an ideal were to exist—never mind what criterion would define that ideal for now—it would simply be a surrogate or a substitute for the moral expert and we know that she has no room for moral experts in her theory. Nor, of course, does Annas have a way of deciding which of the different and differing moral experts to rely upon—unless, of course, the Relativism Objection is circumvented (in which case she will have to abandon the Partiality Principle). More, we will need to ask: in virtue of what is the fully virtuous person virtuous? The answer, of course, cannot be in terms of moral rules. If the answer is—I adapt what follows from Annas (67)—the fully virtuous person is one who does the right action, and the right action is that which a fully virtuous person would perform, we will be confronted anew by the Circularity Objection; and, we shall not have the explanation Annas is so understandably seeking.
Again, Annas writes of how “virtue ethics tries to improve our understanding” (70) and one might be forgiven if one reads “improve” as a hint at something objective in our understanding. I did not quite understand modern art, American Indian history, or differential calculus in high-school as well as I did in college (and there is still room for improvement). My understanding of those subjects improved; it improved in the direction of the ideal understanding of the subject. Isn’t it likewise when we are trying to improve our understanding of ethics? If there is such a thing as ‘the truly brave person’ or ‘the ideally virtuous person’—for example, as Annas acknowledges—then there is something objective that I am trying to understand when I am trying to understand what it is to be ideally just, brave, or virtuous.
Let us try an entirely different path. Suppose that Annas’ view is more akin to the view of the Stoics:
The Stoics hold this openly, indeed blatantly, claiming that the ideal wise person is the only person who really sees and hears properly, knows the truth, can run a household, be a king, be rich, be handsome and so on and so on. As they put it in consciously paradoxical ways, the ideal is the norm. We cannot, then, find out about human nature just by observing people. It is a theoretical concept, and we need the right theory to get it right. And similarly with virtue. Virtue is displayed in its proper form only in the completely virtuous person, but since this is an ideal we have no direct empirical access to this. We have two theoretical concepts, and it is unsurprising that we have to work at and with both of them together.Footnote 65
The ideal person—being an ideal—is nowhere to be found in the actual world; the rest of us are merely imperfect copies of that ideal.Footnote 66 But, only such a person constitutes a norm, not just for some human beings, in a particular society, at a particular moment in their history, but for all human beings, down to and including “the remotest Mysian.”Footnote 67 In the Stoic ethical account, nature and virtue (the two theoretical concepts) jointly play the lead role. One cannot get the right idea about nature without having developed some virtues; and, one cannot develop proper virtue without the right idea of nature. In short, virtue cannot be defined independently of nature nor can the latter be defined independently of virtue.Footnote 68 If and when the two notions—nature and virtue—are correctly defined, one will emerge with exactly one ideal, one norm. Consequently, there is no threat of relativism.
Notice, too, that the developmental account is also found in its incipient form in Stoic ethics, an account not only about nature and virtue, but about reason as well. Thus, “the Stoics see the development of reason which leads to adopting the moral point of view as a natural part of the development of human nature.”Footnote 69 So, reason, whole and complete, is not something we possess at the start; then we possess only the reason-embryo. Reason blossoms as the agent becomes more and more virtuous. The trinity of virtue, understanding of nature, and reason move in lock-step. But, at the end of it all—should one reach that coveted end—the virtuous person will arrive at the ideal, not just any old ideal upon which he, the fully virtuous person, has left his indelible signature.
It is true that each pair of teacher and apprentice may proceed at their own pace—there is differential teaching, learning, and development: the apprentice better understands what nature is when he becomes more virtuous and better understands what it is to be virtuous the more he understands nature. This is no less true of the teacher. Each teacher-apprentice pair may reach the ideal in approximate ways—one pair, perhaps, may be further away from the ideal than others. But, there is no threat that each pair will find or arrive at different ideals by traversing different paths. The ideal is the norm, not several ideals. Clearly, then, it is plausible to speak of “the truly brave person,” as Annas does.Footnote 70 There is no threat of relativism.
This is entirely plausible—after all, there are a fair number of similarities between Kant and the Stoics.Footnote 71 But, the question is whether Annas can subscribe to it. Let me put this in terms of a dilemma. Either there is one, single, ideal, ethical norm or there aren’t any. If the second disjunct is true, then there is no way of escaping from the Relativism Objection canvassed earlier: you follow one path, I another; your understanding improves in the direction of your path, mine along my path; and, finally, your path will instruct you as to what it is to be ‘the truly brave person,’ my path will do likewise, but differently, for me. Ultimately, no one path is better than another (perhaps, not even for one single agent, who can haply change his path, given the Partiality Principle).
If the first disjunct—say, the Stoic view—is true, then we do escape from relativism, but such a doctrine dovetails rather poorly with Annas’ central claims. For example, first, if there is a single, ideal norm, exemplified in the fully virtuous person, then it is that ideal—something which will function like a manual—that will tell us what to do; but, Annas cannot abide such a view. Second, if there is the ideal norm, it will displace the virtuous person from the position of eminence in Annas’ theory. This is because the ideal norm will occupy that position; and, that is the norm that instantiates the Impartiality Principle while entailing the rejection of the Partiality Principle. At best, the virtuous person is important epistemically, a vehicle through which the lesser virtuous person will have access to the ideal. Third, the virtuous person has a better chance of learning and knowing what that ideal is and then teaching it to the rest of us, and we for our part, as we become more virtuous, will understand that ideal better in due course. But, that ideal norm is an objective norm; it is not of our making—there is nothing present of the ubiquitous individual in it—even if we have no access to that ideal without being virtuous.
IV. The Ethical Nature of the Moral Apprentice
Consider two rather simple cases:
Case 1: There are two individuals, Sarah and Joseph, in a certain society. Sarah is the acknowledged moral expert, the fully virtuous personFootnote 72 and Joseph is the self-professed moral apprentice.Footnote 73 Joseph is conscientious and has a virtuous disposition and, when he is confronted by a moral problem, he is a careful scrutinizer of his own heartFootnote 74 to make sure that he does not act from shabby or morally-tainted motives. But, Joseph is young and inexperienced, and sometimes errs, meanders, miscalculates; Sarah intervenes and assiduously explains what he did wrong and why it was wrong. Slowly, but surely, Joseph catches on. He learns to make his own decisions even more carefully than before, having imbibed a variety of moral principles (with provisos and all); in short, he has slowly learnt the art of acting skillfully in a morally right manner. In time, he, too, becomes a fully autonomous moral agent—a fully virtuous person—with his own understanding, expressing that understanding in his right actions. Let us make the assumption (favourable to Annas) that subsequently there is very little divergence in the decisions of Joseph and Sarah, even if there is not complete convergence. The “fully virtuous person is the ideal that the beginner in virtue is aspiring to be” (71). Joseph, the moral apprentice, aspired to be like Sarah when she was his moral guide; he has now approximated this ideal and can dispense with Sarah.
Let us distinguish two stages and we will treat them in successive order: (a) the learner or apprentice stage and (b) the expert stage. Our task is to determine whether the Circularity Objection or the Relativism Objection is escaped in either stage.
(a) The Learner or Apprentice Stage. This example captures the three items in Annas’ account: the virtuous person (Sarah, the moral expert), right action (unspecified), and the relevant developmental process (the process Joseph, the moral apprentice, undergoes under Sarah’s able moral guidance).
In this learner stage, then, when Joseph falters, Sarah comes to the rescue, aids and corrects, whereupon Joseph does the right thing. Should not the rightness of Joseph’s act be ultimately traced to Sarah’s virtuousness—for what else is there? Consequently, if we are to determine what makes for Joseph’s right action, our sole focus should be on Sarah’s virtuousness. (Ex hypothesi, Sarah is the moral expert and so the element of development in her case can be safely set aside. If Sarah’s moral expertise is itself in question, then, of course, Annas herself has no way to proceed further.) But, if Sarah’s virtuousness is the focus, then Annas can no longer show us a way out of the Circularity Objection. In order to measure the moral worth of Joseph’s actions, we must determine Sarah’s standing as a virtuous person; but, we have no way to measure the moral worth of Sarah’s standing as a virtuous person without measuring her actions (given the Partiality Principle, we cannot measure them by appealing to some objective moral rules); and, conversely, there is no way of measuring Sarah’s actions without referring to her standing as a virtuous person. The Circularity Objection is fully restored.
(b) The Expert Stage. Perhaps, leaving Joseph out of the picture distorts the essentials of Annas’ theory. It is true that Joseph needs moral development, not Sarah; but, once Joseph’s moral development is complete and he is his own autonomous person, then the rightness of his subsequent actions is determined by his being a virtuous person. Then, at last, Annas might claim, we do “have an account of right action which was explanatory” (67). So, let us bring Joseph back into the picture. Say that Joseph has performed an action and it is claimed that the action is right; we ask for an explanation: why is his action right? There is no other element in the situation to which we can appeal to explain the rightness of the act other than Joseph himself; or, more specifically, Joseph’s moral understanding.
We have two alternatives. (1) It is entirely legitimate to ask why we should rely upon Joseph’s moral understanding to determine the rightness of an action just as it would be legitimate to ask the credentials of a builder if we are asked to rely upon the builder’s understanding to build something. The etiology of Joseph’s moral understanding would be quite pertinent, especially, in view of Annas’ developmental account, but this would lead us back to the role model or the virtuous person in the learner or apprentice stage. Or, (2) we can simply take Joseph’s moral understanding as the starting point in the explanation of the rightness of the action; but, then it would be a starting point that can offer no defense or explanation for what makes Joseph’s action right. Clearly, both (1) and (2) are indefensible. In (1), we are led back to the Circularity Objection. In (2), we are led back to the Relativism Objection, by now the all-too-familiar route.
Case 2: There are no moral experts in a certain society; in fact, it is a rather nasty society. By an unknown psychological process, Joseph has come to acknowledge his moral deficiency and vows to uplift himself. There is no one—no Sarah—for him to rely upon; he is to rise by his own bootstraps. Joseph learns to be conscientious—although he has conscience which, of course, cannot be acquired;Footnote 75 he learns to scrutinize his own heart to ensure that he is not engaged in self-deception. Now, Joseph is able to overcome the usual obstacles posed by youth and inexperience, and slowly learns to avoid making the moral mistakes he had made before. He learns to make his moral decisions even more carefully than before, having taught himself a variety of moral principles (with provisos and all) and learnt the difficult art of skillfully performing right actions. In time, he becomes a fully autonomous virtuous person with his own understanding, expressing that understanding in his actions.
There is moral development, in this case, even though there are no moral experts. Have we, at last, escaped the Circularity Objection or the Relativism Objection? Let us ask: how is Joseph’s moral development to be understood? One can measure the degree of Joseph’s virtue or virtuousness at each stage in his development against an objective moral standard but that would presuppose that such a standard is available to measure the actions of a person independently of virtuous persons. This is hardly a presupposition consistent with Annas’ theory of virtue ethics committed as it is to the Partiality Principle. Well, there are no virtuous persons in this example; that is its point: to show that “reason commands how men are to act even though no example of this could be found.”Footnote 76 Then, the only alternative that is left is this: one can measure the degree of Joseph’s virtue at each stage in his development against the degree of the rightness of his actions; and, one can measure the degree of the rightness of his actions in terms of the degree of Joseph’s virtue at each stage in his development. But, this argument too obviously succumbs to the Circularity Objection.Footnote 77 And, if we had someone other than Joseph—say, Paul—who was raised differently and acted differently, then we could show, as before, how this argument is also susceptible to the Relativism Objection.
V. Conclusion
Virtue ethics, and in particular Annas’ version of it, is a pretty powerful theory: it has uncovered a host of moral phenomena or described the moral phenomenology more accurately, or in more exquisite detail, and offered an array of arguments that purport to support and explain those phenomena than have many other moral theories. Undoubtedly, then, many of virtue ethics’ insights and arguments must eventually be absorbed into an adequate moral theory. Kant’s moral theory is hardly faultless; but, it offers more hope to serve as an adequate fundamental moral principle than does Annas’ theory of virtue ethics. Annas’ theory assigns to the virtuous person the primary role of the ultimate arbitrator in moral matters (as Kant does not). Consequently, her multi-faceted theory, including the Partiality Principle and the developmental account, cannot escape the following consequence: it is susceptible either to the Circularity Objection or the Relativism Objection. Or both.
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 three anonymous reviewers of this journal.
