From the Nicomachean Ethics to the Grant Study

[Here as promised is a first draft of the paper I’ll be giving this Saturday at the annual conference of the Association for Core Texts and Courses in Plymouth Harbor, Massachusetts. Papers for the conference are supposed to be short, non-technical treatments of a core text or two appropriate for undergraduate teaching, along with a rationale for teaching them. This year’s theme is the relation between the arts and sciences in undergraduate education. Comments are welcome, though I probably won’t see them until next week. I’ll add hyperlinks next week as well. This discussion was quite helpful to me in thinking things through.]

From the Nicomachean Ethics to the Grant Study: Ancient Greek Ethics Meets Modern Psychiatry

Irfan Khawaja
Felician College
Association for Core Texts and Courses
Plymouth Harbor, Mass. (April 11, 2015)

  1. Introduction

George Vaillant’s Adaptation to Life is a classic of contemporary behavioral science; Aristotle’s Nicomachean Ethics is one of the founding texts of ancient Greek moral philosophy.[1] Both texts implicitly address the same topic, but each does so in ways that fundamentally contradict the claims of the other. Given this, it’s a useful (and entirely Aristotelian) exercise to read the two books in tandem, using the one to challenge and correct the claims of its rival. The resulting inquiry leaves us with a better sense of the strengths and weaknesses of both behavioral science and moral philosophy, and leaves us with some difficult questions as well.

  1. Some similarities

Aristotle’s Nicomachean Ethics is a study of the nature of eudaimonia, or human flourishing, and the role played, primarily by arête (virtue) and secondarily by tuche (luck) in the realization, across a human lifespan, of the highest forms of flourishing. Starting with the assumption that there is some fundamental aim that we all seek, that the aim is eudaimonia, that arête is integral to it, and that tuche is at least relevant to it, Aristotle adopts what is typically called a dialectical approach to his inquiry: he lays out common beliefs about the relevant phenomena, raises some puzzles about them, and sketches a coherent and systematic account of them that resolves as many of the puzzles as possible.

If we had to reduce Aristotle’s thesis to a few sentences, it might be this: Virtue is integral to human flourishing, as, in a different way, is luck. Virtue depends on (good) luck in the sense that virtue requires relatively favorable conditions, especially socio-political conditions, for its realization; Aristotle insists in the Politics that it depends in fact on (an idealized version of) the Greek polis, but the more general point is that it depends on some relatively specific set of socio-political (and even geographic) conditions, whatever those turn out to be. Given relatively favorable background conditions, however, virtue transcends luck, as well; virtue allows us to rise above the disfiguring effects of misfortune and to regulate the potentially corrupting effects of good fortune and so, to achieve our flourishing despite them. As far as flourishing is concerned, then, virtue is the most causally fundamental (in Aristotle’s sense of “causal fundamentality”) of the variables that produce flourishing: barring highly exceptional circumstances, the virtuous person is most apt to flourish; the vicious person has no hope of doing so. Moreover, for the most part (though the exceptions to and qualifications of the rule are themselves notable), the choice between virtue and vice is one that is essentially up to the agent.

Aristotle goes a bit further. There is on his view, a hierarchy in forms of life, considered as definitely structured ways of living one’s life, where some fundamental aim dictates the traits of character one inculcates (and/or has inculcated in one). Each of these lives has different implications for the practice of virtue; in a sense each dictates the practice of slightly different virtues, or different conceptions of virtue. The very best such life is the life of philosophical contemplation. The second-best is the practical life of political activity in the pursuit of justice. Both forms of life are virtuous, and both produce human success, but the very best life presupposes native aptitudes possessed only by a very small minority of people; such people come as close as humans do to the blessed status of divinity, achieving (or approximating) a kind of timeless immortality in their highest and most strenuous acts of contemplation. Most men, lacking the aptitude for sustained philosophical activity, are forced at best to dabble in contemplation but devote their life’s task to the pursuit of justice, achieving nobility but not the quasi-immortality afforded the philosopher. Finally, it’s essential to Aristotle’s view that both forms of life are, in different ways, social. The philosopher’s life requires extended periods of solitude, the activist’s life is more socio-politically engaged, but neither can function outside of the context of the polis and the social life it enables.

George Vaillant’s Adaptation to Life is one of three book-length reports on the Grant Study of Adult Development, conceived in 1937 as an attempt to produce a naturalistic psychiatric study of what makes for psychological health in psychologically healthy people.[2] The study chose 268 specially selected Harvard sophomores from the classes of 1939, 1941, 1942, and 1944, and resolved to follow them essentially across their full lifespans, recording their life circumstances, and the success or failure of their adaptations to them. Adaptation is the early-to-mid-life discussion of a 95-person subsample of the original cohort. The book is a longitudinal study conducted essentially by the methods of interview and questionnaire; after administering a standard battery of initial medical and psychological tests, the researchers followed each participant over time, conducting the equivalent of a series of psychoanalytic in-take sessions across several decades. That said, the book proceeds on an eclectic mixture of Freudian assumptions about psychodynamics, behaviorist assumptions about method, relativist assumptions about morality and broadly (though sometimes narrowly) liberal assumptions about politics.

If we had to reduce Vaillant’s thesis to a few sentences, it might be this: successful adaptation to life is a function of two basic variables, (a) which ego defense mechanisms a person adopts, and (b) the kind of relationships to other people that he has. An ego defense mechanism is an unconscious psychological mechanism for handling unwanted or unpleasant facts, whether about oneself or one’s environment. While all ego defense mechanisms have the same essential function, various categories of defense mechanism differ in kind and manifestation from others, forming a hierarchy that ranges from “mature” to “neurotic” to “immature,” and finally to “psychotic” defense mechanisms (where “mature” denotes the most mentally healthy and adaptive responses, and “psychotic” the most unbalanced and maladaptive). A person with a preponderance of mature ego defense mechanisms who gets married, has children, and has an active network of friends and wider social engagements will be successfully adaptive in life.

In this sense, Vaillant’s conception of ego defense mechanisms plays a role analogous to Aristotle’s conception of the virtues: like Aristotelian virtues, they depend on and transcend the vicissitudes of luck. Luck tends to confer the capacity to adopt more mature ego defense mechanisms, but such mechanisms, once adopted, tend to smooth over life’s roughest edges. And though luck affects the adoption of ego defense mechanisms, the tendency is not absolute: some people have a good start in life but enact maladaptive defense mechanisms; others have a bad start, but enact adaptive ones.

The differences between Vaillant and Aristotle are pretty stark, but the similarities, through subtler, are also pretty striking. Both studies focus on success, treating pathology as derivative and secondary and drawing their conclusions about the nature of success unapologetically by empirical observation on specimens of that success—the best and brightest adult males, in both cases. Both adopt a methodology that is in some sense empirical, though each would disagree with the other about the proper methods and criteria of empirical inquiry into the subject-matter. Both authors tell us straightforwardly that empirical inquiry leads to difficult conceptual puzzles in need of resolution; whereas Vaillant wrestles inconclusively with these puzzles, treating them as (lengthy) footnotes to the main discussion, Aristotle claims to resolve them, treating them as the basic focus of the discussion itself. Both conceive of successful “coping” strategies or modes of life in hierarchical terms, whether in terms of a hierarchy of ego defense mechanisms (Vaillant), or a hierarchy of lives (Aristotle). Both have similar attitudes toward luck, suggesting (with some qualifications) that proper adoption or exemplification of the right life-strategy is more fundamental to a successful life than being the beneficiary or victim of good or bad luck, respectively. Finally, both stress the importance of sociality—whether in intimate and more widely socio-political contexts–to their respective conceptions of successful human functioning. Both value bouts of solitude, but take isolation to be dysfunctional or unhealthy.

Whatever the (substantial) differences between them, substantive and methodological, both Aristotle’s Ethics and Vaillant’s Adaptation are studies of (what each author takes to be) the essence of human success. Put somewhat differently, both are treatise-level prefaces to the attempt to manualize human success. They tell us what success is so that we can, drawing on their insights, produce a systematic account of how to achieve it. Though both discuss methodological issues, and touch on the objections that a skeptic might make, neither text betrays much specifically methodological anxiety (though Adaptation sometimes appears to). Both texts are in effect pitched to an audience that shares the author’s (and what he takes to be his society’s) basic normative and philosophical outlook. Aristotle does not, like Plato, spend a great deal of the Nicomachean Ethics worrying about sophistical challenges to the very possibility of ethics; Vaillant does not, like some of the radical critics he mentions in the book, worry that the Grant Study as a whole might rest on some fatal error. Aristotle assumes that we all seek eudaimonia; Vaillant assumes that we all want success in life. Neither book is intended for the malcontent who rejects its common sense starting point.

  1. Aristotle versus Vaillant: psychopathology and the role of morality

There are many important differences between the Nicomachean Ethics and Adaptation to Life, and correspondingly many ways of coming at those differences. For present purposes, I want to focus on each author’s conception of psychopathology, and in particular on the role played by “moral” concepts within each author’s theory of psychopathology.

At one level, the distinction seems obvious enough. Aristotle insists that virtue is integral to human flourishing, and that vice subverts it. Virtue and vice are paradigmatically moral concepts, so it seems to follow that moral concepts are central to Aristotle’s conception of psychopathology. By contrast, Vaillant tries his best to divorce the book’s discussion of adaptation and personality from specifically moral assumptions. The book’s message is by design (though not always in execution) morally neutral; the topic under discussion is adaptation to life, not right action, and no assumption is made that morally right behavior is adaptive, or that immoral behavior is maladaptive. This fundamental difference between Aristotle and Vaillant is, I think, highly instructive, laying bare some fundamental differences in approach between moral philosophy and behavioral science that both philosophers and behavioral scientists ought more explicitly to confront.[3]

Consider first Aristotle’s psychopathology,[4] as described in the first several chapters of book VII of the Nicomachean Ethics. Aristotle starts by making a five-fold distinction between virtue (arête), continence (enkrateia), incontinence (akrasia), vice (kakia), and brutishness (theoriotes), treating the five phenomena as mutually exclusive of one another, and though not quite exhaustive of logical space, then exhaustive of most of it. Virtue and continence are good states, virtue being better than continence; incontinence, vice, and brutishness are bad states, incontinence being slightly better than either vice or brutishness, but all three being qualifying as pathological. To state the obvious (from an Aristotelian perspective): virtue and vice, and continence and incontinence, are incompatible states involving contrary normative verdicts. A virtuous person cannot be vicious, or vice versa; a virtuous person is good and a vicious one, bad; and what is good cannot be bad or vice versa. A continent person cannot be incontinent or vice versa; a continent person is good (though not as good as a virtuous one), and an incontinent person is bad (though not as bad as a vicious or brutish one).

That said, from another perspective, virtue and vice do have something in common, albeit a similarity that only is one if we abstract entirely from the rest of their nature: both are wholehearted.* The virtuous person is virtuous because he wants at every psychological level to be virtuous, and feels no inner conflict at the practice of virtue; he’s not tempted by vice because he finds it repulsive and pointless. There is, from his perspective, nothing attractive about vice because there is no authentic pleasure to be gotten from it. Meanwhile, the vicious person is unapologetically vicious, not just to others, but internally. He’s sufficiently corrupt to take pleasure in the practice of vice, to think that practice right, and to have no sense of shame, guilt, or regret about indulging in it. He’s not tempted by virtue because he finds it repulsive and pointless; there is from his perspective nothing attractive about virtue because there’s no authentic pleasure to be gotten from it.

Something similar may be said, mutatis mutandis, of continence and incontinence. What they share in common (in the same eccentric sense of “share in common” as above) is the phenomenon of (intense) inner conflict. Both continence and incontinence are states of half-heartedness, confusion, self-doubt, inner division, dissociation, neurosis, low self-estimate, and self-defeat. The continent person is self-controlled in the sense that he does the right thing, and does it for the right reason. He knows what is right, and does what is right, but is haunted by psychic pain at doing so. Some (obscure) aspect of his desires rebel at doing what he knows is right; in consequence, he knows that some aspect of his desiderative make-up is disordered. This knowledge induces him to act against (some of) his desires and in favor of those that incline in the reverse direction. The incontinent person is in (something like) the same psychological state as the continent but lacks precisely the continent person’s self-control. The incontinent knows what is right, but finds himself doing what’s wrong, knowing that it’s wrong, and feeling psychic pain at the failure (qua failure). Like the continent, he knows that some aspect of his make-up is disordered; unlike the continent, he cannot control that aspect, but ends up controlled by it, and regretting the out-of-control nature of his existence.

To the extent that there is such a thing as “Aristotelian psychotherapy,” then, it applies (or would apply) to the continent and incontinent. The virtuous are without psychological flaw, and therefore have no need for therapy. The vicious and brutish are (in different ways) incorrigible and therefore beyond the reach of therapy. By contrast, the continent and incontinent occupy an intermediate zone between virtue and vice, and their very half-heartedness (especially the continent person’s) gives leverage for moral reform.

Vaillant’s psychopathology is in one sense a more complicated, and in another sense a much simpler affair than Aristotle’s. In the complex sense, Vaillant’s official psychopathology (summarized in Appendix A of the book) consists of eighteen ego defense mechanisms divided four ways (psychotic, immature, neurotic, mature). This taxonomy corresponds to almost nothing in the Nicomachean Ethics, and represents a conceptual incommensurability in the outlook of the two authors. In the simple sense, however, if we distinguish the five mature ego defenses from all thirteen of the non-mature ones, it turns out that in a certain sense, Vaillant’s psychopathology maps “perfectly” onto Aristotle’s. One of Vaillant’s basic assumptions, paradoxically drawn in part from Freud and in part from behaviorism, is that wholehearted psychological states indicate psychological health (regardless of the moral verdict we might give them) while half-hearted ones indicate neurosis (regardless of the moral verdict we might give them).

In Aristotelian terms, Vaillant’s assumption yields a very strange result. On Vaillant’s view, since Aristotelian virtue and vice are both wholehearted, both indicate mental health. Correspondingly, since continence and incontinence both manifest indecision, both indicate illness or dysfunction. On Vaillant’s view, then as far as mental health is concerned, it doesn’t matter whether a person is virtuous or vicious; what matters is that he is not continent or incontinent. At best, virtue is slightly preferable to vice because vice tends to flout social norms, and violation of social norms tends to impede social adjustment. But the tendency is far from absolute: if a person could indulge in vice without risk of detection by others, or indulge a vice commonly indulged in by many members of one’s society, there would be no psychological basis for preferring virtue over vice, continence over incontinence, or either continence or incontinence over vice.

I’ve described Vaillant’s assumption as partly Freudian and partly behaviorist, but it’s the specifically behaviorist rationale for the assumption that’s relevant here because it’s the behaviorist rationale that makes Adaptation a bona fide work of behavioral science. A basic axiom of behaviorist method is that we ought always to be able to operationalize the variables we use in a study of human behavior. To “operationalize variables” in this context means to identify each variable with a measurable magnitude that is expressed in overt, observable behavior. There is no way to operationalize the Aristotelian distinction between virtue and vice, or between continence and incontinence. Virtue may be good, and vice may be bad; continence involves self-control and incontinence its loss. But goodness and badness in the relevant sense are not susceptible of behavioristic measurement, and the distinction between self-control and loss of control is entirely introspective.

By contrast, the distinction between whole- and half-heartedness is an operationalizable distinction, detectable and measurable via a standard intake session in therapy. We can, in other words, conduct a detailed longitudinal survey and simply ask the experimental subject what he feels, or else do a therapy session and get the therapist to record what she thinks he really feels. Goodness and badness is too variable, and continence and incontinence too subtle, to observe and record. But conflict is something we can observe and record, as is wholeheartedness. And that, in fact, is what Vaillant and his colleagues did record.

The point is that for Vaillant, the desire for operationalization drives the basic assumptions of the study. To be is to be the value of a measurable variable. From a behavioral science perspective, if you can’t measure something, it may as well not exist. Since there is no measurable difference between virtue and vice, the distinction between them ceases to exist. Since there is a measurable difference between {virtue/vice} on the one hand and {continence/incontinence} on the other, that’s what ends up mattering. Since internal harmony is for Vaillant obviously preferable to internal conflict, virtue/vice trump continence/incontinence and indicates health or illness, as appropriate.

From an Aristotelian perspective, of course, Vaillant’s assumptions and claims are preposterous. In equating virtue with vice, Vaillant equates the good with the bad, and in doing so, runs roughshod over obvious moral distinctions. If we take the basic distinctions and platitudes of common belief as epistemically authoritative—as from an Aristotelian perspective, we should–we’re forced to conclude that the Grant Study, a supposed triumph of twentieth century behavioral science, is a moral and methodological travesty. An apparent triumph of longitudinal data-collection has been traduced by a conceptual framework that nullifies the effort put into systematizing it. Far from telling us what makes for adaptation to life, Adaptation to Life (from this perspective) offers a reductio ad absurdum of contemporary psychiatry. A psychiatry that tells us that virtue and vice (or continence and incontinence) are equally healthy psychological states, or (putting things another way) that treats the distinction betwen virtue and vice as essentially irrelevant to mental health, is epistemically bankrupt.

What we need (an Aristotelian might continue) is a new conceptual framework altogether, one that starts with basic distinctions between virtue, vice, continence, incontinence, and brutishness (or illness), that presupposes a basic conception of the individual virtues themselves, and that then operationalizes the causal relationship between conceptions of those things properly conceived (on the one hand) and mental health and dysfunction properly conceived (on the other). From this perspective, Vaillant’s supposedly path-breaking study merely foreshadows the over-medicalized chaos of DSM-5, which makes indiscriminate “disorders” of human life itself–virtuous, vicious, and everything in-between.[5]

From a contemporary behavioral science perspective, however, the Aristotelian complaint seems like an uninformed sort of dogmatism and question-begging. It’s easy to understand and perhaps sympathize with the folk psychological rationale of Aristotelian defenders of virtue (a behavioral scientist might say). They’ve inherited a certain conceptual framework from their Master, and want to preserve it all costs, invoking the “authority” of their “common beliefs,” regardless of the scientific credentials of those beliefs. So behavioral science aside, they want to insist on “commonsense grounds” that there is a “fundamental difference” between virtue and vice, that virtue promotes well-being, that vice subverts well-being, that continence is closer to virtue than incontinence, and by implication that continence is more productive of well-being than incontinence.

There’s a dilemma here (the criticism continues). Either these claims assert a priori conceptual connections, or they assert empirical ones. In the first case, Aristotelian moral philosophy looks like an elaborate but dogmatic language game whose basic assumptions may be driven by “commonsense intuitions,” but bear no relation to psychological reality as it exists outside of the game. Yes, in the game, virtue promotes well-being and so on, but it doesn’t follow that any such generalization obtains in the world outside of the game. That’s what we need science to tell us. Meanwhile, in the second case, Aristotelians claim to be offering empirical assertions about the psychological world, not offering analyses of the concepts in their heads. But in that case, they need to find some way of operationalizing and confirming the causal claims that they want to make. And they conspicuously lack this.

If virtue really promotes well-being, then there must be some sense in which virtue is (or conceptions of virtue are) an independent variable, and well-being or flourishing a dependent one. And conceptually, we then have to distinguish virtue from well-being to show how the one thing contributes to something distinct from it. Either the asserted causal connections obtains or not. If not, the theory is finished. But if so, how so? The connections ought to be measurable by established procedures of behavioral science—and if not by those procedures, then by some hitherto unheard-of procedures that pass scientific muster. But Aristotelian virtue ethicists not only have not produced such studies or such a science, they have insisted on claims flatly at odds with the possibility of coming up with one. A standard claim in the literature is that virtue “constitutes” well-being in a non-instrumental, non-causal way. If so, Aristotelians cannot in principle produce a conception of well-being that is distinct from virtue, and cannot conceive of the one as an independent and the other as a dependent variable. The result seems less like a viable theory than an anti-empirical fantasy insulated from the possibility of confirmation or falsification.

  1. The upshot

So ends the beginning of what ought to be, and would likely be, a very long argument. I’ve driven the discussion to this aporetic conclusion, not because I think that the disagreement between Aristotelians and behavioral scientists is irresolvable, but because I think it’s a disagreement that needs explicitly to be thrashed out in an interdisciplinary way, starting with the preceding disagreements, and working out from there.

As things stand, however, those disagreements are unlikely to take sustained, institutionalized form. There is, to be sure, some cross-disciplinary discussion between ethics and psychology at the professional level. But interdisciplinary inquiry of that sort is a specialized taste for people who have the taste, not a significant aspect of how philosophy and psychology are nowadays taught at the undergraduate level. Very few Departments of Psychology would insist that their undergraduate majors grapple with a text like the Nicomachean Ethics; Aristotle seems “unscientific,” “dated” and “obscure.” The disciplinary emphases and action lie elsewhere. Equally few Departments of Philosophy would insist that their majors spend much time on an empirical text like Adaptation to Life. Empirical work of that sort seems analytically unrigorous and mired in messy detail. Again, the emphases and action lie elsewhere. As long as we insist on seeing undergraduate study as a means of reproducing our professional fields, we’re apt to reproduce the existing structures of those fields, and perpetuate the assumptions behind them.

It’s worth asking a different question, however: whether the structures and assumptions behind our disciplines make for mental health or human flourishing in the world beyond those disciplines. That clarifying question, I’d suggest, leads us to confrontations of the sort I’ve sketched here—and to answers as unhinted-at here as they are worth pursuing. That, at any rate, is my pedagogical conjecture and proposal. As things stand, we’re in danger of creating psychology students mesmerized by operationalism and over-medicalization but insensitive to the common sense of “folk psychology” and ethics, and philosophy students with the reverse weaknesses. Aristotle and Vaillant are grist for the mill of that long and complicated session.[6]


[1] Aristotle, Nicomachean Ethics 2nd ed. Tr. Terence Irwin (Indianapolis/Cambridge: Hackett Publishing, 1999). George E. Vaillant, Adaptation to Life (Cambridge and London: Harvard University Press, 1977).

[2] For a good non-technical overview, see Scott Stossel, “What Makes Us Happy, Revisited,” The Atlantic (April 24, 2014),

[3] Though I lack the space to discuss it here, it’s plausible to think that the difference between the two authors conceals a subtle agreement as well: while Aristotle adopts a moralized conception of psychopathology, and Vaillant a morally neutral one, the gap between them narrows a bit if we keep two facts in mind. For one thing, Aristotelian virtue is a “moral” concept on some but not every conception of morality; depending on how one conceives of “morality,” there is a case to be made that Aristotelian virtue treads the boundary between moral and non-moral concepts rather than falling clearly on one or other side of that divide. Second, while Vaillant professes an adherence to moral neutrality throughout the book, he seems at times whether to honor his own stricture in the breach rather than the observance. At times, Vaillant himself seems to be lapsing into the view that specifically moralized ego defense mechanisms, like altruism, are central to happiness, and that paradigmatic vices, like racism or callousness, are indicative of maladaptation. The topic is too complex to handle here, however, and suggests that the discussion in the text is oversimplified.

[4] I borrow the suggestion that NE VII is a “psychopathology” from Francis Sparshott, Taking Life Seriously: A Study of the Argument of Nicomachean Ethics (Toronto: University of Toronto Press, 1996), chapter 3 (“The Pathology of Practical Reason”).

[5] For discussion, see Gary Greenberg, The Book of Woe: The DSM and the Unmaking of Psychiatry (New York: Blue Rider Press, 2013).

[6] Thanks to Ann Guillory, Kate Herrick, and David Riesbeck for helpful discussion on the issues of this paper.

*It occurs to me that Nicomachean Ethics IX.4 is something of a counter-example to this claim (1166b1-25). I may be conflating the vicious person’s unapologetically taking pleasure in vice with the vicious person’s wholeheartedly pursuing vice. I need to think that one through.

7 thoughts on “From the Nicomachean Ethics to the Grant Study

  1. Glad to get the props. It’s good to see that whole discussion was productive for someone other than me.

    Here’s a few very unrefined thoughts. First, Aristotle is committed to maintaining, at least formally, that the virtues are whatever traits of character will promote, preserve, and (partially) constitute our well-being when we act in accordance with them. In principle, the account of the virtues he gives us is open to revision in light of a better understanding of well-being. Though the account of the virtues is in part a specification of what human well-being would look like, empirical approaches like Vaillant’s could conceivably lead an Aristotelian to revise the theory of the virtues on the grounds that some set of traits promote happiness better than the ones Aristotle and his groupies have tended to emphasize. So I think Aristotle, of all the most famous and influential moral philosophers, is perhaps most open to learning from psychological approaches like Vaillaint’s. But, second, if the “function argument” does anything at all, it shows that subjectivist theories of well-being are mistaken. And this seems to be a strike against empirical approaches like Vaillant’s. Even if all the function argument did was show that, whatever human well-being turns out to be, it must be an active realization of human capacities, that would be enough to cast considerable doubt on any efforts to generalize findings like Vaillant’s to show that some way of life is better than others. Vaillant’s method seems to require nothing more than subjective satisfaction. Aristotle gives us some reason to think that this is insufficient.

    That said, he also seems to think that even the vicious person will not be subjectively satisfied, precisely because he is vicious. There are some straightforward reasons to doubt whether this could be true. But I wonder whether the cases that come to mind as counter-examples are really instances of vicious characters or, rather, cases of brutishness — people whose psychology is so distorted and perverted that they don’t even experience dissatisfaction with vice.

    But perhaps I’m just a moralist. Some have said as much.


    • Thanks. I think the paper was itself unrefined. I did it in a hurry last week, in one take, between classes. For reasons I’ll describe in a separate post (they’re characteristically tragic-comic) I didn’t even manage to get to the conference (which is happening right now as I write this, in New Jersey).

      My own inclinations are Aristotelian, so I’m inclined to find Vaillant’s account preposterous. That said, I was trying to find a way to structure the (aporetic) beginnings of a notional dialogue between an Aristotelian virtue ethicist and contemporary behaviorally-influenced psychiatrist/psychologist. Hence the structure of the paper, and my neutral-sounding summaries of each view.

      I’d grant both of the points you’ve made, but let me quote a bit from Vaillant to give you the sense of why he wouldn’t, and why most contemporary psychologists would be baffled by them. This is from Adaptation, pp. 48-52.

      First, since [mental] health is relative, the men [in the study] were only compared with each other. To argue about the comparative mental health, maturity, and ability to adapt of Fedor Dostoevsky and Richard Nixon would be a fruitless project….The subjects must serve as foils only to each other.

      Second, an effort was made to reduce the outcomes of the men, their childhood adjustment, their character type, the adequacy of their defense mechanisms, even their marriages, to numerical scores…[W]hen they can be systematically derived, numerical scores serve to codify value judgments so that they cannot be subsequently altered and reinterpreted to suit the investigator’s whim.

      Third, although statistics, numbers, ‘controls’, ‘blind’ ratings, are tedious, they are necessary in order to combat the distorting effects of preconception. Health is a value-laden concept; statistical association and experimental method help filter value judgments from fact.

      Fourth, the sample is patently drawn from a narrow spectrum of the population; in no sense is it representative.

      Fifth, these men have been studied by the techniques of long-term follow-up….[L]ongitudinal follow-up and multiple interviews permitted some control over the halo effect that makes the best outcomes look better and the worst outcomes look worse than they really are.

      I think most psychologists/psychiatrists would regard that as the methodological “basic minimum” for genuinely scientific study in clinical psychology. From an Aristotelian perspective, however, the assumptions involved range from “mixed bag” to “disaster area.” Because they do, it’s a substantial theoretical project to figure out how Aristotelian virtue ethicists are supposed to get their foot in the door as practitioners (rather than mere consumers, or cherry-picking consumers) of clinical research in the Vaillant genre. Crudely put: when do we meet the mainstream assumptions half-way, and when do we adopt an adversarial posture that treats their assumptions as the equivalent of sophistry? In other words, we need a 21st-century version of Aristotle’s Topics.

      Looking at Vaillant’s assumptions, the fourth and fifth are innocuous. The first is simply wrong, and the second and third, though legitimate in principle, are in this context implicitly motivated by problematically Humean assumptions (fact/value). Still, it’s one thing to reject the prevailing assumptions and another to fashion research methods that reflect the correct ones. I need to read a lot more before I can generalize, but my preliminary sense is that there’s a long way to go before we develop anything like an Aristotelian clinical psychology. I know that Robert Campbell and Jose Duarte are working on issues of this sort. I know Robert’s work better than I know Jose’s, however.

      Until that happens, I think people will tend to infer that the coherence that obtains from the combination of {contemporary psychology and preference-satisfactionist conceptions of well-being} confers greater epistemic warrant on both while casting doubt on objectivist conceptions of virtue ethics a la Aristotle. Not that I agree with the inference, but I can see where it’s coming from. (In fact, preference-satisfactionism vs Aristotelianism with respect to happiness is the subject of a symposium in Reason Papers that’s coming out on Monday, on Christine Vitrano’s The Nature and Value of Happiness. Vitrano takes the preference-satisfactionist line I’ve just described, against John Kleinig who defends an Aristotelian position.)

      Anyway, I guess my point there is just that there is a lot of bridgework to be done.

      On virtue, vice, and subjective satisfaction/dissatisfaction: I belatedly made the realization you describe in your last full paragraph after remembering NE IX.4, 1166b1-25. I interpret Aristotle to be saying there that the vicious person is psychologically torn,* but (to differentiate him from the akratic), wholeheartedly believes that the vicious thing is the right thing to do. So the situation is not quite what I described in the text of the paper, where the vicious person struts the stage of life with a “bright eye and gleaming coat,” joyful, internally untroubled, but totally evil. (Of course, if he did, Vaillant’s response would be: well, from a mental health perspective, then, evil is pretty healthy.)

      I could use a reading assignment here, if you have one. I’ve gotten so used to differentiating Aristotelian flourishing from “modern subjective conceptions of happiness,” that I’m losing my grip on the relation between virtue-realization and preference-satisfaction in Aristotle.

      *After-thought: Of course, the question arises, why is he torn? Or why must be be? The quick answer is that he lacks the relevant kind of love for himself, but then the question re-appears in a different form a ways down the regress: how do we, without circularity, differentiate the forms of self-love so that form X leads to a fully integrated life without regrets, while form Y leads to a disintegrated life with them?


  2. I won’t advance this as a considered interpretive thesis, but as I’ve read the passage in the past, I wouldn’t say that the vicious person is torn. He ends up regretting some of his choices and wishing things had gone differently, but he doesn’t find himself torn between what he thinks he should do and what he feels like doing (that’s why he’s not akratic). Read in one way, I think this is a rather plausible thesis. Vicious people are likely to be unsuccessful, their desires and commitments are likely to conflict, and they are therefore likely to be dissatisfied with their lives and regret how things have turned out in the past. But of course there are certain institutional and cultural arrangements that can shield people from the negative consequences of their actions, and people can be remarkably successful at avoiding any feelings of regret. I think it’s important that Aristotle emphasizes that vicious people in particular are not subjectively satisfied with their lives when they are alone, and hence they strive to avoid solitude. But these days of course one can avoid being alone even when nobody’s around by watching television or messing around on the Internet (ahem). So it may be possible to avoid that dissatisfaction indefinitely, by avoiding opportunities for reflection. Perhaps another way to put the thought is that it may be possible for the vicious to avoid dissatisfaction, but there’s something they’ve got to work to avoid. It still seems too easy to imagine people who really just aren’t dissatisfied, but I think it would be sufficient to show that there is a regular causal connection even if it can be broken. And it may matter that the claim is supposed to apply to vicious people who nonetheless still retain some sort of attraction to what is actually good, however distorted their conception of it and their desires have become — hence my comment earlier about vicious rather than brutish characters.

    If you want a reading assignment, I can’t think of a better discussion than the final chapter of Stephen White’s Sovereign Virtue. But I might be prejudicially biased in its favor. There may be a better discussion somewhere, but that’s the one that comes to mind.


    • “It still seems too easy to imagine people who really aren’t dissatisfied…” Actually, a classic case is General Dyer, the mass murderer I mention in my April 13 post. What’s alarming about him is that, outwardly at least, he not only seems unrepentant about his actions, but seems entirely unperturbed by the circumstances surrounding them. He seems like the kind of guy who might have a clear conscience about killing a couple of hundred people in cold blood. I’m guessing that Dyer is the sort of guy that Kant had in mind when he said that it’s “the coolness of a villain” that makes him “not only more dangerous but also immediately more abominable in our eyes…” (Groundwork, 394).

      I also think that Dyer-type cases (as revealed through his testimony to the Hunter Commission) are what motivate the turn away from Aristotelian ethics and toward some form of deontology. The implicit thought is that if virtue benefits its possessor, then vice must harm the possessor; so the viability or plausibility of Aristotelianism turns on a demonstration of the all-things-considered harmfulness to the practitioner of vice. There’s a puzzle here about what exactly counts as a demonstration of that sort, but it’s easy to think that the only or best way to demonstrate harm to the practitioner of vice is to find internal dissatisfaction in him, so that if no such dissatisfaction can be found, the view fails to meet the relevant burden of proof. And from that (presumed) failure we’re pushed to deontology.

      I basically agree with your interpretation of NE IX.4, but there’s something about it that doesn’t seem quite right to me. I can’t put my finger on it, however, so I’ll have to return to it some other time.

      Speaking of procrastination, I’ve been meaning to read the Stephen White book for a long time, so maybe I should quit messing around on the Internet and get to it.


  3. I think the thesis is probably doomed if we interpret it as applying on the level of particular actions. People too often simply don’t regret actions that they shouldn’t have taken — and, when they do regret them, their regret is too contingent on things having turned out otherwise than they’d planned. But I take it that the thesis isn’t about particular actions, but about how the vicious regard their lives more generally. So too, I don’t think the idea is that people blame themselves for doing what they shouldn’t have done, but that they are just dissatisfied with how things turned out; it may be that they should blame themselves, but that instead they find ways to externalize the source of failure. As I read the passage, the main idea is that genuinely vicious people can’t be successful overall, and so can’t be satisfied with their lives on reflection; that is consistent with them being satisfied with any number of particular actions and with their refusing to acknowledge that they are culpable for how things have gone. Even read in that way, though, there are some obvious problems. For one thing, if a vicious person lives in a society with cultural norms and institutions that reward vice, he may have very little to regret. For another thing, people’s ability to rationalize their actions shouldn’t be underestimated; Seneca was probably closer to the truth when he wrote that even recognizing one’s own faults is a sign of improvement. But the really worrisome thing about cases like you mention is the possibility they raise that even vicious people might be able to have fully coherent practical commitments that they successfully pursue. In some cases I think it’s fairly clear, or at least highly plausible, that the coherence is purchased at the expense of psychopathy. But it’s an empirical question whether or not it must be, and however astute an observer of human behavior Aristotle might have been, he was not in a position to pronounce on this topic definitively.

    I am unmoved, though, by the line of thought you sketch out for heading toward deontology rather than Aristotelianism. As I’m sure you already agree, the Aristotelian view doesn’t make subjective satisfaction criterial of the good, and so in this respect it is in no better or worse a position than the deontological alternative. Deontologists purport to identify reasons that bind rational agents independently of their subjective motivations. So do Aristotelians; it’s just that the Aristotelians have more plausible reasons to offer. Most importantly to my mind, the Aristotelian view claims to show that rational agents should act in some ways and not in others because doing so is good for them; deontologists, insofar as they don’t adopt a basically Aristotelian view of rational agency, take themselves to be offering reasons that supposedly bind rational agents without regard to whether the actions those reasons enjoin benefit the agents themselves. Immoralists and psychopaths will never be moved by argument, but if all we can offer them are “reasons” that make no connection to their own good, then I’m afraid we’ll have to admit that they’re quite reasonable not to be moved.


  4. Pingback: Spring 2015 Issue of Reason Papers Is Out | Policy of Truth

  5. Pingback: Call for Participants: 11th Annual Felician Ethics Conference | Policy of Truth

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s