Professional Documents
Culture Documents
(Routledge Studies in Ethics and Moral Theory 21.) Aristotle. - Peters, Julia - Aristotelian Ethics in Contemporary Perspective-Routledge (2013)
(Routledge Studies in Ethics and Moral Theory 21.) Aristotle. - Peters, Julia - Aristotelian Ethics in Contemporary Perspective-Routledge (2013)
(Routledge Studies in Ethics and Moral Theory 21.) Aristotle. - Peters, Julia - Aristotelian Ethics in Contemporary Perspective-Routledge (2013)
Perspective
R~~J!~~~~"P
NEW YORK AND LONDON
First published 2013
by Routledge
711 Third Avenue, New York, NY 10017
Simultaneously published in the UK
by Routledge
2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN
Routledge is an imprint of the Taylor & Francis Group,
an informa business
© 2013 Taylor & Francis
The right of the editors to be identified as the author of the editorial
material, and of the authors for their individual chapters, has been asserted
in accordance with sections 77 and 78 of the Copyright, Designs and
Patents Act 1988.
All rights reserved. No part of this book may be reprinted or reproduced or
utilised in any form or by any electronic, mechanical, or other means, now
known or hereafter invented, including photocopying and recording, or in
any information storage or retrieval system, without permission in writing
from the publishers.
Trademark Notice: Product or corporate names may be trademarks or
registered trademarks, and are used only for identification and explanation
without intent to infringe.
Library of Congress Cataloging-in-Publication Data
Aristotelian ethics in contemporary perspective / edited by Julia Peters.
p. cm. — (Routledge studies in ethics and moral theory ; 21)
Includes bibliographical references and index.
1. Aristotle. 2. Virtue. 3. Ethics. I. Peters, Julia, 1978–
B491.E7A74 2012
171′.3—dc23
2012027696
ISBN: 978-0-415-62341-4 (hbk)
ISBN: 978-0-203-07276-9 (ebk)
Typeset in Sabon
By Apex CoVantage, LLC
Contents
PART I
Themes in (Neo-)Aristotelian Virtue Ethics
PART II
Beyond (Neo-)Aristotelian Virtue Ethics
THE CHAPTERS
Part I of the volume opens with a decidedly critical piece. In his chapter
‘Aristotle on Virtue: Wrong, Wrong, and Wrong’, Thomas Hurka advances
penetrating criticisms of some of the core theses of the Aristotelian approach
Introduction 3
to virtue. Hurka challenges the Aristotelian tendency to blur the distinction
between the good and the right by making the virtues, which are constitu-
tive of a person’s goodness, objects of praise or blame. He puts into question
the Aristotelian doctrine of the mean and the idea that vice can always be
explained in terms of either excess or deficiency. Most importantly, he chal-
lenges what he calls the foundational egoism of Aristotelian virtue theory,
according to which a virtuous person’s ultimate reason for being virtuous
is grounded in their concern for their own flourishing. Hurka contrasts his
criticism with a sketch of his own recursive theory of virtue, which is opposed
to the Aristotelian approach in crucial respects and thus suggests itself as
an attractive alternative to it. Anthony Price’s reply to Hurka, ‘Aristotle on
Virtue: A Reply to Hurka’—which can also be read as an independent piece
reflecting on Aristotle’s conception of the motivation underlying virtuous
action—argues that contrary to Hurka’s charge, the Aristotelian virtuous
agent’s concern for his eudaimonia is not culpably egocentric. Rather, it is
expressive of an essential aspect of morality: namely, the fact that each agent
must be centrally concerned with, and take responsibility for, his own good
or bad action.
The exchange between Hurka and Price is followed by a block of chapters
on eudaimonism. Christoph Halbig’s chapter ‘The Benefit of Virtue’ offers a
critical examination of the thesis that the virtues benefit their possessor. Hal-
big looks at different versions of the claim represented in the contemporary
neo-Aristotelian literature and reaches the conclusion that the most plau-
sible and defensible version is much weaker than the one favored by most
neo-Aristotelian virtue ethicists. Daniel Russell and Mark LeBar, in their
chapter ‘Well-Being and Eudaimonia: A Reply to Haybron’, defend Aris-
totelian accounts of well-being against Daniel Haybron’s criticism.8 They
argue that Aristotelian eudaimonism, which makes individual well-being at
least in part dependent on the possession of features that are essential to our
human nature, can be defended against Haybron’s objections and is more in
tune with our intuitions about human well-being than Haybron’s alternative
individualist account. In her chapter ‘Virtue, Personal Good, and the Silenc-
ing of Reasons’, Julia Peters suggests a way of reconciling McDowell’s claim
that the virtuous never have to make genuine sacrifices when acting virtu-
ously with the truism that even the virtuous agent (sometimes) has a reason
for regret when forgoing a personal good for the sake of virtue.
Three chapters take up the theme of naturalism. John Hacker-Wright, in
‘Human Nature, Virtue, and Rationality’, offers a defense of neo-Aristotelian
naturalism by arguing that its appeal to human nature not merely estab-
lishes the fact that humans are essentially rational agents, but also can func-
tion as a basis for deriving substantive norms for human rational conduct.
Philipp Brüllmann, in his chapter ‘Good (as) Human Beings’, argues that
neo-Aristotelian naturalism—in particular in the version defended by Rosa-
lind Hursthouse—implies a problematic tension because it puts a conceptual
constraint on our conception of the moral good that is incompatible with the
4 Julia Peters
function it is supposed to play in testing the correctness of individual moral
judgments, in particular, judgments about which character traits are virtues.
Edward Harcourt’s chapter ‘Attachment Theory, Character, and Natural-
ism’ draws on attachment theory—a prominent empirical theory of child
development—in order to suggest a ‘modest’ version of neo-Aristotelian
ethical naturalism that conceives of human ethical life as continuous with
our lives understood in psychobiological terms. This modest version, Har-
court argues, is to be preferred to more ambitious forms that seek to identify
judgments about virtues and vices with judgments about natural perfections
and defects.
Nancy Snow, in ‘Notes Toward an Empirical Psychology of Virtue: Explor-
ing the Personality Scaffolding of Virtue’, draws on empirical psychology
in order to offer a response to the situationist challenge of the Aristotelian
conception of (virtuous) character. Snow seeks to show that contrary to the
situationist challenge, empirical psychology provides support for the kind of
global character traits that, on Aristotelian conceptions, constitute the virtues.
In her chapter ‘Natural Virtue and Proper Upbringing’, Candace Vogler
offers a reading of the (neo-)Aristotelian view that proper upbringing is
essential to the development of virtue and of the insight underlying correc-
tive accounts of the virtues. Proper upbringing, Vogler argues, facilitates the
development of natural virtue, which is a genuine, yet inchoate form of vir-
tue, limited to constraining deviant natural impulses. Full virtue, in contrast,
has to be combined with the possession of right practical reason. However,
even full virtue maintains a certain corrective function since it is one mark
of a virtuous character that its possessor is open to moral self-correction.
Timothy Chappell, in his chapter ‘Kalou Heneka’, draws on Aristotle in
order to defend the view that the category of the Noble or the Beautiful,
which is often evoked in the explanation of virtuous action, constitutes a
category sui generis of reasons for action that need not be reduced to the
category of moral or prudential reasons in order to carry explanatory force.
The second part of the volume contains three chapters devoted to explor-
ing alternatives to the Aristotelian approach to virtue. Christine Swanton’s
‘A New Metaphysics for Virtue Ethics: Hume Meets Heidegger’ draws on
Heidegger—in particular, his conceptions of aletheia and of ‘horizons of
disclosure’—in order to develop the metaphysical scaffolding for a Humean,
response-dependent type of virtue ethics. Erasmus Mayr, in his chapter ‘A
Kantian Plea for Virtues?’, presents a Kantian framework for thinking about
virtue, arguing that there is a distinctive and irreducible role to be played
by the virtues—understood as character traits a human being needs in
order to lead a good life—even within Kant’s ethics. In ‘Toward a Humean
Virtue Ethics’, Lorenzo Greco gives an outline of a Humean virtue ethics
that stresses the crucial differences between a Humean and an Aristotelian
approach to virtue, but emphasizes at the same time that the peculiarities
of Hume’s account have to be considered as potential strengths, rather than
weaknesses of his virtue ethical position.
Introduction 5
NOTES
REFERENCES
Themes in (Neo-)Aristotelian
Virtue Ethics
1 Aristotle on Virtue
Wrong, Wrong, and Wrong
Thomas Hurka
This account was widely accepted in the late nineteenth and early twentieth
centuries—in Britain by Hastings Rashdall, G. E. Moore, W. D. Ross, and
others, in Europe by Franz Brentano and his followers.2 It treats virtue as a
higher-level moral concept, involving a relation to items falling under other,
independently applied moral concepts. More specifically, it sees the virtues
as intrinsic goods that involve morally fitting attitudes to items with other
moral properties, and the vices as evils involving unfitting attitudes.
The account’s first proponents were consequentialists and therefore took
all the virtues and vices to involve attitudes to items falling under the conse-
quentialist concepts of good and evil. One of their claims was that if some-
thing is intrinsically good, then having a positive attitude toward it, that is,
desiring, pursuing, or taking pleasure in it—in short, loving it—for itself is
another intrinsic good and a form of virtue. Thus, if your pleasure is intrin-
sically good, my desiring, pursuing, or taking pleasure in it is also good
and an instance of virtue, more specifically of benevolence. By contrast, if
something is intrinsically evil, loving it for itself is another evil and vicious;
10 Thomas Hurka
thus, my desiring, pursuing, or taking pleasure in your pain for itself is evil
and, more specifically, malicious. The fitting and therefore virtuous attitude
to an evil such as your pain is negative, involving desire for or pursuit of its
absence, or pain at its presence; this hating your pain for itself is good and
involves the virtue of compassion. But hating something good, as when I
enviously want your pleasure to end, is vicious and evil. Attitudes whose ori-
entation matches the value of their object—positive to positive or negative
to negative—are virtuous and good, while ones that oppose it are vicious.
There can also be deontological virtues. If an act is right, my wanting to
perform it because it’s right is fitting and therefore virtuous—it involves
conscientiousness, or a Kantian good will. And it’s likewise virtuous to hate
doing what’s wrong. But whether its object is good or right, a virtuous atti-
tude need not care about it as good or right. If your pleasure is good, my
wanting it because it’s good is virtuous, but so is my wanting it just because
it’s a pleasure and independently of any thoughts about goodness. Likewise,
my hating lying is virtuous not only when I think of lying as wrong but also
when I just don’t like lying. An attitude to something good or right for the
properties that make it so is virtuous even when it doesn’t think of them as
good- or right-making.
A complete higher-level account must also say how virtuous or vicious
different attitudes are. Here it’s guided by an ideal of proportionality, which
says it’s best to love objects in proportion to their degrees of goodness or
evil. Thus, a fully virtuous person will be more pleased by another’s intense
pleasure than by her mild pleasure, and by as much as the first pleasure is
more intense; he’ll likewise be more anxious to relieve a worse pain. Some-
thing similar holds for deontological virtues. If some act’s being an instance
of lying does more to make it wrong than its promoting pleasure does to
make it right, he’ll be more averse to it as an instance of lying than drawn
to it as a promoting of pleasure.
However exactly it’s developed, the higher-level account treats the moral
virtues as intrinsically good, so they have value not just instrumentally, or
for the other goods they promote, but also in themselves. Being benevolent
by itself makes your life better and being malicious makes it worse. But the
account also makes virtue in several ways a secondary moral concept. First,
as a response to items falling under other moral concepts, it can’t be the
only or main such concept; unless other things are independently good or
right, there’s nothing for it to care fittingly about. Second, as so understood
virtue plays only a minor role in the evaluation of actions. Imagine that you
can give either a large pleasure to one person or a small pleasure to another.
Given the ideal of proportionality, it’s most virtuous to desire the larger plea-
sure more than the smaller and therefore to produce the larger pleasure. But
the claim about virtue isn’t needed to establish that you ought to produce
the larger pleasure. That already follows from the fact that it’s the greater
good, or from that plus the claim that you ought to produce the most good
you can. That in doing so you’ll also act from the most virtuous motive may
Aristotle on Virtue 11
be an additional reason to do the independently right act, but it can’t change
what this act is; that already follows from the facts that make your motive
best. Finally, and departing from many of the account’s proponents, I think
virtue is a lesser intrinsic good in the sense that it always has less value than
its intentional object. Compassion for another’s pain is good, but it isn’t
more good than the pain is evil; it can’t be better for there to be pain and
compassion for it than no pain and no compassion. Likewise for vice: a tor-
turer’s malicious pleasure in his victim’s pain isn’t as evil as the victim’s pain.
If you can eliminate only one of the two, you ought to eliminate the pain.
This is a brief sketch of a ‘higher-level’ account of virtue, and when we
turn to Aristotle’s account, we find several points of similarity. He too thinks
moral virtue is good in itself, contributing to a desirable life not just instru-
mentally but in its own right. He also thinks virtue is a matter of your atti-
tudes broadly conceived, of your desires, motives in acting, and pleasures
and pains. An act’s virtuousness depends not on its effects or conformity to
external moral rules but on inner states such as its motive and accompanying
feelings. But on other central issues he’s mistaken.
First a smaller point. In the Nicomachean Ethics Aristotle says the virtues
and vices are traits for which we’re praised and blamed (1105b31–1106a1,
1106a7).3 Since he recognizes that praise and blame are appropriate only for
things under our voluntary control (1109b30–33), he must hold that virtue
and vice are voluntary, and he defends that view in NE III.5. But his argu-
ments for it are unpersuasive.
In one passage he seems to argue that it’s always in our power to act
virtuously (1113b3–6), but a virtuous action must be done from a virtuous
motive, and someone who’s vicious can’t now produce a virtuous motive in
himself.4 He also argues that even if a vicious person can’t now act virtu-
ously, he’s responsible for his vicious action because he could have avoided
developing his bad character in the past: vicious people ‘are themselves by
their slack lives responsible for becoming men of that kind’ (1114a3–5).
But this claim is hard to square with his insistence on the importance for
moral virtue of the right childhood training and education (1095b4–12,
1103a15–18, 1103b3–6, 24–25, 1104b11–13,1105a1–2, 1179b24–27). If
you were raised badly by vicious parents, how could you start to develop
virtuous desires, and if you couldn’t start, how can you be blamed for not
having them now?
The concepts of praise and blame, like those of right and wrong, presup-
pose voluntariness: you can’t have acted wrongly or be to blame unless you
could have done otherwise. But no such requirement governs the concepts of
good and evil. A serendipitous pleasure is good even if no one voluntarily pro-
duced it, and pain evil even when it’s no one’s fault. The higher-level account
12 Thomas Hurka
makes use of only these last two concepts. It says virtue is intrinsically good
and vice evil, and they can be so regardless of how they came about. If Hume
and Mill were right that we have innate tendencies to be pleased by others’
pleasure and pained by their pain, the account says we’re naturally virtuous
and good, though we deserve no credit for this. If we innately delight in oth-
ers’ pain, as a bleaker view has it, we’re naturally vicious but not blameably
so. Aristotle’s claim that virtue is praised and vice blamed applies the wrong
concepts to them, forcing him into implausible arguments about voluntary
control. Those arguments aren’t needed if virtue and vice are instead said to
be just good or evil.
Another issue concerns the primary locus of virtue. We make virtue ascrip-
tions at two levels, one more global and one more local. Speaking globally,
we may say someone has the character trait of generosity or is a generous
person. More locally, we may say a particular act was generous or a particu-
lar feeling malicious. Is one of these two types of ascription primary? Do we
first understand the virtues as traits of character and count individual acts
or feelings as virtuous only when they issue from such traits? Or do we first
identify individual motives and feelings as virtuous and understand a virtu-
ous character as one that tends to produce them?
The higher-level account takes the second view, ascribing virtue proper-
ties first to occurrent states such as individual desires, acts, and feelings and
only then to dispositions. However, Aristotle takes the first view. He defines
virtue as a state of character (hexis) (1105b20–1106a13) and says that to
be done virtuously an act must issue from a ‘firm and unchangeable char-
acter’ (1105a33–34), otherwise it may be ‘in accordance with the virtues’
(1105a29) but it can’t be fully virtuous. Aristotle doesn’t think the mere pos-
session of virtue is the highest good; that comes only in the active exercise of
virtue, as in particular virtuous acts (1095b32–34,1098b33–1099a6). But
they’re only done virtuously if they issue from a stable character.
I think this view is both false to our everyday understanding of virtue
and morally mistaken. If you see someone kick a dog just for pleasure, do
you say ‘That was a vicious act, on condition that it issued from a stable
disposition to perform similar acts on similar occasions’, or just ‘That
was a vicious act’. Surely you say the latter. Your remark doesn’t concern
only the kick’s physical properties; it turns essentially on the motive from
which it was done. But it concerns only its motive at the time, indepen-
dently of any longer-lasting trait. Or imagine that a friend who normally
doesn’t do this gives $20 to a homeless person from concern at the time
for his welfare. If you say ‘That was uncharacteristically generous of you’,
you don’t contradict yourself. Or imagine that we’re a military commit-
tee deciding whether to give a medal for bravery to a soldier who threw
Aristotle on Virtue 13
himself on a hand grenade, knowing it would kill him and in order to save
his comrades. If an Aristotelian says ‘This is a medal for bravery, and we
can’t know whether his act was brave unless we know whether he would
have acted similarly a week before or a week after’, we’ll throw him out
of the room.5
Nor is the issue here just one of terminology. ‘Virtue’ is an evaluative term,
in that to call something virtuous is to call it somehow good, and Aristotle’s
claim that acts not expressing a virtuous character aren’t done virtuously
implies that they aren’t fully good: since they don’t involve the ‘exercise of
virtue’, they can’t make the same contribution to your good as ones that do.
(Perhaps they make no contribution.) And that seems wrong. Considered
just in itself and apart from the other things co-present with it in a life, an
out-of-character act of generosity or courage seems every bit as good as one
based in a stable disposition. The second act may be accompanied by more
acts of similar value in the same life, and that life may be better as a whole,
perhaps even in part because it contains enduring virtuous dispositions.6 But
Aristotle’s claim that the in-character act is by itself better is unpersuasive.
Both analytically and evaluatively, the primary locus of virtue is occurrent
desires, actions, and feelings apart from any connection to more stable traits.
A further issue concerns Aristotle’s differentia for the virtues among traits
of character, his doctrine of the mean. It says that every virtue is a mean
between two vices, and every vice an excess or deficiency with respect to
the same feeling as concerns some virtue. Thus the virtue of temperance is a
mean with respect to the desire for physical pleasure, a desire the excess of
which is self-indulgence and the deficiency of which is insensibility. Cour-
age is a mean with respect to fear, of which the excess is cowardice and
the deficiency rashness. Many present-day Aristotelians distance themselves
from the doctrine of the mean, but I think something like it can be part of
an adequate account of virtue. It can’t be the whole, however, most clearly
because of what it says about vice.
By taking all the vices to involve excess or deficiency, the doctrine implies
that there are no basic human impulses that are always evil: each is such
that in a proper or medial form it’s virtuous and good. But this leaves out
the worst forms of vice, such as malice and cruelty, which involve desire
for or pleasure in another’s evil. No form of these feelings is good; all their
instances are bad.7 The higher-level account makes traits like malice its cen-
tral cases of vice because they involve the positively unfitting attitudes of
loving an evil or hating a good. However, they’re excluded by the doctrine
of the mean, and it’s therefore no surprise that they don’t figure in Aristotle’s
main catalogue of vices in NE II–IV. These books discuss self-indulgence,
cowardice, profligacy, and other vices but not the positive desire for harm
14 Thomas Hurka
to others that’s intuitively the worst vice of all. Aristotle does mention this
desire in the Rhetoric (1382a1–16, 1386b33–1387a1, 1387b22–24), but
that work doesn’t contain the doctrine of the mean, and when that doctrine
appears in the NE, vices like malice don’t. Aristotle may seem to allow for
these vices when he says that not all feelings admit of a mean since some
such as spite and envy have names that already imply badness (1107a9–13).
But his explanation is that if we attach a name to the excess or deficiency
of some feeling, such as ‘gluttony’ to the excessive desire for food, there
will be no mean with respect to it because there’s in general no ‘mean of
excess and deficiency’ (1107a25). That’s precisely how he understands spite
and envy, as the excess and deficiency of another feeling that can be virtu-
ous (1108a35–b6). He continues to assume that our basic impulses all have
medial forms and therefore continues to exclude the worst vices.
I said the doctrine of the mean can figure in an account of virtue, and
it can in particular express the ideal of proportionality. Thus, a desire can
be ‘in a mean’ if it’s proportioned to its object’s value, wanting it neither
more nor less than its degree of value compared to other objects makes
appropriate. As so understood the doctrine can explain ‘vices of dispro-
portion’ such as cowardice and selfishness. A coward is vicious because he
cares much more about his comfort or safety than about some significantly
greater good, such as the preservation of several people’s lives, that he could
secure by risking it. By contrast, a rash person cares too little about his safety
because he risks it for much smaller goods, and a selfish person wants his
own pleasure much more than the greater goods of other people, which is
again disproportionate.
But this use of the doctrine of the mean isn’t available to Aristotle because
it doesn’t fit the general structure of his ethical view. This leads to the most
important objection to his account: that it gives the wrong explanation of
what the virtues are, resulting in a wrong and even repellent picture of the
virtuous person’s psychology.
EXPLANATORY EGOISM
EGOISTIC MOTIVATION
As well as saying virtuous agents choose acts for their own sakes, Aristotle
says they act for the sake of the kalon, often translated the ‘noble’ or ‘fine’
but with aesthetic connotations of the beautiful.20 This raises some addi-
tional as well as some familiar issues.
Because kalon is an evaluative concept, to choose an act as kalon is to
be motivated by an explicitly evaluative thought, as you need not be if you
choose an act for its own sake. If Aristotle thinks motivation by the kalon is
necessary for virtuous action, his account excludes a kind of action allowed
as virtuous by the higher-level account and on many views paradigmatically
virtuous: where you choose an act for properties that make it right but with-
out thinking of them as right-making, as when you relieve another’s pain just
because you want it to end and without any thought of your act as required.
If Aristotle denies that this kind of act is virtuous, his account is excessively
intellectualist in the same way as Kant’s, which finds moral worth only in
acts done from duty and not in ones that are simply compassionate.21
Another issue concerns the aesthetic connotations of kalon. Is choos-
ing an act for its beauty not again choosing it for an inappropriately self-
centered reason, one focused on the aesthetic quality it can add to your life
rather than on any benefits it will give others? Sidgwick read Aristotle this
way, saying his virtuous agent makes ‘a deliberate choice of virtuous acts
for the sake of their intrinsic moral beauty, and not for any end external to
Aristotle on Virtue 21
the act’, so ‘The limits of Aristotle’s Liberality are not determined by any
consideration of its effect on the welfare of its recipients, but by an intuitive
sense of the noble and graceful quality of expenditure that is free without
being too lavish; and his Courageous warrior is not commended as devot-
ing himself to his country, but as attaining for himself, even amid pains and
death, the peculiar kalon of a courageous act’.22
The objection implied here is, however, too quick. Since being kalon is
a supervenient property, any act that’s kalon has non-evaluative properties
that make it so, and to choose it as kalon is to choose it believing it has those
properties. What are they?
Aristotle is characteristically disappointing on this topic, making only
vague and even contradictory statements. Sometimes he suggests that an
act is made kalon by properties it has just as an act and independently of
its motive, as when he says a liberal person will ‘give for the sake of the
noble, and rightly; for he will give to the right people, the right amounts,
and at the right time’ (1120a24–6; also 1120b3–4, 1121b3–7, 1147a29–32,
1151b18–21, 1177b16–18). At other times whether an act is kalon seems
to turn on its motive, as when he says the end of courageous action is con-
formity to a courageous state of character, which is noble (1115b20–22), or
that it’s especially noble to act in the face of great danger, which you’re then
not deterred by (1115a24–31, 1169a21–26), or to benefit another without a
view to repayment (1162b36; also 1171b20–23, Rhetoric 1366b35–67a5).
And of course an act could need both types of property to count as kalon,
though Aristotle never explicitly says this. It’s surely central to an adequate
account of virtue to specify clearly what non-evaluative properties a virtu-
ous person chooses her acts for, but Aristotle’s discussions of the kalon do
not do that.
We can, however, consider the two main possibilities. One is that an act
is made kalon by properties it has apart from its motive, which can include
its being likely to benefit another person.23 Even if this is Aristotle’s view,
however, it still faces the objection that the agent’s primary concern is his
own giving of the benefit rather than the resulting state of the other, such
as her being free of pain. (Recall that in X.7 the benefactor thinks the effect
he produces is noble because it depends on his action.) And the view is hard
to reconcile with the aesthetic aspect of the kalon since merely instrumental
properties, though they can by themselves make an act worth choosing,
don’t usually by themselves make it beautiful. (This may have been part of
what motivated Sidgwick’s reading.) If I cut off your leg to save you from
dying or upbraid you harshly because that’s the only way to improve your
character, what’s remotely beautiful in what I do? There can be aesthetic
quality in achieving an end in an especially elegant or efficient way, but
not all instrumentally good acts do that. An act can also be beautiful if it’s
‘fitting’ to its situation, as an act of gratitude can be to a previous benefit;
Ross suggested this reading for all ancient ethical uses of kalon.24 But as
C. D. Broad argued, while the concept of the ‘fitting’ is appropriate for some
22 Thomas Hurka
moral considerations such as gratitude and promise-keeping, it isn’t appro-
priate to that of promoting good consequences, which involves the different
concept of ‘utility’.25 To choose an act just because it will have good effects,
as a virtuous person often does, isn’t to choose it for a property that can
plausibly make it kalon.
The other possibility is that acts are made kalon by their motive. This
better fits the aesthetic side of the kalon since the motive of an action is
intrinsic to it, and a good motive can be said, at least on the higher-level
view, to ‘fit’ the value of its object. Moreover, several commentators have
ascribed this kind of view to Aristotle.26 But as well as still not addressing
the objection about valuing virtuous acts over their effects, the view makes
virtuous motivation implausibly complex. A virtuous person, it holds, first
has a base-level virtuous desire, for example, to relieve another’s pain. Then
he sees that an act done from that motive will be kalon and forms a sec-
ond, higher-level desire to do it because it will be kalon, or because it will
have that initial virtuous motive. Must virtuous action always have this self-
reflective, double motivation? Does it even often have it?
And there’s again a question about self-centeredness. If the virtuous agent
has two desires, one to relieve another’s pain and the other to do an act moti-
vated by that desire and therefore kalon, which is his main or most strongly
motivating desire? A parallel question can arise after he acts: what’s he most
pleased by then, that he relieved another’s pain or that he acted from the vir-
tuous desire to do so? Aristotle’s answer to both questions seems to be that
the higher-level, self-reflective attitude is the stronger one. He says countless
times that virtuous agents act for the sake of the kalon, which on the view
now under consideration is to do an act because it will have another virtu-
ous motive, and speaks much less often of agents’ doing acts because they’ll
benefit others. Bernard Williams called an agent ‘morally self-indulgent’ if
‘what the agent cares about is not so much other people, as himself caring
about other people’, or if he ‘focuses disproportionately upon the expression
of his own disposition’.27 If Aristotle’s virtuous person chooses an act pri-
marily as kalon, where that depends on its having another virtuous motive,
he’s self-indulgent in Williams’s sense.
It’s therefore not only Aristotle’s descriptions of characters such as the
courageous person on the battlefield, the megalopsychos, the philautos, and
the benefactor especially pleased by what he produced that give an unattrac-
tively self-centered picture of virtuous motivation. The same follows from
some of his more general claims, such as that a virtuous person chooses his
acts ‘for their own sakes’ or for having the quasi-aesthetic quality of being
kalon. My main argument has been that this self-centeredness isn’t a lapse
on Aristotle’s part but an expectable consequence of his overall ethical view.
On that view, recall, all my acts are chosen as means to a chief good that’s
my eudaimonia, so anything choiceworthy for me must contribute to my eudai-
monia. But no state of another person, such as her being free from pain, can do
that; my life can’t be better because of something true of someone else, and as
Aristotle on Virtue 23
a result no such state can be good in a way that by itself gives me a reason to
act. What can give me a reason is only something true of me, such as that an
act will be one of my relieving your pain, or one in which I act from a virtu-
ous motive. It therefore can’t be surprising that those are the primary foci of
Aristotle’s virtuous agent. He isn’t pleased or pained by states of others uncon-
nected to his own agency because those states aren’t relevant to his good. And
he doesn’t first want a good of another, such as her being free from pain, and
only then want to do an act that will produce it; he first wants to do that act
and will only value its result because it’s one he produced. The whole structure
of Aristotle’s view pushes his virtuous agent to look mainly at his own acts and
own motives in a way Ross said involves ‘self-absorption’.
There’s a natural explanation for these facts. As C.C.W. Taylor has
argued, Aristotle developed his account of virtue in a society still influenced
by a Homeric conception of the good or admirable person as essentially
competitive, wanting to be superior to others in aspects of life attended with
honor, pleased with himself when he is superior, and therefore more self-
focused than anyone we today could see as fully virtuous. Hence Aristotle’s
jarring-to-us descriptions of ‘virtues’ like magnificence and megalopsychia,
while foreign to our ethical outlook, fit that of his Greek society.28 I would
extend Taylor’s point by saying the same influences led Aristotle to posit
an ultimate goal for ethical life that’s similarly egoistic, involving for each
person only features of his life and not giving ground-level importance to
what happens to others. Like more specific features of his account of virtue,
the underlying structure of Aristotle’s view reflects an agonistic Greek ethos
that’s some distance from our moral thought today.
CONCLUSION
I’ve argued that Aristotle wrongly thought virtue is praised and vice blamed,
wrongly made the primary locus of virtue dispositions rather than occurrent
mental states, and wrongly excluded, with his doctrine of the mean, the
worst moral vices. But my main criticism has been that his account of virtue
is objectionably egoistic, especially as compared to the higher-level account.
This last contrast is worth making more abstractly.
We can distinguish two general approaches to the concept of virtue, which
can be called the outside-in and the inside-out. The outside-in approach
takes there to be values or, more generally, normative factors outside a per-
son’s motives and attitudes and holds that the virtues involve appropriate
responses to those factors. What makes an attitude virtuous is its relation
to something outside itself and often outside the agent, as when its object
is another person’s pleasure or freedom from pain. This externally-based
explanation of what makes the virtues virtues goes with a picture of virtu-
ous motivation as likewise externally focused, so a virtuous person cares
most about his virtues’ objects, such as another’s pleasure or pain, and only
24 Thomas Hurka
secondarily about his own virtuous motives in pursuing them. The inside-
out approach, by contrast, doesn’t relate the virtues to external values since
it doesn’t recognize any. It just says the virtues are good states of the person,
or intrinsic constituents of an overall good or eudaimon life for him. It there-
fore can’t explain why a given virtue such as benevolence is one; it can only
assert that it is. And it goes with an internally-focused picture of virtuous
motivation, where the virtuous person cares primarily about his own virtue
and its expression and only secondarily about the states of others his acts
can, if successful, bring about.
The higher-level account illustrates the outside-in approach and Aristotle’s
the inside-out, and I’ve tried to show that on several crucial points the former
is more attractive. It gives better explanations of why the virtues are virtues
and of why we should treat others in the way the other-regarding virtues would
lead us to: the ultimate reason isn’t that this will make our lives better, but that
it will make the others’ lives better. It also gives a better picture of the virtuous
person’s motivation as externally rather than internally focused. For a long
time the work of Rashdall, Moore, Ross, and other moral philosophers of
their era was ignored and even denigrated. As a result their higher-level account
of virtue was also ignored, and accounts modeled on Aristotle’s attracted the
bulk of philosophers’ attention. But the higher-level account is by far the more
illuminating of the two; in comparison, Aristotle’s is a dead end.
NOTES
REFERENCES
Socrates well instances Aristotle’s brave man who, in the presence of great
danger, is only ‘gently’ or ‘mildly’ afraid (ērema, Eudemian Ethics [EE] III.1
1228b29, 38); though not without apprehension (which is why he keeps look-
ing around), he is free of disturbance (he is atarachos, Nicomachean Ethics
[NE] III.8 1117a19, III.9 1117a31). Like the megalopsychos or man of proper
pride, he maintains an even step and a level voice (IV.3 1125a12–16). And
all this not idly, but in order to save his own life and (as Alcibiades’s narra-
tive emphasizes) that of his companion, Laches. A different agent might risk
danger gratuitously in order to display his sang-froid; Socrates is a sensible
28 Anthony Price
man and not a swashbuckler. The danger is inescapable, and he faces it with
composure in order to save two lives.
That is his external goal. (In a happier situation, this would have been
victory.) He may also have an internal goal: to show himself brave, and
thereby—as Aristotle would have it—to achieve the fine or noble (to kalon).
Practical deliberation starts, as I read Aristotle, from a concrete goal that
an agent adopts in context. In the EE, the general statement ‘Those who have
no target before them are not in a position to deliberate’ (II.10 1226b29–30)
comes immediately after a specific example: ‘The carrying of goods is a cause
of walking if it is for the sake of that that a man walks’ (b28–29). Socrates’s
goal was to save his own life and that of Laches. Had his goal been different,
say to kill a Bœotian, he would have behaved differently.
However, saying this cannot suffice to show his action to be brave, for
deliberation that is simply oriented toward achieving a certain concrete goal
succeeds through a mere ‘cleverness’ (deinotēs) that ‘is such as to be able to
do the things that tend towards the mark we have set before ourselves, and
to hit it’ (NE VI.12 1144a24–26, tr. Ross). For courage as a virtue, closely
linked to practical wisdom, the agent needs two further capacities: his initial
selection of an end must manifest a good character sensitive to context, and
in thinking through how to achieve that end he must exercise a good judg-
ment about what is worth pursuing or enduring for the sake of what. (This
may lead him to discard his initial end in favor of another.) Aristotle pairs
these two when he writes, ‘Virtue makes the end right, practical wisdom
the things towards this’ (1144a7–9). This somewhat cryptic remark is best
understood by reference to EE II.11, where we read, ‘Does virtue make the
goal or the things towards the goal? We say that it is the goal, for of this
there is neither reasoning nor a logos’ (1227b23–25). The agent’s character
reveals itself in what occurs to him as an external goal worth achieving in the
situation in which he finds himself. He then calculates how he may achieve
this, or—if there are alternative ways or means—best achieve this (NE III.3
1112b15–17).
Obstacles of various kinds may then obtrude. It may turn out simply
impossible to achieve the goal in the context: ‘If we come to an impossibil-
ity, we give up the search, e.g. if we need money and this cannot be got’
(1112b24–26). A different impossibility is if the goal can be achieved but
only in a way that is ethically out of the question. Less dramatically, the cost
of achieving this goal may be such that it is better to pursue another goal less
desirable in itself but cheaper to achieve in context. Estimating this requires
reasoning that is not purely instrumental, and yet also falls within practical
wisdom. We read in the De Anima [DA], ‘Whether one shall do this thing or
do that thing it is the work of reasoning to decide. And such reason neces-
sarily implies the power of measurement by a single standard; for what one
pursues is the greater good’ (III.11 434a7–9). What Aristotle means here by
‘measurement by a single standard’ (heni metrein) is not clear. It can hardly
mean a universal standard that applies in every context; it might mean a
Aristotle on Virtue 29
single standard relevant to the present context, or (more realistically still) a
single standard for each comparison that has to be made on the way toward
a final arbitration.
Such complications lie, I believe, behind a sentence that has been much
debated: ‘If it is characteristic of the practically wise to have deliberated
well, excellence in deliberation will be correctness with regard to what
conduces the end of which practical wisdom is the true supposition’ (VI.9
1142b31–33). On the face of it, this contradicts statements that the end is
provided not by practical wisdom, but by virtue. It may then be suggesting
that practical wisdom has two distinct spheres of operation: it identifies
first the best end (either in general, or in context), and then the best means
(certainly in context). However, we do better justice at once to the other
evidence I have cited, and to the realities of the case, if we notice that the
sentence places wisdom’s role in the selection of an end within deliberation,
and not in advance of it. I take Aristotle’s thought to be that truth in the
selection of an end is only reliably achieved when an exploration of possible
ways and means has identified which goal is best practicable in context.
So understood, Aristotle is faithful to the familiar phenomenology of
practical thinking. An agent doesn’t start his deliberations by (as it were)
closing his eyes and determining an a priori starting point for deliberation.
Nor does he achieve the feat, to be ruled as impossible by Hume, of infer-
ring an ‘ought’ of decision from an ‘is’ of description. Rather, a reflective
inspection of his situation prompts a selection of a provisional goal; this is
followed by a thinking through of possible ways and means that may discard
or revise the goal, or confirm that, in context, it is achievable in a way that
is acceptable.
Nowhere, here or elsewhere, is there is any indication of a general and
effective decision procedure. (Anyone who reads DA III.11 in isolation as a
gesture in the direction of one must find the gesture hollow.) Rather, there is
a frequent emphasis upon the variability of circumstances and the absence
of any universal rule (NE I.3 1094b11–27, V.10, IX.2 1164b27–1165a14).
Aristotle envisages no superordinate end by reference to which subsidiary
goals might be assessed. He displays no interest in innovation that could
achieve neither theoretical truth (since it would not be a contribution to
theory), nor practical truth (which would require its catching on and being
realized in practice). To characterize what the agent is trying to achieve in
general terms, we have to turn to internal goals. In a situation of danger,
whatever other virtues may be operative, the well-intentioned agent aims to
act courageously, and thence well, and nobly or finely (kalōs). This demands
attention to whatever considerations may be relevant to action in context:
‘The man who faces and who fears the right things and from the right motive,
in the right way and at the right time, and who feels confidence under the
corresponding conditions, is brave; for the brave man feels and acts accord-
ing to the merits of the case and in whatever way the logos directs’ (III.6
1115a17–20). What he achieves by doing justice to the situational variables
30 Anthony Price
can be identified from case to case by reference to external goals, but in gen-
eral terms only by reference to internal goals. Thus Aristotle links the exer-
cise of overall judgment to the agent’s achievement of an ethical end: ‘The
end of every activity is conformity to the corresponding state of character.
This is true, therefore, of the brave man as well as of others. But courage is
noble. Therefore the end also is noble; for each thing is defined by its end.
Therefore it is for a noble end that the brave man endures and acts as cour-
age directs’ (III.7, 1115b20–24; cf. in general VI.5 1140a25–28).
In Aristotle’s view, therefore, virtues are not to be construed instrumen-
tally as dispositions that measurably maximize the good effects of action.
Rather, they are dispositions to act well in ways that can be roughly clas-
sified within spheres of action or feeling (as courage involves responses to
danger involving two passions, fear and confidence), but which all involve
a sensitivity to whatever considerations merit attention (whence the unity
of the virtues). The core of what the agent is choosing and doing (which is
a ‘this’ for the sake of ‘that’, viz. a way or means for the sake of an end,
EE II.10 1226a11–13) is specified by a piece of reasoning that Aristotle
takes to be syllogistic in form (cf. NE VI.12 1144a31–2, De Motu Anima-
lium [DMA] 7). Yet this is the tip of an iceberg of perceiving and thinking
that involves the whole situation and has as its target nothing less than acting
as, in context, is good and best. The agent thus displays a complex practical
orientation that is all of a piece: it is only by pursuing the best external goal in
context, with attention to other relevant costs and benefits, that he can aspire
to realize his standing aspiration to live and act well.
II
Action in Aristotle thus possesses a double teleology: for instance, A risks his
own life in order to save B’s life—and all this (means and end taken together)
for the sake of acting bravely, finely, and well. Here risking his own life is
instrumental toward saving B’s; without any such worthwhile goal it would
not be brave, but rash. And it is only if he risks his own life for another’s
that we have a paradigm of acting finely. Aristotle finds it characteristic of
the decent agent that ‘he does many things for his friends and country, and
if necessary dies for them’ (NE IX.8 1169a18–20).1
We have to bring together two general claims. eudaimonia is always the
final end of action (or more precisely, as becomes clear later, deliberate
action): ‘It is a first principle; for it is for the sake of this that we all do all
that we do’ (I.12 1102a2–3). Equally, acting virtuously involves choosing
to do what is virtuous for its own sake (II.4 1105a28–32). Any apparent
tension between these two statements can be allayed by noting that it is
by and in doing what he does (say, risking his life in the right context for a
good goal) that the agent succeeds in being eudaimōn in the sense of acting
well. Within the agent’s motivation we can distinguish these two elements:
Aristotle on Virtue 31
instrumentally he risks his life as a means to achieving an external goal;
constitutively he does all that as a way—and perhaps the only way open to
him in context—of acting well.
Aristotle at once confirms this, and apparently confuses it, when he is
arguing in X.7 in support of his own opinion that even better than ethical
action is the contemplation of scientific truth. One may half regret this chap-
ter as a piece of higher salesmanship. Taken on its own (and in separation
from X.8), it might be interpreted as taking back the NE’s valuation of ethi-
cal action as a great good in itself and replacing that by an eccentric claim
that only intellectual contemplation is itself a way of being eudaimōn. In the
course of pursuing his argument, Aristotle says the following:
On the face of it, this is not only inconsistent, but incoherent: if, for exam-
ple, military action can itself achieve nobility, how can it fail to be desir-
able for its own sake? (The kalon is not an instrumental value.) However,
coherence can be restored if we recall the exact wording of II.4: acts that
are in accordance with (kata) the virtues may fail to be done virtuously.
For example, if I risk my life in battle in a way that might well benefit my
comrades, but without the right motivations (I may not intend their benefit,
or, even if I do, I find no intrinsic value in acting so, but view it as a fatal
chore), I fail to act virtuously in the full sense and hence to act well. What
accords with the virtues, but fails in itself to achieve an intrinsic value, is
just what I do concretely (which I have been describing vaguely as ‘risking
my life’). That has no value that is independent of its instrumental role. This
differentiates it from contemplating whose value is only intrinsic and which
can be appreciated as being incommensurable in value just so long as it is
permissible in context. The evaluation of virtuous action is therefore more
complex than that of contemplation: risking one’s life, taken as such, is only
instrumentally valuable, that is, in serving a good external end; if it does
pursue that extrinsic value in context, we have a more complex act—risking
one’s life for a good end—that may well (in the absence of any overriding
counter-consideration) possess the intrinsic value of acting well. We could
put Aristotle’s point as follows: contemplating truth, taken as such, is closer
to eudaimonia than is risking one’s life, taken as such; we have more to
add in the second case than in the first if we are to identify a case of acting
well and achieving eudaimonia. How far this justifies placing contempla-
tion above virtuous action is debatable; yet, so read, it is intelligible—and
confirms my understanding of Aristotle.
32 Anthony Price
The ethical structure that I have identified has two further features that
we should note. First, it is not implied that A values B’s life less than he val-
ues his own acting well (as if, in a different situation, he would rather that
B died than that he should, on a single occasion, fail to act well). It is saving
B’s life that he pursues for the sake of acting well. Part, though not all, of the
content of this is that, if the situation were such that saving B’s life would not
be acting well, he would not pursue it (or only in error). Secondly, a question
whether A acts well in order to save B’s life does not arise. Adding that A not
only risks his life but acts well for B’s sake would reiterate the same motiva-
tion with nothing gained except confusion. If A were so motivated even in
acting well, it should follow that he not only acts well, but also acts better
than well since he is acting well out of a noble, because altruistic, motive.
But then even acting better than well should be enhanced if it also is gener-
ously motivated—which would yield an infinite sequence of bootstrapping
achievements. A safer rule is that motivation cannot thus be reduplicated one
level up: the altruistic motive attaches to the act of risking his life whereby the
brave man acts well, but not to acting well itself. Acting well is not the sort of
thing to be enhanced in this way. It itself cannot be other-interested—which
should absolve it of the charge of being self-interested.
This already gives us an understanding of Aristotle’s claim in I.7 that,
while many things are worth pursuing for their own sake, only eudaimonia
is fully final or perfect (teleios) in being pursued for its own sake, and never
for the sake of anything else (1097a33–34). Yet it becomes evident that, so
read, this evaluation of eudaimonia is restricted: that its value is final in this
sense does not do much to show that its value is great. Compare a connected
concept: acting commendably. It is doubtless better to act commendably
than to act otherwise, and acting commendably is a thing achieved by having
the right motivations and not a thing that becomes valuable, or even more
valuable, by being well motivated itself. And yet, as a value, it seems minor
even if it is not epiphenomenal: meriting commendation is not a mere side-
effect (as perspiration may be of physical motion in battle) and may well be
a goal of the agent; yet it is hardly a salient goal and one without which his
attitude to his action would be significantly different.
What makes it important that, by risking his life for a good cause, the
brave man is thereby acting well? This must be some feature of his action
that connects with another formal feature of eudaimonia, which is its self-
sufficiency: ‘The self-sufficient we now define as that which when isolated
makes life desirable and lacking in nothing; and such we think eudaimonia
to be’ (I.7 1097b14–16).2 This should not be read as implying that, when an
agent consciously acts well, he must suppose that, within the context of his
present action, nothing could have gone better. That can’t be what Aristotle
means when he later remarks that the good man ‘has nothing to regret’ (he
is ametamelētos, IX.4 1166a29). There he may merely mean that the good
man has never to regret how he acted, wishing that, even with the informa-
tion he then possessed, he had decided and acted otherwise. I.7 must be
Aristotle on Virtue 33
claiming more than this, but not that nothing could have enhanced his life at
the time. (Very trivially, his immediate experience might have been better still
either without a pang of toothache, or with a praline in his mouth.) Exactly
what more Aristotle has in mind must connect with his idea that the agent
who acts well may also be acting finely or nobly. This is evidently in part a
hedonic concept. There is a pleasure to be found even in dying as a hoplite:
‘The end which courage sets before itself would seem to be pleasant, but to
be concealed by the attending circumstances, as happens also in athletic con-
tests’ (III.8 1117a35–b3). One may compare a hope in I.10 that, even in great
misfortunes, nobility may ‘shine through’ (1110b30–31). A more transparent
pleasure accompanies the life and death of a hero: ‘He would prefer a short
period of intense pleasure to a long one of mild enjoyment, a twelvemonth of
noble life to many years of humdrum existence’ (IX.8 1169a22–24). How-
ever, more needs to be said to explain what this unique value is that can pro-
vide so rare and rich an enjoyment.3 Here, I feel, Aristotle rather lets us down:
it remains unclear precisely what constitutes to kalon and how it supervenes
intelligibly upon the conditions for acting virtuously laid down in II.4.4
There are thus at least two aspects to Aristotle’s conception of the agent
who ultimately acts for the sake of acting well. In part, this requires that he
is not putting on blinkers in pursuit of some limited goal but trying to do
justice to however many, and varied, considerations arise within the context
of his action. Additionally, it finds a special and intrinsic value in action that
achieves this goal. It is within action (and certain other activities), and not
in the results of action, that the best of life is to be found. In Alcibiades’s
anecdote, it is Socrates, and not Laches, who had the better day. We may or
may not find this credible. Yet it is just what Aristotle committed himself to
when he argued that the human good is nothing other than ‘activity of soul
in accordance with virtue’ (I.7 1098a16–17).
III
He did not reason that, since in all likelihood an innocent man would die
anyway (and why not then one rather than two?), he was free to save his
own skin. Instead, he decided to act justly, where ‘acting justly’ denotes a
way of acting open to him that he ranked incomparably above other ways.
The thought that evil may come into the world, but must not through me, is
not culpably egocentric. It places my agency, where it must be for each agent,
at the center of my life. What it precisely does not do, which would indeed
be egocentric, is to displace that concern by another that is focused upon
personal pleasure or advantage. To love one’s neighbor as oneself is not to
be indifferent as to whether I do wrong, or he does. Such an attitude would
be not impartial, but irresponsible. Yet to take responsibility for one’s action
is not to suppose, insanely, that it matters more from some non-agential
point of view (say God’s, or—less intelligibly—Sidgwick’s point of view of
the universe) if I act badly, than if someone else does. That is not the reason
why one kicks oneself if one has acted badly oneself, and not if another has
acted badly—which rests rather upon the logic of kicking oneself. Personal
responsibility is inescapable, and inescapably first-personal.
In refusing to collaborate in arresting Leon, even at the likely cost of
losing his own life, Socrates was acting in accordance with what is known
as ‘Democritus’ Maxim’ (B 45), roughly ‘It is worse to do wrong than to
be wronged.’ This is most familiar to us from early Plato (e.g., Crito 49b,
Gorgias 469c) but is also explicit in Aristotle: ‘Acting unjustly is the worse
[of acting unjustly and being unjustly treated], for it involves vice and is
blameworthy’ (NE V.11 1138a31–32). The thought is not that, for each
agent, it is impersonally worse that he should wrong anyone else than that
anyone else should wrong him; rather that, morally speaking, it is better for
him, or from his point of view, to be a victim than a villain.5
Such an attitude respects the integrity of individual agency and is central
and essential to morality as we know it. However, it may root this thought
deeper in human nature if we view it as the ethical transformation of a basic
truth about human responses. Richard Wollheim introduces the idea that
self-concern is basic, and not derivative from other concerns, by imagining
the following alternatives:
1. If man endues death in order to escape greater pains, this is not courage
(EE III.1 1229b12–14). Of course, different goals are appropriate to differ-
ent virtues. The health that is a goal of temperance (NE III.11 1119a16) is
doubtless the agent’s own; by contrast, the liberal are loved for being useful
(IV.1 1120a22), evidently to others.
2. I do not here discuss the debated lines that follow (1097b16–20); on them,
see my Virtue and Reason in Plato and Aristotle, 52–53.
3. It may appear a further element of egocentricity that this pleasure is taken
in one’s own activity and not, say, in the benefit to a recipient. However,
this is inevitable since the pleasure is a form of enjoyment, and I cannot
enjoy another’s sense of relief in being saved from danger (though I might
enjoy perceiving it). However, it would be wrong to infer that I lack the
attitudes toward his well-being that one would expect of someone willing to
risk his life on another’s behalf. Aristotle recognizes a sympathetic pleasure
or pain (sunchairein, sunalgein, IX.4 1166a7–8, IX.10 1171a6–8, 29–30).
He disparages men who resemble women in liking to receive sympathy (X.11
1171b10–11) but respects the attitude of a mother separated from her chil-
dren for whom it is a sufficient consolation to see them flourish, even if she
is unknown to them (VIII.8 1159a28–33).
4. I say a little in Price, op. cit., 68.
5. These things are illuminatingly discussed in Müller, ‘Radical Subjectivity:
Morality versus Utilitarianism’, 115–132.
6. Wollheim, The Thread of Life, 237.
7. See his Anarchy, State, and Utopia (Oxford: Basil Blackwell, 1974), 28–35.
8. The congratulatory self-awareness of the megalopsychos (IV.3) is not some-
thing that modern Aristotelians would wish to defend.
REFERENCES
Primary
Plato Apology
Crito
Gorgias
Symposium
Aristotle De Anima (DA)
De Motu Animalium (DMA)
Eudemian Ethics (EE)
Nicomachean Ethics (NE)
Secondary
Müller, A. W., ‘Radical Subjectivity: Morality versus Utilitarianism’, Ratio 19 (1977),
115–132.
Nozick, R., Anarchy, State, and Utopia (Oxford: Basil Blackwell, 1974).
Price, A. W., Virtue and Reason in Plato and Aristotle (Oxford: Clarendon Press,
2011).
Wollheim, R., The Thread of Life (Cambridge: Cambridge University Press, 1984).
3 The Benefit of Virtue
Christoph Halbig
II
After this much too brief look on the two relata of the connection of virtue and
happiness, let us now proceed to the analysis of the connection itself. I would
like to distinguish between five kinds of such a connection in decreasing order
of strength:
(i) Being overall virtuous is both necessary and sufficient for the happiness
of the virtuous person.
40 Christoph Halbig
(ii) Being overall virtuous is a necessary condition on other goods being
of prudential value for their possessor and thus contributing to her
happiness.
(iii) Being overall virtuous is itself a basic kind of prudential value.
(iv) Possessing some virtue or other is a necessary condition on other
goods being of prudential value for their possessor.
(v) Being overall virtuous is a contingent, personal project whose suc-
cessful realization is of prudential value for its possessor.
Let me make some comments on each of them in turn. Both the strongest
and the weakest kind of connection can be dismissed quite swiftly:
The weakest one (v) would altogether trivialize the connection between
virtue and happiness—virtue would come out as just one project among
others that could as well be replaced with the pursuit of vice. Its prudential
value would not lie in itself but in its being the object of a personal project
whose successful realization would be rewarding as such and thus of pru-
dential value quite independent of its content.
The strongest one (i) is represented by the Stoic thesis that virtue is the only
kind of value in the strict sense of the word—other kinds of value like health
are at best so-called preferred indifferents. One might object at this point that
virtue as the only bearer of genuine value falls into the category of the good,
not into the category of the good for, and thus does not bear on the question of
prudential value at all. This, however, would hardly be consistent with the Stoic
claim that a life of virtue is a life that is ipso facto worth living—it is rewarding
for the person leading it, even if that person is bereft of all preferred indifferents
and is exposed to many of the dispreferred ones. Virtue would thus both be
necessary and sufficient for happiness; there can be no sacrifice in the service of
virtue because virtue is simply the only kind of intrinsic value in the game. The
Stoic theory of value, however, needs no separate discussion here since even the
somewhat weaker kind of connection (ii), which does grant genuine value to
other entities beside virtue, will, as we shall see now, not stand critical scrutiny.
The second but strongest (ii) kind of connection represents what in recent
debates is often labeled as a moralized conception of happiness. According to
it, no other goods (than virtue) like health, achievement, etc., are of prudential
value to their possessor if they come at the expense of virtue. And no other evil
(than vice) is of prudential disvalue to its possessor if it is suffered in following
the requirements of virtue. That implies that sacrifices required by virtue can-
not count as losses in happiness. In contemporary philosophy, such a position
has been defended, for instance, by D. Z. Phillips, who holds that ‘death for
the sake of justice is not a disaster’8 and that for the virtuous person who dies
for the sake of justice, death itself even turns into ‘a good’9; John McDowell
claims that the idea of happiness characteristic of the virtuous person
III
But it is not just the benefit side that is unlikely to figure prominently in
the deliberative perspective. Virtue-ethics is notoriously beset by the problem
of self-indulgence. If it is, as virtue-ethics holds, the virtues that provide the
48 Christoph Halbig
sources of reasons for action, then it should be expected that in deliberating
over what to do, we try to track those very sources of our reasons for action.
This, however, seems morally repugnant. A compassionate person should
genuinely care for the person in need, not for himself caring about such a per-
son qua possessing the virtue of, say, compassion; it is the former and not the
latter that should take center-stage in his deliberation.31 In order to avoid this
problem of self-indulgence, virtue-ethics might even be compelled to become
what Derek Parfit32 labels a self-effacing theory, thus positively discouraging
its adherents from deliberating in its own terms.33 Without being able to go
deeper into the problem of self-indulgence, it seems uncontroversial that the
virtues, even if of crucial importance for the deliberative stance of the agent,
should not be the focus of its content—thus turning the agent’s attention to his
own character in a morally questionable way. The person who suffers a loss
in the service of virtue should, in her own deliberating perspective, suffer that
loss not for the sake of her own virtue, but rather for the sake of demands of
the moral situation that the virtue in question is concerned with (in the case of
compassion it is concerned with the misery of one’s neighbor).
Both parts of the ‘does virtue benefit its possessor?’ question thus seem
alien to the first-person perspective of the deliberating agent. In that perspec-
tive, prudential aspects are considered only in part and under the constraints
mentioned previously, whereas virtue effaces itself by directing the agent’s
attention away from herself towards the demands of the situation that call for
her action.34 Looking back on her own past actions, the agent might of course
take stock in a third-person perspective: If she comes to the conclusion that
virtuous action regularly implies heavy losses in terms of her happiness, such
an observation might encourage her to rethink the normative relevance of vir-
tue on the one hand, happiness on the other, and the connection between both
of these terms. Or she might adopt a critical stance toward the contingent
social, political, or economical circumstances she finds herself in.
In this chapter, I have tried to disentangle the various ways in which
virtue does contribute to the benefit of its possessor. Losses and tragic dilem-
mas, however, remain possible. To what extent they become actual depends
on those contingent but also (at least in part) malleable circumstances. So,
notwithstanding the various levels pointed out in this chapter on which
virtue and happiness are connected, and notwithstanding the complex ways
in which these levels interact, it seems safe to state at least one thing: At each
historical moment it seems to take much virtue and considerable sacrifices
particularly from the virtuous to keep the virtuous life from becoming too
demanding in terms of happiness.35
NOTES
1. Cf. Nietzsche, Die fröhliche Wissenschaft, § 21, ‘An die Lehrer der Selbstlo-
sigkeit’.
2. Cf. Hursthouse, ‘Virtue Theory and Abortion’, 226.
The Benefit of Virtue 49
3. For instance Adams, A Theory of Virtue; Driver, Uneasy Virtue; Hurka, Virtue,
Vice and Value.
4. Cf. Slote, ‘The Virtue in Self-Interest’, 264f.
5. Cf. Hursthouse, On Virtue Ethics, 526. For a critical discussion of such an
approach cf. Copp and Sobel, ‘Morality and Virtue. An Assessment of Some
Recent Work in Virtue Ethics’, 526ff., and Zagzebski, ‘The Admirable Life
and the Desirable Life’, § 1.
6. Cf. Williams, Ethics and the Limits of Philosophy, 46.
7. On the dimension of depth cf. Foot, Natural Goodness, 86ff.
8. Phillips, ‘Does It Pay to be Good?’, 50.
9. Ibid., 51. See also ibid., 60, where Phillips claims that the ‘man who chooses
justice’, even if he does not profit as the rogue might well do in terms of
acquiring power, wealth, etc., ‘has accomplished all.’
10. McDowell, ‘The Role of Eudaimonia in Aristotle’s Ethics’, 369. In ‘Some Issues
in Aristotle’s Moral Psychology’, however, McDowell has added that eudai-
monia/happiness marks out just ‘one dimension of practical worthwileness’
(122). So a serious injury, for instance, incurred in the pursuit of virtue might
well turn out to be a genuine loss along some other dimension. Nevertheless,
McDowell leaves no doubt that the dimension of eudaimonia/happiness is
privileged with respects to the other dimensions—it marks out ‘excellence par
excellence’ (123), which in turn implies that for someone who has learnt to
appreciate that dimension ‘nothing else matters for the question what shape
one’s life should take here and now, even if the upshot is a life that is less desir-
able along other dimensions.’ In a similar vein, Müller, speaking of a ‘secret’
contained in the virtuous person’s attitude toward her happiness, denies that
the virtuous really believe that their goodness could actually deprive them of
their happiness, cf. Müller, Was taugt die Tugend?, 191.
11. McDowell, ‘The Role of Eudaimonia in Aristotle’s Ethics’, 369.
12. Philipps, ‘Does It Pay to be Good?’, 60.
13. Slote (‘The Virtue of Self-Interest’, 274), who is not in the least tempted by the
hang-tough strategy in the case of spoilt goods, at least implicitly subscribes
to what I have labeled the asymmetry-thesis by noting that such a strategy
would be ‘even more implausible’ in cases of prudential disvalues neutralized
or turned into prudential values by being required by virtue.
14. Cf. Aristotle, Nicomachean Ethics (NE) X.5, 1175b26–28.
15. Cf. Haybron, ‘Well-Being and Virtue’, 10, for a more detailed argument that
tries to prove the incompatibility of the no-loss-at-all thesis with the demands
of self-respect.
16. And even the deliberating agent had better be aware that there is such a
prudential disvalue involved, even if he immediately goes on to silence them
in terms of their normative force—otherwise the distinction between, for
instance, the courageous and the rash agent could hardly be made (the coura-
geous agent unlike the rash one is aware of the serious risks he exposes herself
to in acting virtuously). For McDowell’s theory of silencing see ‘Virtue and
Reason’, § 3 and ‘The Role of Eudaimonia in Aristotle’s Ethics’, 370.
17. Cf. Foot, Natural Goodness, 97, who argues that in such a situation, the
virtuous agent would not describe herself as sacrificing her happiness but as
realizing ‘that a happy life had turned out not to be possible for him’.
18. This point is made by Zagzebski, ‘The Admirable Life and the Desirable Life’,
65.
19. As does Foot loc. cit. (n. 35).
20. Cf. Aristotle, NE, 1153b19.
21. Such a general requirement is at least hinted at in Foot, Natural Goodness,
96, who holds that ‘humanity’s good can be thought of as happiness, and
50 Christoph Halbig
yet in such a way that combining it with wickedness is a priori ruled out’.
(Although some paragraphs earlier Foot allows the combination of a wicked
character and a happy life as at least a conceptual possibility, ibid. 92.)
22. Cf. Dancy, Ethics Without Principles, 184–187.
23. For the possibility of such trade-offs between virtues and other prudential
goods that might lead parents to recommend ‘a mixed virtuousness’ to their
children see Copp and Sobel, ‘Morality and Virtue. An Assessment of Some
Recent Work in Virtue Ethics’, 527ff.
24. For an ‘argument from lack of sympathy’ (for the vicious person who has
exactly the same amount of prudential values as the virtuous person, with the
exception of virtue itself) to a rejection of virtue as a fundamental category of
prudential value see Hooker, ‘Does Moral Virtue Constitute a Benefit to the
Agent’, 149–155.
25. For a discussion of the dimension of depth see Foot, Natural Goodness, 86ff.
26. At the heart of Michael Slote’s so-called Platonic elevationism lies the thesis that
such a connection can be shown not just for some elements, but indeed for ‘every
element of human well-being’ [my emphasis, C.H.]. In case the theory of the
unity of the virtues holds true, Slote’s Platonic elevationism would of course
collapse into what he labels Aristotelian elevationism, that is, the thesis ‘that all
elements of personal well-being must be compatible with virtue taken as a whole’
(Slote, ‘The Virtue in Self-Interest’, 274)—since the individual virtues required
for some element of well-being would then in their turn presuppose all the others.
27. Cf. Slote, ‘The Virtue in Self-Interest’, 276f.
28. For a more detailed discussion of such a line of argument that tries to show
that the virtues make a constitutive contribution to moral knowledge as a
crucial dimension of the fundamental prudential value of knowledge, see
Hooker, ‘Does Moral Virtue Constitute a Benefit to the Agent’, 146f.
29. Cf. Slote, ‘The Virtue in Self-Interest’, 277f., who holds that ‘knowledge con-
stitutes a distinctive form of personal good, and counts as wisdom, only when
it takes courage to acquire it’. Ibid., 278.
30. For this distinction see Haybron, ‘Well-Being and Virtue’, 20. Ibid., 19ff.
Haybron makes some interesting observations on how Aristotle’s point of
departure from a first-person, goal-setting perspective might have come in
the way of developing an independent theory of well-being.
31. This problem, however, does not seem to beset all the virtues in the same
way; it seems that someone could act justly for the very reason of being just,
whereas it seems unlikely that someone could act compassionately for the
reason of being a compassionate person (he should definitely not care about
himself ). Cf. Williams, Ethics and the Limits of Philosophy, 10f.
32. Cf. Parfit, Reasons and Persons, 24.
33. For a discussion of the self-indulgence problem and its possible consequences
for virtue-ethics cf. Hurka, Virtue, Vice and Value, 246ff.; Cox, ‘Agent-based
Theories of Right Action’; Keller, ‘Virtue Ethics is Self-Effacing’.
34. The person and her character might of course be included in that situation.
35. Be it only to keep the virtuous life attractive for those who are still on the way
toward virtue and who might be deterred by circumstances that would make
the practice of virtue ipso facto heroic (by, for instance, severely punishing
acts of basic decency).
REFERENCES
Adams, Robert M., A Theory of Virtue (Oxford: Oxford University Press, 2006).
Aristotle, Nicomachean Ethics.
The Benefit of Virtue 51
Copp, David, and David Sobel, ‘Morality and Virtue. An Assessment of Some Recent
Work in Virtue Ethics’, Ethics 114 (2004), 514–554.
Cox, Damian, ‘Agent-based Theories of Right Action’, Ethical Theory and Moral
Practice 9 (2006), 505–515.
Dancy, Jonathan, Ethics Without Principles (Oxford: Oxford University Press, 2004).
Driver, Julia, Uneasy Virtue (Cambridge: Cambridge University Press, 2001).
Foot, Philippa, Natural Goodness (Oxford: Oxford University Press, 2001).
Haybron, Daniel M., ‘Well-Being and Virtue’, Journal of ethics and social philosophy
2 (2007), 1–27.
Hooker, Brad, ‘Does Moral Virtue Constitute a Benefit to the Agent’, in How Should
One Live?, ed. R. Crisp (Oxford: Clarendon Press, 1996), 141–155.
Hurka, Thomas, Virtue, Vice and Value (New York: Oxford University Press, 2001).
Hursthouse, Rosalind, On Virtue Ethics (New York: Oxford University Press, 1999).
———, ‘Virtue Theory and Abortion’, Philosophy and Public Affairs 20 (1991),
223–246.
Keller, Simon, ‘Virtue Ethics is Self-Effacing’, Australasian Journal of Philosophy 85
(2007), 221–231.
McDowell, John, ‘The Role of Eudaimonia in Aristotle’s Ethics’, in Essays on Aris-
totle’s Ethics, ed. A. O. Rorty (Berkeley: University of California Press, 1981),
359–376.
———, ‘Some Issues in Aristotle’s Moral Psychology’, in Ethics (Companions to
Ancient Thought, IV), ed. St. Everson (Cambridge: Cambridge University Press,
1998), 107–128.
———, ‘Virtue and Reason’, Monist 62 (1979), 331–350.
Müller, Anselm Winfried, Was taugt die Tugend? (Stuttgart: W. Kohlhammer Verlag,
1998).
Nietzsche, Friedrich, Die fröhliche Wissenschaft, in Kritische Studienausgabe Bd. 3,
1887.
Parfit, Derek, Reasons and Persons (Oxford: Oxford University Press, 1984).
Phillips, D. Z., ‘Does It Pay to be Good?’, Proceedings of the Aristotelian Society 65
(1964), 45–60.
Slote, Michael, ‘The Virtue in Self-Interest’, Social Philosophy and Policy 14 (1997),
264–285.
Williams, Bernard, Ethics and the Limits of Philosophy (London: Fontana 1985, 3ed.).
Zagzebski, Linda T., ‘The Admirable Life and the Desirable Life’, in Values and Vir-
tues, ed. T. Chappell (Oxford: Clarendon Press, 2006), 53–66.
4 Well-Being and Eudaimonia
A Reply to Haybron
Mark LeBar and Daniel Russell
CONCLUSION
NOTES
REFERENCES
Haybron, Daniel, ‘Happiness, the Self, and Human Flourishing’, Utilitas 20 (2008),
21–49.
———, The Pursuit of Unhappiness (New York: Oxford University Press, 2008).
———, ‘Well-Being and Virtue’, Journal of Ethics and Social Philosophy 2 (2007),
1–27.
Hursthouse, Rosalind, Beginning Lives (New York: Blackwell, 1987).
———, On Virtue Ethics (New York: Oxford University Press, 1999).
Kraut, Richard, ‘Two Conceptions of Happiness’, The Philosophical Review 88
(1979), 167–197.
LeBar, Mark, ‘Good for You’, Pacific Philosophical Quarterly 85 (2004), 195–217.
Ridge, Michael, ‘Reasons for Action: Agent-neutral vs. Agent-relative’, in Stanford
Encyclopedia of Philosophy, ed. Edward N. Zalta (Fall 2008 Edition), http://
plato.stanford.edu/archives/fall2008/entries/reasons-agent/.
Sumner, L. W., Welfare, Happiness, and Ethics (New York: Oxford University Press,
1996).
5 Virtue, Personal Good, and
the Silencing of Reasons
Julia Peters
II
(P2) If one misses something that one has no reason to pursue, then that
is no loss.
The virtuous person’s reasons for pursuing the personal good have been
silenced by virtuous considerations. Accordingly, she has no reason to pur-
sue the good in question. In missing something she has no reason to pursue,
she suffers no loss. It follows that she suffers no loss in forgoing the personal
good in question.9
Before looking at the premises of McDowell’s argument in more detail,
we need to consider how exactly the argument can be directed against the
challenge sketched previously. The core claim of the challenge is that the
practice of virtue may demand sacrifices from the virtuous person, such
72 Julia Peters
that virtuous action may deprive the virtuous person of personal goods. In
this way, the practice of virtue may undermine the virtuous person’s happi-
ness. In contrast, McDowell’s argument attempts to show that where it may
appear that a virtuous person endures a loss when forgoing a personal good
for the sake of virtue, this is really not so: her reasons for pursuing the good
have been silenced, that is to say, she has no reason to pursue it; when she
misses something she has no reason to pursue, that constitutes no loss, no
personal sacrifice for her. Forgoing personal goods for the sake of virtue is
therefore not really painful for the virtuous person, or does not undermine
her happiness.
What, then, can be said about the premises of the argument? I take it that
one important consideration underlying P1, the claim that reasons against
acting virtuously are silenced rather than outweighed from the virtuous per-
son’s point of view, is the following one. Unless we assume that for the vir-
tuous person, considerations opposing a requirement of virtue are silenced,
rather than outweighed, it would in certain extreme cases be impossible to
explain why people act virtuously at all. Consider again the courageous sol-
dier on the battlefield. If he was to weigh personal gains against losses, how
could he possibly come to decide that the loss ensuing from his action—the
loss of his life—is outweighed by the gain he expects from it? Or, similarly,
how could the just person hope to outweigh the loss of her life by acting
justly? McDowell’s point is that as long as we assume that the virtuous per-
son registers potential goods forgone for the sake of virtue on the loss-side of
a balance, we will not be able to account for her decision to embark on the
virtuous course of action in extreme cases. Virtuous action in extreme cases
can only be explained if reasons against acting virtuously are not expected to
be outweighed, but are simply silenced.10 On this view, then, considerations
of virtue are not so much understood as giving rise to certain reasons that
can then be weighed against other reasons, but rather as putting a constraint
on what can count as a reason in the first place.
A second, ancillary thought underlying P1 may be that the notion of
silencing is needed in order to draw a distinction between two types of moral
agent, one of which is more excellent—deserves greater moral praise—than
the other: the virtuous and the merely continent agent. Both the virtuous
and the continent moral agent end up doing the same thing, but the structure
of the practical consideration preceding their choice of action is significantly
different in each case. Consider again the soldier on the battlefield. One
important mark of his virtue—his courage—is that he is not tempted by the
possibility of saving his life by acting cowardly: running away and abandon-
ing his fellow soldiers is simply not an option for him. Accordingly, he stays
on to fight. But he might also have reached the decision to stay and fight in
a different, less straightforward way: he might have weighed the reasons for
and against staying to fight, and he might have been seriously tempted to
run away and save his life—while nevertheless deciding that in the end, he
has stronger reasons to stay. One might say that in this case, his decision to
Virtue, Personal Good, and the Silencing of Reasons 73
stay is preceded by a period of inner division: he is torn between different
options. If this was the case, he would not be truly virtuous, but merely con-
tinent. While reaching precisely the same practical decision as the continent
agent, the virtuous agent knows no such inner division, and this constitutes
part of his moral excellence, over and above the fact that he also ends up
doing the right thing. The notion of silencing captures precisely this differ-
ence in practical deliberation between the two types of moral agent: the con-
tinent agent takes (prudential) reasons that speak against acting virtuously
into consideration and is torn between different courses of action; for the
virtuous agent, in contrast, countervailing reasons are silenced.11
The thought underlying P2, the thesis that if one misses something one
has no reason to pursue, one does not suffer a loss, is a claim about the
relation between goods and reasons. This becomes more obvious when one
focuses on the reverse of P2: As long as one suffers a loss when missing or
forgoing something, one also has a reason to pursue it. In other words, as
long as something constitutes a good, such that losing it has to be considered
a genuine loss, one also has a reason to pursue it. Accordingly, where one has
no reason to pursue something—where one’s reasons to pursue it have been
altogether silenced—it no longer constitutes a good, its goodness must have
been ‘cancelled’. For the virtuous person, then, the goodness of a supposed
personal good is conditional on its being in accord with considerations of
virtue. Something can be a genuine personal good for her only insofar as its
pursuit does not flout considerations of virtue.12
In spite of the initial plausibility of its premises, however, the conclusion
of this argument appears problematic. Consider again the case of the soldier.
Even if McDowell is right in pointing out that the truly courageous soldier
will not consider himself as having any reason for saving his life instead
of staying on the battlefield to fight for his city, and even if the fact that
he has no reason ‘cancels’ the goodness that his life has for him, somehow
the intuition lingers that the soldier nevertheless suffers a loss or makes a
sacrifice in giving his life for his city. This intuition can be articulated more
accurately by the following consideration. For all his willingness to give
his life in fighting for his city, the soldier would certainly prefer a world in
which he could have both, save his life and fight for the city, to the actual
one, where the two are incompatible. For it is not the case that the soldier
does not in general value his life; rather, it is merely under the present cir-
cumstances, the actual world being as it is, that the value of preserving it
has been cancelled for him. But it is not clear how McDowell’s argument
can account for this intuition and explain the soldier’s hypothetical prefer-
ence for a world in which he could both save his life and fight for his city.
According to McDowell’s account, the soldier loses nothing when forgoing
the opportunity to save himself from personal harm by running away. Hence
there seems to be no reason why he should prefer a world in which he could
have both—both escape personal harm and fight for his city—to the actual
one. If we accept McDowell’s argument, then, it seems to imply that the
74 Julia Peters
virtuous person necessarily has to have a kind of fatalist attitude: whether
the exercise of his virtue conflicts with his pursuit of personal goods or not
makes no difference to him, for even if he has to forgo a personal good in
the name of virtue, this constitutes no loss for him.
McDowell’s argument presents us with a problem: while its premises—or
the considerations underlying them—appear plausible, its conclusion nev-
ertheless seems to violate a strong intuition we have regarding the losses
potentially suffered by a virtuous person as she acts in the name of virtue. In
the remainder of this chapter, I want to suggest a way to solve this problem,
by showing how it is possible both to do justice to the intuition just sketched
and to accept, in essence, McDowell’s argument. On the whole, my sugges-
tion aims not so much at criticizing McDowell’s argument, but rather at
showing how it can be reconciled with countervailing intuitions. My overall
conclusion will be that McDowell’s argument is successful in offering a reply
to the challenge sketched in the introduction: it is successful in establish-
ing that the exercise of virtue never makes the virtuous person unhappy.
Nevertheless, the crucial point I shall try to make is that it does not follow
from this that a virtuous person does not suffer any losses.
III
What I have in mind can be best introduced by pointing out that there are
(at least) two different ways in which the pursuit of a personal good can
conflict with a requirement of virtue. Consider an example involving the
virtue of temperance: A temperate person is invited by her friends to go out
binge drinking with them. As she is temperate, she of course declines the
invitation. McDowell’s claim that the virtuous person has no reason what-
soever to pursue the personal good in question seems very plausible here. If
the person in the example was seriously tempted by the opportunity to go
binge drinking, and had to weigh reasons in favor of and against seizing this
opportunity, we would be hesitant to call her temperate, even if in the end
she abstained and declined the opportunity—at best, she could be consid-
ered continent in this case. For the truly temperate person, an opportunity to
go binge drinking does not constitute a good whose loss has to be weighed
against the reasons speaking in favor of declining the invitation. Accord-
ingly, the temperate person has no reason to accept her friends’ invitation
and seize the opportunity to go binge drinking.
At first sight, this example may appear to be parallel in all important
respects to the one of the courageous soldier. Like the temperate person,
the courageous soldier has no reason to run away and save his life instead
of fighting for his city. His life does not constitute a good for him, as it can
only be preserved by acting cowardly. Similarly, for the temperate person,
going out binge drinking does not constitute a good, as it involves behaving
intemperately. However, in spite of their similarity, there is an important dif-
Virtue, Personal Good, and the Silencing of Reasons 75
ference between the two cases. To go binge drinking means to go drinking
with the explicit intent of drinking excessively, that is, intemperately. Hence
binge drinking is an inherently intemperate activity—there is no such thing
as practicing binge drinking in a temperate way. Accordingly, the temperate
person never has a reason to go out binge drinking, for doing so is always
in conflict with considerations of virtue. It follows, according to P2, that an
opportunity to go binge drinking does not constitute a good for the temper-
ate person under any circumstances. In contrast, for the courageous soldier,
his own life is usually of great value, and he usually has a reason to preserve
it. It is just that in the situation he is in, as he can only preserve his life by
acting cowardly, that his reasons for preserving it are silenced, such that his
life no longer constitutes a good for him. Only under these specific circum-
stances, in which his reasons for preserving his life have been silenced, does
it no longer constitute a good for him.
I want to suggest that it is crucial to pay attention to the difference
between these two examples, for this difference holds the key for reconcil-
ing McDowell’s argument with the intuitions mentioned previously about
losses suffered by the virtuous person.
It seems correct that with regard to the temperate person who forgoes an
opportunity to go binge drinking, it does not make sense to speak of her suf-
fering a loss or making a sacrifice. If she is temperate, then not going binge
drinking is not a sacrifice for her. We saw previously that the reason why one
would be inclined to hold that even the courageous soldier suffers a genuine
loss in giving his life for his city is that he would certainly prefer circum-
stances in which he could preserve his life and fight for his city to the actual
ones, in which he can only have the latter. But the same consideration does
not apply in the example of the temperate person. Since there are no cir-
cumstances under which she can both go binge drinking and be temperate—
because binge drinking is intemperate under any circumstances—it makes
no sense to hold that she would prefer such circumstances. In this case, then,
there is no lingering intuition that even though the virtuous person has no
reason to pursue a certain good, she may nevertheless be suffering a loss.
In contrast, in the case of the courageous soldier, the fact that his life does
not constitute a good worth pursuing for him is due to the circumstances
he finds himself in. But this makes it possible to distinguish between two
different senses in which he can be said to suffer a loss. On the one hand,
he can be said to suffer a loss in failing to pursue something that constitutes
a good in the situation he is in. According to McDowell’s thesis P1, the
courageous soldier does not suffer a loss in this sense because his life, which
he fails to preserve, does not constitute a good for him in his situation. It is
not his failing to pursue something that constitutes a good in his situation,
then, that makes the soldier suffer a loss. But on the other hand, the soldier
can be said to suffer a loss in being deprived by the circumstances of the
goodness of something that would usually, under different circumstances,
constitute a good for him. If the situation was different, if the circumstances
76 Julia Peters
were more fortunate, his life would constitute a great good for the soldier. As
the circumstances are, however, his life has lost its goodness for him, or the
goodness of his life has been cancelled. But this must certainly be considered
as a loss by him: something that would usually be of great value has lost its
value, or has been made unavailable as a good. His life has been deprived
of goodness.
More generally, where it depends on the circumstances whether the virtu-
ous person’s reasons for pursuing a certain good are silenced or not, there is
always room for her to suffer the kind of loss just described with regard to
the case of the courageous soldier. If her reasons to pursue the good in ques-
tion are silenced, then it constitutes no loss for her not to pursue the good.
But insofar as the fact that the good in question is no longer a good for her
is due to unfortunate circumstances, it does constitute a loss for her that the
circumstances have deprived something that usually constitutes a good for
her of its goodness, or that they have made unavailable as a good what is
usually a good for her.
One can also express this thought in terms of the regret the virtuous
person may feel and the reasons she has for feeling it. Where the virtuous
person’s reasons for pursuing a good are silenced, she has no reason to regret
not pursuing the good because it does not constitute a good for her (or, in
hindsight, she has no reason to regret not having pursued the good because
it did not constitute a good for her). But if it is due to unlucky circum-
stances that her reasons for pursuing the good are silenced, she has reason
to regret that the circumstances are as they are (or that they were as they
were). Because this is what makes the good in question unavailable to her as
a good, by depriving it of its goodness. Thus the virtuous person’s regret is
not directed at what she herself does or did under certain circumstances, but
at the circumstances themselves. She experiences her loss as being incurred,
not by herself or her own action, but by the circumstances—or more gran-
diosely, by fate.
This makes a crucial difference. Because it means that from the point of
view of the virtuous person, it is not her virtuous action that constitutes the
cause or ground of her loss, but the circumstances. It is not her virtuous
action that undermines her happiness or well-being, in short, but fate.13
IV
We can now come back to the overall theme of the relation between virtue
and eudaimonia. As sketched in the beginning, the thesis that the possession
and practice of virtue is in the best interest of the virtuous looks dubious in
light of the apparently undeniable fact that the virtuous may have to sacri-
fice personal goods in acting virtuously. I presented McDowell’s argument
as implying a response to this challenge. McDowell argues that the virtuous
person, when missing or forgoing a personal good in acting virtuously, does
Virtue, Personal Good, and the Silencing of Reasons 77
not make a genuine sacrifice and does not suffer a genuine loss. For what
she misses or forgoes does not constitute a genuine good for her. However,
the conclusion of McDowell’s argument contradicts the strong intuition that
there is a sense in which the virtuous person, after all, does suffer a loss when
she gives up a personal good in acting virtuously. Otherwise, we could not
make sense of the fact that whenever possible, the virtuous person would
prefer to be able to both act virtuously and pursue the personal good in
question.
In light of the previous discussion, we can now see how this tension can
be dissolved. Giving up a personal good in acting virtuously, the virtuous
person does not suffer a loss in the sense of making a sacrifice, that is, of
depriving herself of a personal good that she could have secured for herself
in the situation she is in, or that would have been available as a good in her
situation. However, she does suffer a loss in the sense of being deprived of
a good by the circumstances: something that usually constitutes a good for
her has been deprived of its goodness or has been made unavailable as a
good. The fact that she has been deprived of a good by the circumstances
gives her a reason for regret. Accordingly, while she would not prefer act-
ing or having acted in a different way than she acts or has acted—namely,
non-virtuously—she would prefer a world in which it was possible for her
to both act virtuously and pursue or secure the personal good in question.
Hence she is precisely not a ‘fatalist’: while she does not regret acting or hav-
ing acted virtuously, she is susceptible to regret directed at the circumstances
in which she finds herself.
McDowell’s argument thus implies a subtle way of refuting the challenge
discussed in the beginning. The argument shows that for the virtuous per-
son, the practice of virtue is never the ground or cause of her unhappiness
or loss; acting virtuously is never what deprives her of a personal good, or
what makes her suffer a loss. In this sense, it is true that she does not suffer
any losses by or through acting virtuously. However, it does not follow that
the virtuous person suffers no loss at all as long as she acts virtuously.14 If
the pursuit of some personal good conflicts with a requirement of virtue in a
certain situation, and the virtuous person forgoes the personal good in that
situation, she does not sacrifice or give up anything that constitutes a good
in the situation she is in. But she is nevertheless being deprived of a good,
and in this sense she suffers a loss.
It is important to bear in mind that the stronger thesis that the virtu-
ous person literally suffers no loss at all cannot be established by means of
the argument alone discussed previously. For McDowell is unfortunately
often read as intending to argue for this stronger conclusion. According to
Rosalind Hursthouse, for instance, McDowell holds that for the virtuous
person, no personal good missed or forgone in meeting a requirement of
virtue counts as a genuine loss because her own virtuous action is all that is
of value to the virtuous person, or at least its value is so much more signifi-
cant to her than that of any personal good that no ‘sacrifice’ of such a good
78 Julia Peters
for the sake of acting virtuously really counts as a sacrifice.15 Hursthouse
argues against this position, pointing out that it only looks plausible—if at
all—in a very limited number of cases. These are cases in which the virtuous
person sustains the ‘loss’ of a personal good as she performs a noble, admi-
rable action—for instance, sacrificing her life for a worthy cause. Here one
might think that if only the cause is great and worthy enough, any personal
losses endured in furthering it may appear insignificant. By contrast, Hurst-
house argues, McDowell’s thesis looks less plausible with regard to cases
in which the virtuous action consists merely in the avoidance of something
base, rather than in the performance of a noble deed: for instance, ‘dying not
to serve some noble cause but only because you have fallen into the hands
of a mad tyrant and, despite his threats, refused to do something wicked’.16
Hursthouse’s thought seems to be that where losses are endured in the
accomplishment of something noble or great, we can somehow make sense
of the notion that from the point of view of the virtuous agent, the losses
are compensated for by the good accomplished through virtuous action.
But this does not seem to be possible in the case where losses are endured
merely for the avoidance of something bad: for here nothing noble or great
is accomplished that could outweigh the losses.
However, this is an unfair reading of McDowell’s position. As I argued
previously, one of the thoughts underlying McDowell’s thesis about the
silencing of reasons, P1, is that if we took the virtuous person to weigh
reasons for and against acting virtuously in cases where virtue and personal
good conflict, this would sometimes make it impossible to explain why she
ends up acting virtuously. On this view, the virtuous person would con-
cede that she is suffering a personal loss in acting virtuously, but hold that
this loss is compensated for by the good attained through virtuous action.
‘In suitably described cases’, McDowell writes, ‘any such claim would be
implausible to the point of being fantastic’.17 Among such cases, presum-
ably, are those in which nothing great or noble is to be attained through
virtuous action but merely something base to be avoided. To escape such
implausibility, McDowell suggests, we should instead assume that the per-
sonal goods in question cease to be personal goods for the virtuous person
when they are seen to conflict with virtue, such that she does not consider
herself as having any reason to pursue them. From McDowell’s point of
view, then, it should make no difference to the virtuous person whether she
forgoes a personal good in order to virtuously accomplish something noble,
or in order to virtuously avoid something base. In both cases, the good
in question simply ceases to be a good since its pursuit is in conflict with
considerations of virtue, hence it is no question for her how its loss may be
compensated for.
Nevertheless, there is an important intuition underlying Hursthouse’s
complaint. In a case such as the one involving the mad tyrant, we are inclined
to insist that there is some loss that is endured by the virtuous person as she
forgoes the personal good in question, or that she has a reason for regret.
Virtue, Personal Good, and the Silencing of Reasons 79
After all, would she not prefer circumstances to have been different—would
she not prefer not to have fallen into the tyrant’s hands—such that she could
have saved her own life and acted virtuously? In contrast, one might think
that there may be cases in which a virtuous person even welcomes an oppor-
tunity to sustain a personal loss in acting for a great, noble cause. However,
it is possible to do justice to this intuition in the way suggested previously.
If a mad tyrant confronts the virtuous person with a choice between either
dying or doing something wicked, the virtuous person will choose to die.
Her life, saved through a wicked action, would have no value for her, hence
she suffers no loss in not preserving it by acting wickedly. But she neverthe-
less suffers a loss, and has a reason for regret, since she is being deprived, by
the circumstances, of some good: the circumstances are such that her own
life has lost its goodness for her. For her, her loss is not induced by her virtu-
ous action, but rather brought about by the circumstances.18
The strong thesis that, for the virtuous person, conflicts between virtue
and personal good result in no personal loss at all could only be established
on the basis of an assumption to the effect that the only thing that is of value
for the virtuous person is her own virtuous or non-virtuous action. If this
was the case, it would be true that the virtuous person was safe from suffer-
ing any losses at all as long as she only acted virtuously: the only thing that
could potentially constitute a loss for her was her failure to act virtuously.
But this is not McDowell’s view, at least not according to the argument
discussed previously. This argument establishes a more modest claim: The
virtuous person does care about her own personal good, in addition to her
virtue or virtuous action. But from her point of view, the loss of a personal
good is never suffered as a result of or due to her virtuous action. Rather, she
experiences such losses as imposed on her by the circumstances. Hence it is
the way in which they experience such losses, rather than the fact that they
do not experience them at all, that is distinctive of the virtuous.19
NOTES
REFERENCES
How can Hursthouse reject the thought that nature determines how
humans should be yet think that the same considerations that grounded
the four ends in plants and animals also ground the normative status of
the four ends in humans? She gives no new arguments to support such
a status for the four ends in the case of humans.7
Copp and Sobel see Hursthouse’s argument as incomplete at best, and they
are skeptical of anyone’s ability to complete it. Indeed, there is a missing com-
ponent to this account of ethical naturalism, but it can be filled in, and was,
Human Nature, Virtue, and Rationality 85
at least in part, by Philippa Foot. The crucial missing piece is that the appeal
to human nature serves not only to tell us that we are rational, but also to
define what it is to reason well. Neo-Aristotelian ethical naturalism should be
understood as a thesis about rationality, according to which practical rational-
ity is species-relative. Our reasoning cannot ignore what we need as human
beings and yet still claim to exhibit practical rationality. Hence, Hursthouse
can claim that as rational animals we are freed from a certain kind of obedi-
ence to nature, while maintaining that nature has some normative role for
us; nature is normative over our reasoning, but not directly over our action.
When Foot states that human beings go for what we see as good rather than
the good that we see, she adds that what we see as good is inevitably informed
by a conception of our form of life. Making that conception explicit and
subjecting it to criticism is an essential part of moral reform, for ethical natu-
ralists. When Hursthouse’s claims about impersonal benevolence are placed
against the background of a proper understanding of ethical naturalism, a
version of her argument against impersonal benevolence can go through, or so
I will argue here. I will first argue for a way of conceiving human nature that
is crucial to the interpretation of ethical naturalism defended here.
The idea of human nature may enter our practical reasoning explicitly in
the form of a major premise, such as ‘It befits a human to overcome fear
for the sake of worthy goals’, or something as general as ‘It befits a human
to attempt to further the survival of the human species’. Yet one might
wonder what recommends such premises as starting points for practical
reasoning. After all, if my identity is to enter into my reasoning at all, per-
haps I would rather appeal to what befits a Buddhist or an American since
these identities might mean more to me than my identity as human, which
seems, after all, to be rather abstract and thin.
Neo-Aristotelian ethical naturalists are not assuming that our reason-
ing must or should appeal to being human in this way. Instead, the view
is that there is a sense in which we inevitably make an implicit appeal to
some understanding of human nature inasmuch as we reflect and act. This
understanding of human nature may be informed by scientific ideas that we
have about human beings, but in principle it is separate from these ideas
and instead shared with others who diverge from us in their scientific views
on anthropology. It is a practical conception of human beings that we must
employ in order to understand ourselves or anyone else as engaged in an
action. Further, this conception of human nature supplies standards of eval-
uation or natural norms, including norms of conduct.
Actions are done by living things, and so understanding a human action,
on the naturalist’s view, is a matter of understanding a particular life form.
As Michael Thompson argues, all living things exhibit, in various ways, a
86 John Hacker-Wright
special kind of agency.8 The growth of a fern is essentially different from
the growth of a puddle of rainwater or a trash heap because it is the fern
itself that brings about its own growth by a process of cell division. We can
provide a description of how such a process is supposed to advance and
recognize when it has gone awry, whereas there is no question of the growth
of a trash heap going wrong. Further, the growth of a fern differs from that
of a rhododendron, even though they both exhibit the same sort of agency
that is characteristic of living things. We must, in principle, be able to discern
the growth of the fern from a growth on the fern that is due to blight, for
example, and such discernment requires a conception of how life is sup-
posed to go for the fern. To characterize something as possessing the agency
distinctive of living things requires bringing to light norms for the life form
of that individual; this is no less true in our own case. Yet the case of human
beings differs in that we obviously stand in a different relationship to our
own life form. With other life forms our views are based entirely on obser-
vation. In the case of the human life form, we must apply an understand-
ing of it whenever we describe ourselves. Hence, our understanding of the
human life form is, in part, internal and not based on observation; I know
something about the category ‘human being’ from my own case because it is
in the background of the fact that I think and act. In this way, the category
‘human being’ is not merely used to gain an understanding of part of the
world, but rather it is a practical concept.
Consider a simple case of action: sawing a plank of wood. There are facts
I must know and practical competencies I need in order to intentionally
undertake this action. Among the facts I need to know are that by perform-
ing such and such actions, I will, if everything goes well, be sawing a plank
of wood. Many such facts I have gathered by watching others perform simi-
lar actions, as well as through trial and error. But I must possess even more
basic knowledge inasmuch as I have seen others perform such actions and
am able to know that I am doing the same. To understand an intentional act
as such is to understand a living thing as being engaged in a specific form of
agency that involves responsiveness to reasons. This is not something that
could simply appear in a ‘rogue individual’, as Thompson puts it; the capac-
ity to respond to reasons to act must characterize my life form.9 To see this,
consider a simpler case like that of eating. For an organism to be regarded
as eating something, one must take it to be ingesting that which is normally
nutritious for its kind as well as absorbing it in such a way as to derive
nutrition from it. Otherwise, we could not say that we saw an organism eat-
ing something, but instead we would be watching a fortuitous occurrence
whereby an organism took in some sort of material that happened to further
its life. Likewise, to see an organism as engaged in an intentional action, we
must posit a non-accidental concurrence of a specific kind of mental event
that consists of forming an aim followed by an initiation of movement such
that, if all goes well, I will bring something about related to fulfilling that
aim. Further, to intentionally undertake an action, I must realize that I am
Human Nature, Virtue, and Rationality 87
doing so. As Anscombe argues, if I am asked why I am sawing a plank of
wood, and my reply is that I did not realize that I was doing so, say because
I was instead simply enjoying the rhythm of the sound produced when the
saw is placed against the wood, then I was not sawing intentionally.10 It must
be a norm for that life form to engage in such actions, otherwise, there will
be a mere coincidence of mental events and bodily jerks.
Action implies a sort of insight into human nature inasmuch as in acting I
implicitly posit that it is normal to initiate movements in order to undertake
the fulfillment of an adopted aim; in other words, in undertaking any action,
we posit that we are members of a life form that normally possesses rational
agency. This bit of knowledge concerning our life form is discerned from
undertaking an action, rather than from observation. Indeed, we could never
observe an action without first possessing this knowledge about our life form.
For we must see someone as setting out to do something knowing what they
are about, and I cannot learn about this capacity from observation. Rather,
I must learn it from acting intentionally myself and take others to be doing
the same. Hence, to see other human beings as acting requires us to frame
them against the background of the same practical conception of the category
human to which anyone adverts whenever he acts. There is some normative
content to this aspect of our self-understanding since problems such as acting
on false premises, weakness of will, and slips are defects of actions, and we
must all regard them as such if we are rational agents. Presumably, if we take
no steps to avoid defective actions, we are defective agents.
This brings us to a conception of human nature that we all apply, at least
implicitly, in our actions. But of course we need not coincide completely in
our understanding of the category; there is surely a great deal more to being
a non-defective agent than simply taking steps to avoid acting on false prem-
ises or to prevent weakness of will. Since we act on reasons, our actions are
susceptible to an open-ended range of criticisms. Our conceptions of virtue
and vice capture some of the reasons to which an agent must respond in
order to attain non-defective agency. For example, generally one is seen as
deficient in charity when she fails to respond to a plea for help for the sake
of avoiding a mild inconvenience and delay. Does ethical naturalism help us
validate this judgment? The following sections will demonstrate that it does.
If the case made in the preceding section is correct, imputing actions to our-
selves or others places us against the background of our form of life. Now I
will argue that our judgments about proper conduct are likewise inevitably
framed as judgments with a certain practical generality, in that they charac-
terize our form of life.
As I pointed out previously, one may think that there are various ‘practi-
cal identities’ from which I could draw norms for my conduct, and even that
88 John Hacker-Wright
one could imagine norms applicable to oneself alone. So perhaps I could
have the thought, ‘This particularly self-serving action befits JHW’. In think-
ing this, I could hold that humans are characteristically charitable, while
being indifferent to this fact. In that case, I regard myself as a bad human
being but an excellent individual, that is, I am excellent as JHW. At least,
this appears to be an option, but is it really? Can I conceive of reasons that
are reasons only for me? The argument can be extended to show that I can-
not; the connection between the aim I hope to achieve and the movements I
initiate is only conceivable as acting on a reason against the background of
a norm for my form of life as shown previously. Whatever I am aiming at
must be intelligible as an aim, and the movement must have some intelligible
connection with the fulfillment of that aim in order for what I am doing to
be comprehensible as an action.
One might think that I can desire anything at all and have any sort of
belief pertaining to the fulfillment of that desire, and therefore anything
at all can be understood as an action given a certain condition of the indi-
vidual. But the identification of a desire designates a process within a living
thing. Again, to see an organism as having a desire one must employ some
understanding of that organism’s form of life. As Anscombe pointed out,
the primitive sign of wanting is trying to get.11 Outside the case of organ-
isms possessing the ability to explicitly tell us that they want something,
we would have to witness an organism trying to get something in order to
attribute a desire to it, or else have found some other physiological signs of
desire. In our own case, when we are aware that we desire something, it is a
matter of conceiving ourselves as being disposed to attain that thing. In any
case, we must appeal to an organism’s life form to attribute desires to it, for
we must be able to see it as trying to get something or as having a disposition
to do so. While I can have a wildly idiosyncratic desire, it must be like other
desires in some ways, and to attribute such a thing to myself invokes some
standards for my form of life.
Similar assertions can be made about beliefs; after all, some species are
evidently not capable of having certain beliefs. Wittgenstein presumably
means to point this out when he asks, rhetorically, ‘A dog believes his master
is at the door. Can he also believe his master will come the day after tomor-
row?’12 By contrast, it may seem that, as language users, we can believe
anything whatsoever; I can, for instance, believe that I am Louis XIV. Again,
the point is not that wildly idiosyncratic and insane beliefs are not possible,
but that at some level knowledge of the life form is involved in identifying
the presence of a belief in an individual. For belief involves species-typical
capacities for registering the way the world is, and even someone who
believes he is Louis XIV has the relevant capacities for determining what it
would be for him to actually be Louis XIV, even if those ideas are mistaken
or if he is mistaken that those conditions are fulfilled. Taking someone to
believe that he is Louis XIV is, among other things, attributing to him those
relevant capacities.
Human Nature, Virtue, and Rationality 89
Still, one might wonder whether all of this sets any limits on what can
be believed, desired, or understood as a reason for action. As I have just
pointed out, the view does permit wildly idiosyncratic desires and beliefs, yet
there exists some formal constraint on how I regard my desires and beliefs
as giving me reasons. I might take a wildly odd desire and belief to give me
a reason: for example, I might believe that by donning a certain necktie,
Napoleon Bonaparte will slide a desired twig of mountain ash under the
door, and I may think that this method will work only for me. So, I might
think I alone have reason to put on this necktie here and now. Nevertheless,
to see this desire as reason-giving is to see the desire as directed at something
that humans sensibly want to acquire. This is not to say that certain desires
are not perplexing; it would be difficult, as Wittgenstein puts it, to ‘find
one’s feet’ with someone possessing the desire for a twig of mountain ash,
and especially someone who thinks he can get it in such a magical way. One
question that may arise is whether the person with such a desire is acting on
a reason or under an irrational compulsion that is irresponsive to reasons.
Although someone can have idiosyncratic desires and beliefs, to take a desire
to provide reason is implicitly to judge that this is something a human can
intelligibly try to attain and that someone can be rationally criticized for tak-
ing the desire as reason under given circumstances. At the very least, I cannot
be perplexed about someone else contriving to get a twig of mountain ash
if it is something I have taken myself to have reason to try to get. Reasons I
take myself to have attach to an interpretation of reasons for humans.
If I cannot understand my reasons to be unique to myself, perhaps I could
accord a special weight to acting on the reason that something would benefit
me, but what would justify such a weighting, if not reasons that are, in prin-
ciple, applicable to everyone? Hence, when I act more selfishly than is gen-
erally considered acceptable, I demonstrate that my conception of a norm
for human life is one with that degree of selfishness; that is, one in which
considerations of benefiting oneself have a wider scope and greater weight.
I have argued that I must frame my desiring and acting on reasons against
the background of my form of life. This is not to say that any desire I have
must be counted as a good for human beings, or that I must aim at some good
in my desires; the limits of what I can intelligibly desire constitute only part
of my conception of my form of life. Acting on a reason places me against a
background of my form of life, but that then brings my acting on a reason
into contact with a broader background of norms for human beings. I may
have what I recognize as a bad desire, and understand myself as acting badly,
against other reasons I know I have not to act on that desire.
Now we are arguing about what is proper to human beings, and this is
the level at which we must argue, ultimately, for what considerations con-
stitute genuine reasons for action. Any argument concerning what reasons
we have will have to come back to certain inescapable facts about human
beings. These facts include the following: we are agents, we die, we are
vulnerable to physical harm, we are entirely dependent on adults through
90 John Hacker-Wright
infancy, and many similarly obvious facts. I believe any viable norm will
have to allow for our survival as individuals and as a species, our happiness,
and our ability to live in groups; these are Hursthouse’s four ends. To add
to her list, viable norms for human beings will also have to allow for the
achievement of rational agency. These are ends that constrain conceptions
of what it is for a human being to act well, as it is unreasonable to propose
norms for a form of life that would not, under normal circumstances, allow
an individual to live, or allow the species to survive, or would require suffer-
ing. Yet the fact that we are agents, I think, is particularly important since it
is an essential part of our self-conception. Defensible norms must allow us
to achieve that which we must rationally take ourselves to be. Human beings
can only reliably become rational agents under certain conditions, and the
norms we propose for our species must be responsive to facts about how we
cultivate rational agency. For example, we need a significant input of care
from one or more adults if we are to attain rational agency, and this point
is pertinent to defending Hursthouse’s case against impersonal benevolence.
Before I flesh out that claim, it is worth noting the connection of the argu-
ment thus far with Foot’s call for a ‘fresh start’ in moral philosophy.13 Foot
argues that we should dispense with any idea of practical rationality that
does not relate to goodness of the will. In her argument for this claim, she
draws on Warren Quinn’s case that practical rationality could not be consid-
ered a virtue if it allowed us to be rationally shameless; the dominant maxi-
mizing conception of rationality surely does just that. Foot follows Quinn
in asking what particular hold such a conception of rationality could have
on us if it requires of us that we act badly. The argument I made previously
strengthens this case by adding that those who think that an alternative
conception of rationality exists are simply mistaken. Good reasons answer
to the human good on this account, though something like a maximizing
account could be sustained by arguing for a maximizing account of the
human good. Yet a good human cannot simply maximize the fulfillment
of her preferences since, as I will elaborate in the following section, organ-
isms guided by such norms could only fortuitously reproduce themselves as
rational agents. Foot is right to hold that morality does not have to answer
to a default maximizing view of rationality, and indeed, maximizing desires
or preferences cannot be a norm for human beings.
NOTES
1. The two major defenses of this naturalistic approach to justifying the virtues
are Philippa Foot’s Natural Goodness and Rosalind Hursthouse’s On Virtue
Ethics. See also Alasdair MacIntyre, Dependent Rational Animals.
2. An important qualification of this claim is that it is a norm for some species to
withhold care under conditions of scarcity. Yet, this point can be folded back
into our characterization of the normal life of that species; in other words,
it is part of the characteristic life of that species to withhold care under such
circumstances. See Andreou, ‘Getting On in a Varied World’.
3. Foot, Natural Goodness, 30.
4. Foot, Natural Goodness, 43.
5. Foot, Natural Goodness, 56.
6. Hursthouse, On Virtue Ethics, 222.
7. Copp and Sobel, ‘Morality and Virtue: An Assessment of Recent Work in Vir-
tue Ethics’, 540. Note that Copp and Sobel are a bit mistaken in framing this
criticism; plants do not have characteristic enjoyments or well-functioning
social groups, and so there are fewer than four ends for some species, on
Hursthouse’s view.
8. Thompson, Life and Action, 43.
9. Thompson, ‘Apprehending Human Form’, 71.
10. Anscombe, Intention, 49.
11. Anscombe, Intention, 68.
12. Wittgenstein, Philosophical Investigations, 183.
13. Foot, Natural Goodness, 5–24; see also Foot, ‘Does Moral Subjectivism Rest
on a Mistake?’.
14. As I write this, genetically engineered male mosquitoes are being released
that produce moribund larvae. Though defective mosquitoes, they are at least
good from our standpoint insofar as they will cause a crash in mosquito
populations and hence reduce diseases that are carried by their species. See
‘Sterile Males for Mosquito Control’.
15. Of course, this puts our social relations in a purely instrumental light, and as I
will argue later, our interest in relationships must be more than instrumental.
16. See Annas, ‘Virtue Ethics: What Kind of Naturalism?’.
17. Annas, ‘Virtue Ethics’, 25.
18. Singer, ‘Should This Be the Last Generation?’.
19. Although Hursthouse does not connect the issue of characteristic enjoyments
to the rejection of impersonal benevolence, she clearly recognizes the point in
On Virtue Ethics, 234.
20. See my ‘Human Nature, Personhood, and Ethical Naturalism’.
21. A point made forcefully by Alasdair MacIntyre in Dependent Rational Ani-
mals, 99–118.
REFERENCES
Virtue ethical projects that draw their inspiration from Aristotle often come
with a special kind of naturalism. This ‘Neo-Aristotelian Naturalism’ (NAN)
is the subject of this chapter.1
What is NAN? In a first approximation, ignoring the varieties, one
could say the following: (i) NAN is a form of ethical naturalism in that it
claims that moral judgments can be grounded in judgments about human
nature.2 (ii) NAN differs from most other forms of ethical naturalism by
its functional, teleological perspective and its focus on the evaluation of
something as ‘good of its kind’, especially when applied to living beings.
(iii) NAN considers moral judgment a variant of this kind of evalua-
tion. According to NAN, the morally good person is good as a human
being.
Recent attempts to assess NAN have mainly focused on its concept of
(human) nature. Opponents of NAN doubt that the functional view of
nature is compatible with the natural sciences, especially with biology
‘after Darwin’.3 But perhaps this doubt is beside the point, for propo-
nents of NAN would respond that their aim is not to offer a biological
account of animals but to spell out what is necessary for us to recognize
something as a living being.4 It seems an open question to what extent
NAN has to take the natural sciences into account. The debate has hit
an impasse.
In what follows, I suggest a different approach to an assessment of
NAN. I will argue that if we follow its lines, we do not only have to
accept a certain picture of ‘human nature’, adequate or not by whatever
standards. We also have to accept a certain picture of ‘moral thinking’.
The functional perspective that is characteristic of NAN imposes a num-
ber of constraints on our consideration of the morally good. These con-
straints turn out to be problematic as soon as we take NAN as a way of
grounding moral judgments.5 Or so I will argue. NAN is interesting in its
own right, but in an important sense, it is a ‘non-starter’.
98 Philipp Brüllmann
II
III
As Hursthouse correctly notes, this premise does not provide action guid-
ance—it does not tell us what to do—unless we know who the virtuous
agent is. So what we need is a specification of the virtuous agent. And to
make sure that the virtue ethical account of right action offers a genuine
alternative to deontology, this specification should not refer to moral rules
or duties. It should not define the virtuous agent as someone ‘disposed to
act in accordance with correct moral rules’ (28–9). Here is Hursthouse’s
suggestion:
P.1a. A virtuous agent is one who has, and exercises, certain character
traits, namely, the virtues. (29)
But of course this premise cannot be the whole story, either. Unless we know
which character traits are the virtues, we do not know who the virtuous
agent is, and thus have no answer to the question of what sort of acts we
should perform. P.1a calls for an account of the virtues. In order to give this
account, Hursthouse introduces the following premise, explicitly labeled as
‘neo-Aristotelian:’
Good (as) Human Beings 99
P.2. A virtue is a character trait that a human being needs for eudai-
monia, to flourish or to live well. (167)
P.2a. The virtues make their possessor a good human being. (167)
Referring to the work of Philippa Foot, Hursthouse asserts that P.2a talks
about human beings as living beings, and she develops an account of good
human beings that starts from good plants and good (social) animals (ch.
9–10). It is with P.2a that NAN comes into play.
IV
Hursthouse claims that P.2a ‘can get off the ground as a criterion for a par-
ticular character trait’s being a virtue’ (192). This criterion might be speci-
fied as follows (this is my own formulation):
[T]he beliefs and putative facts about who can and cannot be relied
on, about whether you can fool most of the people most of the time, or
whether they can easily be manipulated, about what can be discerned
to be a pattern in life, what is to be attributed to good or bad luck and
what is ‘just what is to be expected’—about in short, human nature and
the way human life works—do not fall tidily under either classification
[of empirical vs. moral ‘facts’, P.B.]. Neither side believes what they
believe about how life works on the basis of even local, let alone world-
wide, observation or statistical analysis. The beliefs are part and parcel
of their ethical (or immoralist) outlook. (189)
Yet, it is easy to see how this reply invites another objection (which Hurst-
house deals with as well). If we enrich our concept of human nature with facts
from within an ethical outlook, we are at risk of merely re-expressing our ethi-
Good (as) Human Beings 101
cal beliefs instead of grounding them. We get drawn, it seems, into that vicious
circle that the appeal to human nature was meant to avoid (165).
According to Hursthouse, this second objection operates with the wrong
concept of validation. Referring to good human beings in order to validate
beliefs about which character traits are the virtues does not mean to derive
ethical conclusions ‘from a neutral point of view’. It rather means to follow a
‘Neurathian procedure’ (a term she borrows from John McDowell), that is,
to start from within an acquired ethical outlook and to submit opinions that
are part of this outlook to ‘reflective scrutiny’: ‘And those [particular judg-
ments, P.B.] that were part of the outlook and survived the reflective scrutiny
would not merely re-express it; they would now express, so to speak, that
they had survived the scrutiny’ (166).
VI
VII
VIII
As mentioned previously (section I), NAN differs from most other forms
of naturalism by its functional, teleological perspective and its focus on the
evaluation of something as good of its kind. As will turn out in what follows,
it is this functional perspective that stands behind the idea that the concept
of human nature provides a criterion for the virtues.
NAN’s functional perspective rests on one basic assumption: the assump-
tion that ‘good’ is a logically attributive adjective.12 What does that mean?
Briefly, it means that ‘good’, when used properly, always means ‘good rela-
tive to a certain kind’. This is taken to be true also of cases in which ‘good’
appears in the grammatically predicative way. When ‘good’ is used properly,
the judgment ‘X is good’ is to be understood as elliptical for ‘X is a good
F’. In this respect, the adjective ‘good’ is similar to adjectives like ‘big’ and
‘small’, but different from adjectives like ‘red’ and ‘yellow’. Usually, the
following criterion is used to mark the difference between ‘good’ as attribu-
tive and, for example, ‘red’ as non-attributive: Whereas from ‘X is a red F’,
you can infer ‘X is red’ and ‘X is an F’, you cannot infer ‘X is good’ and ‘X
is an F’ from ‘X is a good F’.13 There is a debate on whether this criterion
really does the job,14 but in the present context we can leave the intricacies
aside. Instead, we should focus on the question of what the idea that ‘good’
Good (as) Human Beings 103
is attributive—provided its correctness—implies according to NAN. Two
implications seem to be of special importance.
First implication. There are cases in which it is unquestionable that
‘good’ is used attributively, as in the judgment ‘X is a good knife’. In cases
like these, the relation between the description of X as a knife and the
evaluation of X as a good knife seems unproblematic because if you know
what a knife is, you already know (in some way or other) what a good
knife is. In the present context, we cannot go into the question of why
exactly this is the case, but obviously it has something to do with the fact
that knives are tools, that is, artifacts designed to serve a certain purpose
or function. Thus (i) ‘X is a good knife’ means ‘X is a knife that serves its
purpose well’; and (ii) to know what a knife is, is to know the purpose
it was designed for and, at least partially, the conditions under which it
serves its purpose well.15 Apparently, good knives offer an example for
an evaluation that is reducible to a description (to know that X is a good
knife is nothing but to know that X cuts well).
Now if it is true that the only proper use of ‘good’ is the attributive one,
and if all the other cases have enough in common with the case of tools, then
the relation between descriptive and evaluative judgments is never problem-
atic because the latter are always reducible to the former. This, in short, is
the reply that NAN offers to G.E. Moore’s ‘Open Question Argument’.16
Second implication. When Neo-Aristotelian Naturalists talk about ‘good’
as logically attributive, they usually talk about standards of goodness. While
the standards of redness are the same no matter which object we call red,
the standards of smallness change with, and depend on, the object that is
called small. If ‘good’ is like ‘small’, then the same holds for the standards
of goodness, that is, the standards of goodness change with, and depend on,
the object that is called good.
Why should we care? According to authors like Philippa Foot, we should
care because this implication guarantees the objectivity of those standards
in the following sense. The criteria for good knives cannot be chosen, so to
speak (we cannot use any standard we like when evaluating knives) because
partially those criteria are determined by what knives are, namely, tools
designed to serve a certain purpose. This, in short, is the objection that Neo-
Aristotelians raise against ethical non-cognitivism, which—in their view at
least—makes consistency the only restriction to an application of evaluative
terms.17 Thus Philippa Foot writes in ‘Goodness and Choice’:
[T]he man who uses these words [‘a good knife’, P.B.] correctly must use
them in conjunction with particular criteria of goodness: those which
really are the criteria for the goodness of knives. No matter what he may
do in the way of choosing knives which are M he cannot say ‘M knives
are good knives’ unless M is a relevant characteristic, or unless he is pre-
pared to show that M knives are also N knives, and N is a characteristic
of the right kind.18
104 Philipp Brüllmann
The basic idea is that using the words ‘a good knife’ correctly implies using
them in conjunction with those criteria that ‘really are’ the criteria for good
knives, that is, in conjunction with objective standards. Let us sum up this
idea by the following thesis:
So, according to NAN, there are two important implications of the assump-
tion that ‘good’ is logically attributive. The first implication is that the rela-
tion between the descriptive and the evaluative is unproblematic. The second
is that there are objective standards of goodness, determined by the object
that is called good.
IX
This summary of NAN is of course far from exhaustive, but it should suf-
fice to illustrate two things: first, what Neo-Aristotelian Naturalists mean
when they claim that there is a ‘common pattern’ to the evaluation of bees
106 Philipp Brüllmann
and wolves on the one hand and moral evaluation on the other; second,
why they find this common pattern attractive. If NAN is correct, then moral
evaluation is just as unproblematic and rests on standards as objective as the
evaluation of knives, wolves, and bees. Let me develop a little further what
this tells us about the theory of NAN.
Apparently, the point of NAN is not that we should take into consider-
ation what a human being is when deliberating on moral questions. This, I
think, is hardly controversial. The point of NAN is that there is a special con-
nection, a link, between judgments about the virtues and judgments about
human nature. As shown in the preceding paragraphs, this link amounts to
a reduction and is based on our mastery of certain concepts like ‘tool’, ‘life-
form’, and (as suggested) ‘virtue’. NAN is naturalism on a conceptual basis.
By claiming that there is a specific, conceptual link between judgments
about the virtues and judgments about human nature, NAN does not only
presuppose a functional concept of human nature. It also determines how
this concept grounds our judgments about the virtues. In a nutshell, it is
because we know how to apply ‘X is a good F’ correctly, and because we also
know the scope of this concept, that we consider the virtuous person as good
qua human being. This, once again, is the point of NAN’s criticism of ethi-
cal non-cognitivism that Hursthouse reiterates herself.25 So there is a close
connection between the features by which NAN has been characterized in
section I. NAN claims that moral judgments can be grounded in judgments
about human nature because it regards moral judgment as a variant of the
evaluation of something as good of its kind.
As we have seen in section IV, the reference to human nature is attrac-
tive for an ethics of virtue because it seems to provide an independent and
objective basis for our judgments about the virtues. Now we can see that
NAN determines the way that happens, namely, as I said, by a reduction. To
understand the point of this remark, it is crucial to distinguish the question of
how, according to NAN, virtue ethics can take advantage of the objectivity of
human nature (a) from the question of how objective NAN’s concept of human
nature actually is (b). It is the first question that I have been dealing with, and
on the whole my suggestion is to characterize NAN rather by (a) than by (b).
XI
XII
XIII
The result of the argument in sections VII–XII is that NAN is not compat-
ible with a Neurathian procedure. It is already by submitting naturalism to
this procedure that Hursthouse detaches it from its Neo-Aristotelian back-
ground. So it seems we are left with two alternatives.
The first alternative is to give up the Neo-Aristotelian way of validating
C (‘A character trait is a virtue iff it contributes to making its possessor a
good human being’) and to regard the latter as one belief of our ethical
outlook that may or may not survive reflective scrutiny. To my impression,
this is what Hursthouse is actually doing;30 and if what I say is correct, her
approach leads to a completely different kind of naturalism, a different way
of taking nature into account when deliberating on ethical questions (which
is true also if C should survive reflective scrutiny).
To fully understand the consequences of this first option, we have to
remember how NAN is connected to the advantages of naturalism for an
Good (as) Human Beings 109
ethics of virtue (section X). If applying a Neurathian procedure implies deny-
ing that C is sufficiently justified on the basis of a priori fixed points (on the
basis of our mastery of certain concepts), it also implies doing without the
specific way in which NAN provides virtue ethics with independence and
objectivity, namely, by reducing judgments about the virtues to judgments
about human nature.31 And this appears to me a decisive shift in the theory.
Be that as it may. It is the second alternative that is more relevant in the
present context, that is, the option to stick with NAN as a way of ground-
ing judgments about the virtues and hence to give up the idea that C is to be
validated by a Neurathian procedure. As I will try to show, this option tells
us something about the constraints NAN imposes on our moral thinking.
XIV
XV
NAN cannot be reduced to the claim that the virtuous person is good as a
human being, for NAN is a theory about how this claim is grounded. It is
a theory that invites (and commits) us to take thinking about good knives
and good bees as a model for moral thinking, which means to impose some
important constraints on the latter.
It is conspicuous that Rosalind Hursthouse, a virtue ethicist who sees
herself in the tradition of Elizabeth Anscombe and Philippa Foot, is will-
ing to drop naturalism, if it should turn out that those character traits that
make us good as human beings are completely different from those we call
the virtues (194–195). But if we think it possible that naturalism might yield
such a result, we have already dropped NAN, as I have tried to show. In this
sense, NAN is a non-starter.33
NOTES
1. The most important proponents of NAN are Philippa Foot and Rosalind
Hursthouse (see Foot, ‘Does Moral Subjectivism Rest on a Mistake?’ and
Natural Goodness; Hursthouse, On Virtue Ethics) who are influenced in
a general way by the work of Elizabeth Anscombe and Peter Geach (see
especially Anscombe, ‘Modern Moral Philosophy’; Geach, The Virtues), and
in a more specific way by Michael Thompson’s investigations into the logic
of judgments about ‘life-forms’ (see his ‘The Representation of Life’). I set
aside here the question of whether it is really adequate to call NAN an ‘Aris-
totelian’ approach. Though it is clear that Aristotle thinks that a virtuous
Good (as) Human Beings 111
person is good as a human being, it is far from clear whether he intends to
ground judgments about the virtues in judgments about human nature. For
some reflections on how NAN transforms Aristotle, see Brüllmann, ‘Laster
als natürliche Defekte?’.
2. Taking ‘moral judgment’ and ‘to be grounded’ in a very wide sense.
3. See, e.g., Millum, ‘Natural Goodness and Natural Evil.’
4. See, e.g., Hacker-Wright, ‘What is Natural about Foot’s Ethical Naturalism?’.
5. This is true even if we take ‘grounding’ and ‘moral judgments’ in a very wide
sense.
6. All references in the main text are to this work.
7. The second one is ‘The virtues benefit their possessor.’ Since Hursthouse
asserts that the two claims are ‘interrelated’ (167), picking out one of them
means to oversimplify her account. But, once again, my aim is not to offer a
discussion of that account but to take some of its features as a basis for illus-
trating a more general point.
8. See Gowans, ‘Virtue Ethics and Moral Relativism’ for an interesting discussion
of this aspect.
9. Cf. Woodcock, ‘Philippa Foot’s Virtue Ethics Has an Achilles’ Heel’.
10. Cf. Foot, Natural Goodness, ch. 3, and Nussbaum, ‘Aristotle on Human
Nature and the Foundations of Ethics’.
11. In other words, Hursthouse tries to combine Foot and McDowell, which is
interesting because the latter develops his ideas (in ‘Two Sorts of Naturalism’)
starting from a criticism of Foot’s NAN.
12. See Foot, Natural Goodness, 2–3; Hursthouse, On Virtue Ethics, 195. The
classic text is Geach, ‘Good and Evil’.
13. See Thomson, Normativity, 1–17 and 233–248 for an elaboration.
14. See Rind and Tillinghast, ‘What Is an Attributive Adjective?’.
15. I say ‘partially’ because there is room for disagreement, and it might take
some deliberation to determine when exactly a knife serves its purpose well.
But importantly, NAN defines the kind of disagreement to be expected and
the kind of deliberation necessary for those who have the appropriate con-
ceptual knowledge (see section XIV).
16. See Geach, ‘Good and Evil’; Foot, ‘Moral Beliefs’ and Natural Goodness,
2–3; and (more explicitly) Thomson, Normativity, ch. 1. Roughly, the idea
seems to be that someone who considers (say) ‘Is a knife that cuts well a good
knife?’ an open question has a lack of conceptual knowledge.
17. See, e.g., Foot, Natural Goodness, 7.
18. Foot, ‘Goodness and Choice’, 133.
19. This is an approximation. Depending on our background assumptions, we
might for instance (i) add ‘she does not know how to apply the concept “X is
a good F” to knives,’ or (ii) reduce the apodosis to ‘she does not know what
a knife is’ (under the assumption that describing something as a knife already
implies taking a normative stance). In the present context we need not decide
these questions. The important point is to see that NAN defines the kind of
mistakes that can be made in the evaluation of knives.
20. See Thompson, ‘The Representation of Life’, and Foot, Natural Goodness,
30–31.
21. See Foot, Natural Goodness, ch. 3, who associates human as animal good with
the goals of survival and reproduction, but emphasizes how complex these goals
turn out to be in the case of human beings. Cf. also the more explicit account of
Hursthouse who spells out four ends to which evaluations of social animals as
good of their kind refer (individual survival, continuance of the species, freedom
from pain and enjoyment, good functioning of the social group; On Virtue Eth-
ics, 202), and then assigns a ‘genuinely transforming effect of our [human, P.B.]
112 Philipp Brüllmann
rationality’ (218) on this basic structure (instead of adding a fifth aim for the
case of an evaluation of human beings).
22. I say ‘more or less’ because in most contexts we do not judge animals as good
of their kind but as good in relation to our needs and goals (‘Good dog!’).
23. See Foot, Natural Goodness, 2–3, 38–39; Hursthouse, On Virtue Ethics, 195.
24. It seems that the meaning of ‘good’ does undergo a change (mysterious or not)
that is indicated by the fact that the opposite of ‘good’ in its non-moral use is
‘bad,’ whereas the opposite of ‘good’ in its moral use (or in one case of moral
use) is ‘evil’. For a criticism of Geach’s position, see Pigden, ‘Geach on Good’.
25. ‘Hare can call a cactus a good one on the grounds that it is diseased and dying,
and choose it for that reason, but what he must not do is describe it as a good
cactus, for a cactus is a living thing’ (On Virtue Ethics, 195).
26. Hursthouse’s reference is McDowell, ‘Two Sorts of Naturalism’.
27. For some remarks, see Harman, ‘General Foundation versus Rational Insight’.
28. This illustration, which somehow undermines the distinction between epis-
temic and motivating reasons, seems problematic to me (see section XII), but
I cannot pursue the issue here.
29. See McDowell, ‘Two Sorts of Naturalism’.
30. It thus seems that the transition from T 1 (good knives) and T 2 (good bees)
to T 3 (virtuous human beings) is not as obvious on Hursthouse’s account as
the theory of NAN seems to imply (see section IX).
31. The point is, within the framework of NAN, judgments about the virtues
are just as independent and objective as judgments about human nature. But
the same does not seem to hold for the situation of a Neurathian procedure
where the former judgments are not reduced to the latter. Here, judgments
about the virtues are independent and objective to the extent that judgments
that survive reflective scrutiny have these properties. Cf. Gowans, ‘Virtue
Ethics and Moral Relativism’, 406–408, for some remarks on NAN and the
Neurathian procedure.
32. It is important to note that this result does not beg the question with respect
to what counts as ‘moral’. It suffices to assume that there are certain con-
tent restrictions as to when a judgment is a moral judgment, whatever those
restrictions may be. Cf. Hurthouse’s worry that naturalism might ‘yield far
too many horrific [results, P.B.] for us to count it as validating ethical beliefs
at all’ (On Virtue Ethics, 194).
33. I would like to thank Anne Burkard, Benjamin Kiesewetter, and Thomas
Schmidt for helpful comments on an earlier draft of this chapter.
REFERENCES
The subject matter of this chapter falls within the triangle marked out by three
Aristotelian ideas—human nature, human excellence, and human well-being.
The chapter is a highly programmatic attempt to introduce some material
from developmental psychology—specifically, from attachment theory—and
to explain why philosophers working somewhere within the Aristotelian tri-
angle have reason to take more of an interest in it than they do now.
I am assuming of course that attachment theory is true: if it isn’t, or doesn’t
make some reasonable claim to be believed, there’s no reason for anyone to
take an interest in it. But if you grant that much—and it is one of the leading
theoretical orientations in developmental psychology1—there are lots of true
theories: why should neo-Aristotelians be interested in this one?
Attachment theory is a theory of child development. Indeed properly
speaking it is a theory of human development, but—partly because children
are easier to study than adults, partly because childhood experience may
be especially important in making us the way we are—attachment theorists
have taken a special interest in the early years. This is already one, rather
general, reason why neo-Aristotelians should be interested in the theory, for
Aristotle’s ethics is (in part) a developmental theory: it aspires to provide not
only a theory about what human excellence is, but a theory about how we
acquire or fail to acquire it. Aristotle himself says that because good charac-
ter is produced by habituation,
So simply insofar as it has much to say about ‘our very youth’, attachment
theory stands a chance of telling us what Aristotle tells us very little about,
namely how we get from there to here, as well as merely that we sometimes do.
But there is a more specific reason than this. Ethics in the neo-Aristotelian
mould sees itself as a naturalistic undertaking: that is, it seeks to locate eth-
ical life in the world as made intelligible to us by natural science. Moreover
there is reason to think that the proper form of such an undertaking is to
Attachment Theory, Character, and Naturalism 115
display the continuity between our second and our first, or between our
ethical and our psychobiological natures.3 Now attachment theorists claim
that some human traits described by the theory arise through natural selec-
tion. Those traits, then, would belong to our first natures, or to ‘an account
of human beings which is to the greatest extent possible prior to ideas of
the ethical’.4 But attachment theory is also a taxonomy of psychological
dispositions, plus a theory about why people (children, adults) have the
disposition(s) of that sort which they do. Furthermore, these psychological
dispositions appear to stand in an explanatory relation to some traditional
traits of character, that is, to some virtues and vices. Attachment theory
therefore looks as if it is well equipped to put some empirical flesh on the
bones of Aristotelian naturalism by making vivid the continuity I have said
this variety of naturalism demands.
So far it looks as if I envisage the flow of ideas as being entirely from
attachment theory to neo-Aristotelian ethics. But that’s not so. As far as I
know, attachment theorists are unaware of the ways in which different forms
of naturalism are debated within ethics and might be surprised at the thought
that their work had anything to do with ethics. But perhaps precisely because
they think that what they are up to is just psychology or evolutionary biol-
ogy (i.e., some sort of natural science), some of them have apparently signed
up unawares to a version of ethical naturalism that goes much further than
the continuity thesis I mentioned earlier. They are in distinguished philo-
sophical company here—the late Philippa Foot’s, for example.5 But I shall
argue that Foot’s ambitious version of neo-Aristotelian naturalism and the
versions of attachment theory that unwittingly subscribe to it both run into
difficulties. Thinking through the connections between attachment theory
and neo-Aristotelian ethics, then, should be a way both of reining in some
of attachment theory’s own more extravagant theoretical ambitions, and of
demonstrating both the prospects (the continuity thesis itself) and some of
the limits of naturalism in ethics.
I shall return to the continuity thesis shortly, but let me focus for now on the
ambitious version of neo-Aristotelian ethical naturalism, as exemplified by
Foot. In this version of naturalism, claims about human virtues and vices,
or excellences and defects, are identified with claims about the way human
beings should be if they are to be good of their kind, or properly suited to
lead our characteristic species life. Here—in contrast to the kind of naturalism
that Moore is famous for opposing, a species of reductionism in which moral
properties are, in one way or another, identified with ‘natural’ ones—it isn’t
a matter of identifying one kind of property with another: only one kind
of property, the excellences and defects themselves, is ever under discussion.
Nor is it a matter of deriving surprising claims about which properties are
116 Edward Harcourt
excellences or defects in humans: it’s assumed we will more or less agree on
that at the start. Rather it’s a matter of arguing that these interesting properties
are the virtues and vices because they have a further property, that of playing a
certain kind of role, namely, that they are necessary (in the case of the virtues)
for us to lead the kind of life that, as members of that species, we’re supposed
to lead. Just as it is an ‘Aristotelian necessity’6—a necessity that ‘determine[s]
what it is for members of a particular species to be as they should be’—‘for
plants to have water or birds to build nests’,7 so ‘for human beings the teach-
ing and following of morality is something necessary. We can’t get on without
it’; again, ‘getting one another to do things without the application of physical
force [and which morality accomplishes] is a necessity for human life’. Cor-
respondingly, a species member who does not do what it is necessary for the
species to do is ‘naturally defective’, so immoral human beings are defective
in just the same sense as birds who fail to build nests, or owls who cannot see
in the dark. And because it’s a matter of plain fact that virtues are necessary
for us, it is also a matter of plain fact that a human being with a given virtue
is excellent, or with a given vice defective. Thus Foot’s ambitious naturalism
is designed to fulfill the cognitivist meta-ethical ambition common to many
ethical naturalisms. Contrast the continuity thesis, whose relation to cognitiv-
ism is looser and won’t be discussed further here.
This ambitious form of naturalism ought to interest empirical investiga-
tors who address questions about the way human beings should be or—
transposing the same idea into a developmental idiom—about optimal
development,8 attachment theorists included. But there’s an interesting dif-
ference. There’s no difficulty in getting moral philosophers to recognize that
what they’re working on are virtues and vices, that is, ethical notions. The
controversy arises when ambitious naturalists try to get people to agree that
these ethical notions are also ‘natural’ notions—that is, that virtue consists
is a perfection of our first nature, or of our nature as we can make it intelli-
gible to ourselves ‘to the greatest possible extent prior to ideas of the ethical’.
With developmental psychologists the sticking point is different: many of
them have no difficulty at all in agreeing that the notion of optimal devel-
opment—human beings turning out as they should—is a psychobiological
one. The surprise to them is that this notion is also ethical. But suppose they
get over their initial surprise, and suppose the concept of optimal develop-
ment, as deployed by attachment theorists, really is an ethical one. Isn’t the
very fact that these developmentalists have been studying it unawares, using
empirical methods, for all that time, evidence for the truth of ambitious
naturalism? I shall argue (in section V) that despite appearances, there’s no
support for ambitious naturalism to be derived from developmental psy-
chology, at least in the form of attachment theory. Roughly, insofar as the
dispositions in which attachment theory deals are ‘adaptive’—even if some
theorists have claimed otherwise—this is not in the sense that they are the
result of natural selection (and so belong to our first natures), but rather in
the sense that they are favorably related to human social life—and perhaps
Attachment Theory, Character, and Naturalism 117
to a particular form of it rather than to human social life in general. Insofar
as more is claimed, attachment theory overreaches itself, in a way that I shall
suggest ambitious naturalism in ethics does, too.
II
III
Although almost all human infants form attachments, not all attachments
are alike in quality, and this fact and its explanation forms the subject-matter
of attachment theory’s ‘individual-difference component’. The first mea-
sure of attachment quality was the ‘Strange Situation’, developed by Mary
Ainsworth.16 This test is administered at either 12 or 18 months, to one
parent–infant pair at a time—and note that at this age, at least, infants can
fall into different attachment types with respect to different parents. To sim-
plify, the Strange Situation proceeds as follows: the mother (let’s say) enters
an unfamiliar room with the infant and settles it down to play with some
toys. A ‘stranger’ (an unfamiliar research assistant) then enters who starts
to play with the infant, and after a short time the mother tells the infant she
is leaving, leaves for three minutes, and then comes back. The separation
and reunion is repeated, with the stranger absent.17 The infant’s behavior is
recorded throughout. Observed infant behavior falls into three recognizable
patterns. (There is also a fourth pattern that was theorized later, but I omit
that, again for simplicity’s sake.) In pattern B, the infant is overtly distressed
when the mother leaves, then seeks proximity with her when she comes back
and is comforted by it, and then resumes playing. In pattern A, the infant
does not express distress when the mother leaves, though it displays other
signs of distress such as more rapid breathing and heart rate, suggesting
Attachment Theory, Character, and Naturalism 119
it is suppressing the expression of distress rather than simply indifferent.
The infant doesn’t show a preference in play as between the mother and
the stranger and is then also relatively indifferent when the mother returns
(e.g., looks or turns away from her) and ‘if picked up . . . makes no effort
to maintain the contact’. In pattern C, the infant may be ‘clingy’ toward
the mother and uninterested in the toys even before she leaves, is immod-
erately distressed when she does leave, but when she comes back doesn’t
calm down and exhibits ‘furious clinging’—‘seeking contact, then resist-
ing contact angrily once it is achieved’. These three attachment types are
labeled ‘insecure-avoidant’ (type A; also ‘deactivating’), ‘secure’ (type B),
and ‘insecure-ambivalent’ or ‘insecure-resistant’ (type C; also ‘hyperactivat-
ing’).18 But though infants were the first to be systematically classified into
attachment types, attachment theory does not apply only to infants: on the
contrary, it is supposed to apply across the life span. Accordingly, other
tests, based both on observation and on interview data, have been devel-
oped for various later stages of life (e.g., the Cassidy-Marvin system and the
Preschool Assessment of Attachment for preschool-age children,19 and the
Adult Attachment Interview20), with roughly the same number of attach-
ment classifications, often with similar names to those used in the Strange
Situation. Though there is some debate about which age-specific test is the
most reliable for a given age, and about the extent to which different age-
specific tests keep track of the same characteristics, there is an evident family
resemblance between the criteria for these age-specific tests.
The second aspect of attachment theory’s individual-difference compo-
nent that I want to draw attention to concerns the further characteristics
with which secure and insecure attachment are associated. These associa-
tions have both a diachronic and a synchronic dimension: the diachronic
dimension concerns the characteristics predicted, at a greater or lesser
distance in time, by a secure infantile attachment history; the synchronic
dimension concerns the characteristics contemporaneously associated with
secure attachment as measured by the test(s) appropriate to the life-stage in
question. On the whole the contemporaneous associations are stronger than
the predictive ones,21 and the predictive associations are weaker the longer
the distance in time and the further removed the ‘outcome domain’ is from
quality of relations with the infantile attachment figure him- or herself. This
is thanks to the fact that attachment classifications can shift quite early in
life (e.g., an infant who is securely attached to its mother aged 12 months
may become insecure if the mother suffers from post-natal depression fol-
lowing the birth of a second child; and an insecurely attached child may
become securely attached to someone if it is fortunate in its adoptive or
foster parents).22 And if they don’t shift, this is likely to be not only thanks
to the infantile attachment history, but thanks to the persistence of the fac-
tors—such as warm relations with parents—that also explain the infantile
attachment classification. Thus if there is a predictive relation to later char-
acteristics, it is likely to be mediated by a variety of further factors.23
120 Edward Harcourt
To summarize some recent findings,24 infant attachment security predicts
a good relationship at least a year later between the child and the attachment
figure, considered in terms of ‘enthusiasm, compliance’, and ‘less frustra-
tion and aggression’ during shared tasks; secure attachment also predicts
harmonious caregiver–child relations over longer periods in the presence
of continued sensitive caregiver behavior. Secure attachment in adulthood,
meanwhile (whether or not itself predicted by secure attachment in infancy),
is correlated with greater sensitivity to one’s own children’s needs and ‘more
warmth and appropriate structuring of learning tasks’25 and, in attachments
to peers, a capacity inter alia to admit vulnerability and need for the other
without ‘continually worrying about the attachment figure’s availability’.26
Moving to the next widest outcome domain, that of other relationships, chil-
dren with secure attachment histories have less conflictual relationships with
peers from preschool to 7 years,27 are less dependent on teachers in pre-
school,28 are less dependent on counselors at summer camp aged 10, and are
more sociable with unfamiliar adults.29 By contrast, the insecurely attached
4-year-old boys exhibit more ‘aggressive, assertive, controlling and attention-
seeking behavior than their securely attached counterparts’. Finally, attach-
ment theory argues for a connection between attachment security and
broader personality traits. The Minnesota Study of Risk and Adaptation
from Birth to Adulthood30 argues for ‘significant associations between early
attachment security and personality characteristics throughout childhood
and adolescence . . . [including] self-esteem, agency and self-confidence, [and]
positive affect’. Securely attached children aged 6 describe themselves gener-
ally in positive terms but are better at admitting flaws—insecurely attached
either are more negative about themselves or do not admit flaws. There
is also an important contemporaneous association between secure attach-
ment and the capacity for emotion-regulation, including in adulthood,31
and between secure attachment and psychological understanding (they are
more ‘proficient at identifying emotions in others, . . . especially . . . negative
emotions and mixed feelings’). Thompson concludes that ‘children with a
secure attachment history are capable of developing and maintaining more
successful close relationships, especially with parents and with peers, than
are insecure children; they develop a variety of desirable personality qualities
in childhood and adolescence [including ‘social problem-solving skills’]; they
are more likely to exhibit constructive forms of emotionality and emotion
self-regulation; and they exhibit more positive self-regard’.32
IV
I now want to use my sketch of attachment theory to explain why the theory
is well placed to flesh out what I earlier called the continuity thesis. Attach-
ment dispositions face both backward toward the psychobiological (thanks
to attachment theory’s normative component) and (thanks to its individual-
Attachment Theory, Character, and Naturalism 121
difference component) forward toward the ethical. As a result, the more we
get to understand attachment, including what explains it and what it explains,
the better we should be able to bring the continuity idea to life and, thus, to
take it beyond a picture of how some naturalistically minded philosophers
think things ought to turn out.
As regards the forward-facing connection, it is striking how generally
the characteristics with which secure attachment is associated, whether
predictively or contemporaneously, seem to be positive characteristics: the
capacity to relate harmoniously to others, co-operativeness, ‘positive affect’,
self-esteem, the capacity to be realistic about oneself and to tolerate one’s
own imperfections, are all apparently more worth having than those typically
possessed by the insecurely attached or by those with insecure attachment
histories (dependence, attention-seeking, low self-esteem, limited capacity for
symbolic play, and so on). It’s thus unsurprising that a great deal of effort
is expended, by parents, educators, therapists, and others, on trying to get
children into attachment category B and making sure they stay there.
However, the characteristics in question are ill-assorted. Some seem either
to be virtues or to imply virtues. Here I am thinking of the capacity for friend-
ship; the capacity to form a realistic appraisal of one’s own excellences and
defects, which is surely a virtue, though perhaps not an Aristotelian one;33
and the capacity to offer help and to ask for help when you need it. Co-
operativeness, meanwhile, if it is not itself a virtue, surely implies the tradi-
tional virtues of trust and, unless the co-operation is very short-term, also
honesty and fidelity to promises. Other characteristics I mentioned seem like
more general character-traits that have sometimes been argued (emotional
self-regulation34) to go with or (self-esteem35) to underlie the virtues, while
some are traits that may be as it were adverbially related both to virtues and
to vices (aggression—good in fighting the local authority to get your child a
place at school, not so good in bullying a colleague into accepting an unfair
workload; the same goes for ‘positive affect’, though one really needs to know
more about what’s meant—if it means ‘the capacity to enjoy life’s goods to the
full’, arguably it is itself a virtue). Empathy is another tricky one, depending,
for example, on whether one thinks cruel people genuinely possess it.
There is a lesson to be learned both from the evaluative asymmetry (as
we might put it) between secure and insecure attachment and from the het-
erogeneity of the positive traits. The evaluative asymmetry seems to show
that secure attachment stands in a privileged relation to the virtues (and
insecure attachment to the vices). But how close is the relation? At one
extreme, the answer would be that secure attachment is virtue; or perhaps
that it’s the disposition that underlies and unifies the virtues. That would be
a highly ambitious direction for neo-Aristotelian ethics to try to go in.36 On
this view, attachment theory would not be an intermediate level of theory
that merely mediates between the biological and the ethical, because the
individual-difference component of attachment theory already is a theory
about the ethical, that is, a theory that stands to virtue and vice as (say)
122 Edward Harcourt
Plato’s tripartite moral psychology does. On a more modest view, secure
attachment would belong at a level intermediate between the biological and
the ethical, with secure attachment occupying the place occupied (roughly)
by Aristotelian ‘natural virtue’,37 a disposition that is not yet virtue but may
turn into it if properly cultivated. That certainly fits the fact that attachment
dispositions are relatively unstable in the early years. One might also take
this more modest thought in a skeptical direction: if secure attachment is
clearly in some sense privileged among the attachment dispositions, the fact
that it’s not straightforwardly related to the virtues (after all the previous
list leaves a great many out) could be used to explain why people are not
typically unified in respect of the virtues (honest without being generous,
or honest to colleagues but dishonest to lovers), or to challenge the Aristo-
telian thought that we’re ‘made for virtue’.38 I can’t develop these lines of
thought here, let alone adjudicate between them, but this doesn’t matter for
my present purpose: whichever way these lines of thought are developed, we
are going to end up with a richer and more realistic version of the continuity
thesis than we have so far.
VI
I want finally to apply the foregoing reflections to raise some questions for
Philippa Foot’s claim that vice is a natural defect in humans and for the
familiar idea that the virtues stand in a privileged relation to well-being.
Whatever its relation to virtue, secure attachment appears to be in some
sense best for the person whose disposition it is. Perhaps this is most clearly
argued in connection with infancy itself. The salient characteristics of secure
attachment in infancy—the freedom to express distress when it is felt with-
out fear of rejection and in the expectation of comfort, the capacity for
the pleasures of ‘affective sharing’45 and warm physical contact, and the
freedom to become absorbed in the environment—are real human goods.
These both reflect the real goods of the kinds of relationship that give rise
to secure attachment, and—especially in the case of the freedom to become
absorbed—make available to the infant a great many other goods in their
turn. They are also goods that are, to varying degrees and in varying combi-
nations, unavailable to the insecurely attached infant. Thus quite indepen-
dently of what, if anything, secure attachment predicts about characteristics
later in life, to describe secure attachment in infancy is to describe a good
infancy in the sense that parallels ‘a good childhood’ or ‘a good life’. But if
the point is especially vivid in connection with infancy, the capacity for good
close relations in later life (which, e.g., balance intimacy and autonomy) is
also associated with secure attachment. One need only remind oneself of
the number of people who refer themselves for psychotherapy because they
find themselves unable to enjoy those goods to gauge the privilege of secure
attachment in relation to well-being.46 This privilege plays well—as far as
it goes—for the Aristotelian association between virtue and well-being if
secure attachment also has a privileged relation to the virtues.
I have also suggested that secure attachment enjoys another kind of privi-
lege, in the sense that the characteristics associated with it seem desirable in
a way those associated with insecure attachment are not. But there is surely
an element of cultural relativity here. One only has to switch context to, say,
ancient Sparta—of legend if not of fact—for it to be quite probably better
(for me) to be insecurely attached: think of the oft-cited insecure-avoidant
trait of precocious self-reliance, useful if one has to spend days on end on
solitary sentry duty. One can make the same point for insecure-ambivalent
Attachment Theory, Character, and Naturalism 125
attachment: in a war zone, where real threats are more or less constant,
insecure-ambivalent unwillingness to allow distance from the attachment
figure is the more ‘adaptive’ characteristic.47 If secure attachment is privi-
leged in respect of suiting us better to social life, the privilege is thus surely
relative to the more or less stable circumstances in which we live. Note how-
ever that this is not to endorse cultural relativism: one of the awful things
about ancient Sparta might be that it prized the reproduction of insecure-
avoidant types, thus leading many of its citizens to miss out on the real
goods of warm personal relations; the same goes mutatis mutandis for war
zones. But if secure attachment bears a privileged relation to some virtues,
the relativity raises a problem for the idea that these virtues are necessary
for our species life, or that lacking them is a ‘natural defect’, for it looks as
if, as long as circumstances are imagined to be appropriately different, some
version of our species life could be carried on just as well if the distribution
of insecurely attached people in the population were the same as the distri-
bution of securely attached people here and now.
The point about relativity can be pressed further. If insecure attachment
would be optimal for a majority of the population in radically challenging
or threatening circumstances, it is surely useful in some of the population
even in our circumstances. Although our circumstances are more or less
stable, our social world is sufficiently complex to make it likely that some
division of labor—made possible by the variety of attachment dispositions
in a given population—is necessary to the form in which we, locally, carry
on our species life. Thus it is surely good in our own fortunate though
imperfect circumstances to have some people around who are risk-takers
(and so presumptively insecure-ambivalent) and some who are precociously
self-reliant (presumptively insecure-avoidant): it is not obvious that the full
range of goods that are available to humans would be realized in a society in
which everyone was cooperative, affectionate, and compliant. Thus it might
not be that the only human analogue of Foot’s naturally excellent wolf who
hunts with the pack is the cooperative person who sits attentively round the
committee table: if the naturally excellent are those who have the charac-
teristics necessary to sustain our species life, then granted the point about
the division of labor, this description might net not only the good committee
person but the odd person who angrily storms out of meetings (or simply,
never attends meetings because they are hatching a plan on their own). The
real human equivalent of the lone wolf or the night-blind owl would rather
be the rare human being who has no disposition to form attachments—a
defect indeed, but whose absence seems to leave just about everything open
as far as virtues and vices are concerned.
At the very least, the conclusion to draw is that since our species life can
be carried on in circumstances that vary greatly in respect of stability and
the presence of threats, there is no single attachment disposition (or single
distribution of different attachment dispositions among a population) that is
necessary for us to do so, and so no attachment disposition48 which is per se
126 Edward Harcourt
a natural defect. So far that skeptical point says nothing about virtue either
way: the extent to which virtue comes in depends on the strength of the
association between virtue and secure attachment. If secure attachment does
have a privileged relation to virtue, it looks as if vices can’t be natural defects
because insecure attachment, too, is an ingredient in the mix necessary for
our species life—it would be as impossible to sustain if no one had that as if
no one was securely attached. But perhaps secure attachment doesn’t have a
privileged relation to virtue. If that is so, then granted the apparently privi-
leged relation of secure attachment to well-being, that would make trouble
for the association between virtue and well-being, though perhaps that con-
nection is in trouble anyway.49 In any case, virtue could be underpinned
psychologically by either a secure or an insecure attachment disposition
(perhaps depending on the virtue, perhaps depending on the circumstances):
the good committee person and the awkward individualist who doesn’t turn
up for meetings might, though in different attachment categories, both be
virtuous. Whether secure attachment does bear a privileged relation to vir-
tue, however, awaits a proper investigation of what in the way of virtues
secure attachment is and is not correlated with—a question that attachment
theorists may not ask in so many words, but on which their data bear in a
multitude of ways.
NOTES
REFERENCES
Virtues do not exist in a personality vacuum, but they co-exist with a con-
stellation of other personality structures and processes that can help or
hinder the development, sustenance, and exercise of virtue. Insights into
some of them are found by adverting to Cervone’s theory of ‘knowledge and
appraisal personality architecture’ (KAPA).
Cervone avers that two aspects of cognition are central for modeling
the architecture of personality: knowledge and appraisal.10 According to
him, knowledge is structural and appraisals are dynamic. He contends that
‘Knowledge consists of beliefs about actual or prospective attributes of per-
sons or the environment. Elements of knowledge, then, are enduring mental
representations of a feature or features of oneself, other persons, or the
physical or social world’.11 In addition, knowledge varies in the extent to
which it is generalized or domain-linked. Cervone’s account of knowledge
is largely consistent with standard philosophical conceptions of knowledge
as beliefs or mental representations.
Appraisals are more complex:
CONCLUSION
The ideas offered in this chapter are sketches that require much further
elaboration. They are offered here as notes toward the development of a
more comprehensive empirical grounding for virtue than that begun in my
earlier work. Here as there, I disagree with the specific conclusions and
visions of personality, cognition, and virtue that philosophical situationists
endorse. Yet, we have to thank them for prodding us to do the empirical
work of grounding virtue.
NOTES
1. Harman, ‘Moral Philosophy Meets Social Psychology: Virtue Ethics and the
Fundamental Attribution Error’, 316.
2. Harman, ‘Moral Philosophy Meets Social Psychology’, 327–328.
3. Doris, Lack of Character: Personality and Moral Behavior, 6.
4. Doris and Stich, ‘As a Matter of Fact: Empirical Perspectives on Ethics’, 121.
5. Merritt, Doris, and Harman, ‘Character’.
6. See Snow, Virtue as Social Intelligence: An Empirically Grounded Theory,
40ff; Narvaez and Lapsley, ‘The Psychological Foundations of Everyday
Morality and Moral Expertise’, 144.
7. See Snow, Virtue as Social Intelligence.
8. See Cervone, ‘The Architecture of Personality’ and ‘Social-Cognitive Mecha-
nisms and Personality Coherence: Self-Knowledge, Situational Beliefs, and
Cross-Situational Coherence in Perceived Self-Efficacy’.
9. Narvaez and Lapsley, ‘Psychological Foundations’.
10. Cervone, ‘The Architecture of Personality’, 186.
11. Ibid.
12. Ibid., 186–187.
13. Ibid., 187.
Notes Toward an Empirical Psychology of Virtue 143
14. Ibid.
15. Ibid.
16. See Cervone, ‘The Architecture of Personality’, and ‘Social-Cognitive Mecha-
nisms and Personality Coherence’; Shadel and Cervone, ‘Evaluating Social-
Cognitive Mechanisms That Regulate Self-Efficacy in Response to Provocative
Smoking Cues: An Experimental Investigation’; and Cervone, Orom, Artistico,
Shadel, and Kassel, ‘Using a Knowledge-and-Appraisal Model of Personality
Architecture to Understand Consistency and Variability in Smokers’ Self-
Efficacy Appraisals in High-Risk Situations’.
17. See Annas, Intelligent Virtue.
18. One might think it too high a standard to impose an Aristotelian conception
of virtue on the kind cashier, but I think it is not too far-fetched to believe
that folk conceptions of virtue express and reinforce central elements of the
Aristotelian conception, such as the notion that virtue requires appropriate
motivation, good deliberation, and regular success in achieving the targets of
virtue. For example, someone who seeks to be kind only in order to ingratiate
themselves with others is not commonly regarded as truly kind but as self-
seeking. Someone whose thoughtlessness always causes her to miss the mark
is regarded as a well-intentioned bumbler, not as a genuinely kind person.
The ‘folk’ often have Aristotelian intuitions about virtue. Philosophical situ-
ationists decry this about the ‘folk’.
19. Cervone, ‘The Architecture of Personality’, 188–196.
20. See ibid., 190–191; see also Cervone, ‘Social-Cognitive Mechanisms and Per-
sonality Coherence’.
21. Cervone, ‘The Architecture of Personality’, 192, 194, figure 6.
22. Ibid., 193–194.
23. Ibid., 194.
24. Ibid., 194, 196, figure 8. In addition to the studies’ reliance on self-reports,
another possible drawback is that the subject population was college stu-
dents. For better or worse, that population seems to be the most readily
available to empirical psychologists.
25. Shadel and Cervone; Cervone et al.
26. Shadel and Cervone, 91.
27. Ibid., 92.
28. Ibid., 93.
29. Cervone et al., 51.
30. Ibid., 51.
31. Ibid., 52.
32. This section draws on Nancy E. Snow, ‘Situationism and Character: New
Directions’; ‘ “May You Live in Interesting Times”: Moral Philosophy and
Empirical Psychology’; and ‘Intelligent Virtue: Outsmarting Situationism’.
33. See Annas, Intelligent Virtue; Kamtekar, ‘Situationism and Virtue Ethics on
the Content of our Character’; and Doris, Lack of Character.
34. Merritt, Doris, and Harman, ‘Character’, 360.
35. Ibid., 263.
36. Ibid., 370–371.
37. John M. Doris, e-mail message to author, January 2, 2012.
38. Merritt, Doris, and Harman, ‘Character’, 375.
39. Ibid., 375.
40. See Snow, ‘ “May You Live in Interesting Times” ’ and ‘Situationism and Char-
acter’.
41. Lapsley and Hill, ‘On Dual Processing and Heuristic Approaches to Moral
Cognition’, 322.
42. Snow, Virtue as Social Intelligence, 43–45.
144 Nancy Snow
43. Narvaez and Lapsley, ‘Psychological Foundations’.
44. Ibid., 144.
45. See Merritt, Doris, and Harman, ‘Character’, 374.
46. Narvaez and Lapsley, ‘Psychological Foundations’, 146–147.
47. Ibid., 147–149.
48. Ibid., 148.
49. Ibid., 145.
50. See Snow, ‘Intelligent Virtue: Outsmarting Situationism’.
REFERENCES
Annas, Julia, Intelligent Virtue (New York: Oxford University Press, 2011).
Cervone, Daniel, ‘The Architecture of Personality’, Psychological Review 111 (2004),
183–204.
———, ‘Social-Cognitive Mechanisms and Personality Coherence: Self-Knowledge,
Situational Beliefs, and Cross-Situational Coherence in Perceived Self-Efficacy’,
Psychological Science 8 (1997), 43–50.
Cervone, Daniel, Heather Orom, Daniele Artistico, William G. Shadel, and Jon D.
Kassel., ‘Using a Knowledge-and-Appraisal Model of Personality Architecture to
Understand Consistency and Variability in Smokers’ Self-Efficacy Appraisals in
High-Risk Situations’, Psychology of Addictive Behaviors 21 (2007), 44–54.
Doris, John M., Lack of Character: Personality and Moral Behavior (Cambridge:
Cambridge University Press, 2002).
Doris, John M., and Stephen P. Stich, ‘As a Matter of Fact: Empirical Perspectives on
Ethics’, in The Oxford Handbook of Contemporary Philosophy, ed. Frank Jack-
son and Michael Smith (New York: Oxford University Press, 2005), 114–152.
Harman, Gilbert, ‘Moral Philosophy Meets Social Psychology: Virtue Ethics and
the Fundamental Attribution Error’, Proceedings of the Aristotelian Society 99
(1999), 315–331.
Kamtekar, Rachana, ‘Situationism and Virtue Ethics on the Content of our Charac-
ter’, Ethics 114 (2004), 458–491.
Lapsley, Daniel K., and Patrick L. Hill, ‘On Dual Processing and Heuristic Approaches
to Moral Cognition’, Journal of Moral Education 37, no. 3 (2008), 313–332.
Merritt, Maria W., John M. Doris, and Gilbert Harman, ‘Character’, in The Moral
Psychology Handbook, ed. John M. Doris and the Moral Psychology Research
Group (New York: Oxford University Press, 2010), 355–401.
Narvaez, Darcia, and Daniel K. Lapsley, ‘The Psychological Foundations of Every-
day Morality and Moral Expertise’, in Character Psychology and Character Edu-
cation, ed. Daniel K. Lapsley and F. Clark Power (Notre Dame, IN: University of
Notre Dame Press, 2005), 140–165.
Shadel, William G., and Daniel Cervone, ‘Evaluating Social-Cognitive Mechanisms
That Regulate Self-Efficacy in Response to Provocative Smoking Cues: An Exper-
imental Investigation’, Psychology of Addictive Behaviors 20 (2006), 91–96.
Snow, Nancy E., ‘Intelligent Virtue: Outsmarting Situationism’, paper presented at
the Pacific Division Meeting of the American Philosophical Association, Seattle,
Washington, April 6, 2012.
———, ‘ “May You Live in Interesting Times”: Moral Philosophy and Empirical
Psychology’, Journal of Moral Philosophy (forthcoming).
———, ‘Situationism and Character: New Directions’, in Handbook of Virtue Ethics,
ed. Stan van Hooft (Durham, England: Acumen Publishing, forthcoming).
———, Virtue as Social Intelligence: An Empirically Grounded Theory (New York:
Routledge, 2010).
10 Natural Virtue and Proper
Upbringing
Candace Vogler
ZÔON LOGISTIKÓN
For better and for worse human life is permeated by our sort of reason—
the talking sort. To suppose that there is such a thing as right reason in the
sphere of the practical—practical wisdom, one might say, or prudentia, or
phrōnesis—is to suppose that we can act well and also that we can err. This
is the sort of thing that Immanuel Kant pointed to in placing human beings
among the addressees of the categorical imperative—the source of good
action in us can, but need not, determine what we do. And it is what Aqui-
nas marked in his initial discussions of those cultivated sources of action,
habits, by remarking, ‘If the form be such that it can operate in diverse ways,
as the soul; it needs to be disposed to its operations by means of habits’.1
That human beings can act ill is no surprise. That whatever a virtue might
turn out to be, talk of virtue is talk of what Aquinas calls ‘habits’ that tend to
practical wisdom, is a commonplace for contemporary Anglophone philoso-
phers working in the area. But how it is that these habits tend to practical
wisdom is less clear, even if we all agree that excellence in the exercise of
practical reason, practical wisdom, shows itself in such habits as individual
justice, courage, and temperance.
Virtues are supposed to be beneficial powers or capacities that work in
accordance with reason to make their bearers good human beings and to
make what their bearers do count as acting well. Virtue gives reason appro-
priate direction in an individual human life, and this shows itself in both
the particular actions of a virtuous adult and in the adult’s overall practical
orientation. It belongs to neo-Aristotelianism to hold, further, that ethical
virtue is not beyond reach for human beings with their wits about them; at
least, it is normally and generally within reach even if many of us fail to get
there. Aristotle certainly thought that it was rare. It is not clear whether he
thought that it was generally possible for people.
I will concentrate on two aspects of some contemporary Anglophone philo-
sophical work on virtue—interest in corrective accounts of virtue (sometimes
called, less helpfully, ‘remedial’ accounts) and a way of reading Aristotle that
has it that a proper upbringing tends to virtue. In order to have a look at these,
146 Candace Vogler
I will rely on Anselm Müller’s discussion of Aristotle’s distinction between
natural virtue and ethical virtue.2 Müller explains this distinction by way of
offering a new interpretation of two still more perplexing aspects of Aristotle’s
work on virtue: his treatment of virtue as a mean state—the notorious doc-
trine of the mean—and his claim that in order to have one virtue, I must have
all of them—the equally notorious unity or reciprocity of the virtues thesis.
I will not be concerned with catalogues of traits that may or may not count as
virtues. A concatenation of psychological traits that happen to be agreeable
or beneficial need not amount to virtue any more than a regime of muscle-
building that targets first one group of muscles, then another, and then
another, need equip us for labor. Virtue is supposed to strengthen us for the
challenges of acting well and living well in roughly the way that cultivating a
strong, healthy body is supposed to equip us for physically demanding work.
The topic at issue—virtue—turns on the cultivation of one’s practical powers
for the sake of acting and living well, not the acquisition of a bundle of lovely
psychological traits. I will accept all of this. I will accept two other Aristotelian
commonplaces about virtue and one more recent commonplace as well:
[The virtues] are corrective, each one standing at a point at which there
is some temptation to be resisted or deficiency in motivation to be made
good.4
She illustrates the point in light of various specific virtues: ‘one may say that
it is only because fear and desire for pleasure often operate as temptations
that courage and temperance exists as virtues at all’,5 and:
As with courage and temperance so with many other virtues: there is, for
instance, a virtue of industriousness only because idleness is a temptation;
and of humility only because men tend to think too well of themselves.
Hope is a virtue because despair too is a temptation . . . With . . . justice
and charity it is a little different, because they correspond not to any
particular desire or tendency that has to be kept in check but rather to a
deficiency of motivation that they must make good.6
Paula Gottlieb, the author of a recent book about Aristotle’s ethics, com-
plains bitterly about Foot on this score.7 Gottlieb thinks that Foot’s view
rests on a picture of human nature as essentially flawed and lacking—a
dubious Christian or Stoical idea that has no place in a properly Aristotelian
virtue ethics. I fear that Foot leaves herself open to Gottlieb’s complaint,
partly by rejecting the reciprocity of virtues thesis. Foot writes:
So far from forming a unity in the sense that Aristotle and Aquinas believed
they did, the virtues actually conflict with each other: which is to say that if
someone has one of them he inevitably fails to have some other.8
In short, unlike Aquinas, Foot rejects the version of the unity of the virtues
thesis that rests upon the reciprocity of the virtues, and there is a hint that
she rejects it because humans tend to be disorderly in their motives, aspira-
tions, and actions and inadvertently at odds with themselves in their efforts
to shape their practical orientations.
Natural Virtue and Proper Upbringing 149
The reciprocity of the virtues thesis is difficult. Anselm Müller gives the
tidiest formulation, reminding us that, for all the attention it has generated,
it is not a topic discussed at length by Aristotle. Here is Müller’s summary
of the view:
The possession of any one ethical virtue, for its application to particular
situations, requires the possession of practical wisdom. But for wisdom
the correct starting points for deliberation are needed, and so therefore
all the virtues. Hence, by the transitivity of requirement, the possession
of any one ethical virtue requires the possession of all the ethical virtues.9
Foot is not alone in finding the thesis implausible, but her rejection of it
seems an outgrowth of her way of advocating a corrective account of virtue.
That is certainly how Terrence Irwin understands some of Foot’s darkest
remarks about virtue. Irwin points out that a corrective account of the virtues
I think that he is right to urge that Foot takes things in this direction. I do
not think that defending a corrective account of virtue inevitably leads to a
rejection of Aquinas’s position, however, partly because the sense in which
virtues are corrective in Aquinas is not happily captured in the phrase ‘piece-
meal remedies’. I don’t know whether one can square a corrective account of
virtue with Aristotle. Aristotle certainly touches upon something connected
to correction in discussing a man who knows the better and does the worse
without becoming vicious. But corrective accounts of virtue hold that the
ordinary and proper operation of fully-fledged ethical virtue is corrective,
not just that virtue can keep us from spiraling toward vice when we are
ungoverned. The corrective in virtue belongs to self-governance, rather than
just kicking in when we fail to govern ourselves appropriately.11
In thinking about the role that virtue has in right reason, it has become fairly
common for philosophers like Foot to move very quickly between Aristotle’s
work on phrōnesis and Aquinas’s work on prudentia. I do not think that it
150 Candace Vogler
is easy to treat the two as a single topic. Writ large, at least, there seems to
be a divergence over the role of intellect in practice. Denis Bradley describes
one pivotal controversy this way:
I will not take up the vexed question about how to widen the interpreta-
tion of the relevant passages in Aristotle in order to make the account look
more like what contemporary Anglophone philosophers seek in seeking
rational foundations for ethics. I will not venture into the fray over whether
to handle the apparent insufficiency by delving more deeply into Aristotle’s
account of desire as a source of human action, or by expanding the account
of reason, or both. It does seem a bit much to argue, as Sarah Broadie does,
that phrōnesis allows ‘continual re-evaluation’ of ends by way of the assess-
ment and re-assessment of means and of means to means and their likely
consequences,13 but, again, this is a question for scholars of Aristotle more
able than I ever will be. I strongly suspect that if corrective accounts of virtue
(as, I take it, Aquinas’s is) make contact with Aristotle, that contact is made
by way of Aristotle’s stress on ethical virtue as a mean state.
My thought is, roughly, this: given some plausible account of the sense
in which the exercise of virtue aims at a mean, we will have found a sense
in which Aristotelian virtue might have a corrective function in its ordinary
operations. Finding something like a corrective function in Aristotelian vir-
tue, in turn, might help us figure out whether an Aristotelian account of
virtue is compatible with the thought that adults can change their ways in
some manner that is regulated by reason, even if we do not think that the
determined scoundrel can be argued into being a good human being.
Müller also provides what I take to be the best account of virtue as a mean
state. His interpretation requires rejection of some of Aristotle’s remarks
about virtue and the mean, stressing that there are two very different senses
Natural Virtue and Proper Upbringing 151
in which Aristotle treats virtue as a mean state. Müller begins by outlining
two aspects of virtue: (1) that virtues are distinguished from each other in
terms of ‘the characteristic dimension of feeling and action in which they
operate’14 and (2) that a fully-fledged ethical virtue is distinct from any ‘nat-
ural virtue’ that shares its characteristic dimension of feeling and action,
and this distinction is a matter of the natural virtue having been ‘shaped by
wisdom’.15
There are various ways in which a practical disposition may be ‘natu-
ral’. The uneasiness that somehow Aristotle’s virtue is nothing more than
the child of a fortuitous marriage of temperament and pre-rational, con-
ventionally sanctioned tendencies in thought and feeling—namely, those
tendencies that happened to be attractive to the adults who governed
us from infancy through adolescence—can be expressed as the worry
that there is no clear line separating natural virtue and ethical virtue in
Aristotle.
Armed with this distinction between two aspects of virtue, Müller
argues that there are two correspondingly different ways of treating virtue
as a mean. On what he calls a ‘one-dimensional’ interpretation of the doc-
trine of the mean, the exercise of virtue involves feeling and acts that are
neither excessive nor deficient. In short ‘this right feeling or action is said
to occupy a position, within the characteristic dimension, that lies between
too much and too little, considering the requirements of a good human
life’.16 Müller contrasts this with what he calls a ‘many-dimensional con-
ception of the mean’, which he associates with the second aspect of virtue.
Müller writes:
Müller calls these ‘critical dimensions’, and argues that, although Aristotle
clearly does not treat these remarks as pointing to two different accounts of
the mean, and although it is in no way obvious how to make sense of the
one-dimensional account, these two ways of assessing intermediacy ought to
be separated, and we ought to accept the many-dimensional account of the
mean. There are problems enough brewing there. For example, it is unlikely
that there will be a distinctive median in each of the many critical dimen-
sions. It is correspondingly implausible to hold that the virtuous person is
the one who tends to hit the mark of doing and feeling neither too much nor
152 Candace Vogler
too little dimension by dimension, even though some of Aristotle’s remarks
seem to point in the dimension-by-dimension direction for the secondary
assessment of the mean. Müller writes:
And even in the examples where it is tempting to think that the idiom of too
much and too little does have a clear point of application, whether one feels
or does too much or too little ‘may depend on the object and other ‘circum-
stances’ of your response’, such that ‘even ‘where the critical dimension can
itself be seen as a variant of the characteristic dimension’, an overall mean
(and virtuous) response need not correlate with an intermediate degree in
that critical dimension’.19
Müller explores various ways of salvaging something of the doctrine of
the mean in the face of the many difficulties he discusses. He concludes that
we should abide by the many-dimensional account of the mean and notice
that the exercise of any one virtue finds its many-dimensional mean only
when it is guided by the claims proper to other virtues: ‘Only then will you
hit the many-dimensional ‘mean’, i.e., correctly decide when and how and
because of what etc. to deploy, and not to deploy, the characteristic response
of that virtue’.20 And that is how a reinterpretation of the doctrine of the
mean sheds light on the reciprocity of the virtues.
So, if friendliness is a natural virtue, then it is not just the abject need to please
everyone all the time. If self-confidence is a natural virtue, it is not the same
as infantile senses of omnipotence or more developed species of arrogance.
If there is natural courage, we will not look for it in the tendency of some
younger children to accept any and all dares issued by older children. In light
of this insight, it is at least possible to counter the tendency on the part of
some contemporary ethicists to suppose that proper upbringing can imbue
human beings with ethical virtue—with fully-fledged, reciprocal, appropri-
ately unified virtue—as though any human being, no matter how generous
and fine, could effect that kind of alteration in any other human being, no
matter how impressionable, needy, unguarded, or eager.
What we can say, armed with Müller’s distinction between ethical and
natural virtue, is that the propriety of a proper upbringing shows itself in
this: proper upbringing contributes to the development of natural virtue in
the young. Whether the youth will then become practically wise, whether
they will be guided by right reason in their maturity, is not settled by looking
to the advantages or disadvantages occasioned by their upbringing.
This is not the conclusion that Müller draws. But it looks to be a conclu-
sion that can be given some plausibility in light of his work, one that can
help us understand how it is (1) that an apparently proper upbringing is
not sufficient for ethical virtue, (2) that there can be virtuous adults whose
upbringing looks to have been about as far from proper as you please, and
(3) that it is nevertheless tremendously important to see to it that children
are well brought up.
RETURNING TO FOOT
AN EXAMPLE
Consider someone who wants to change when she sees that she has a taste
for gossip about those of her colleagues who are not among her friends. She
cultivates avid curiosity about matters that are none of her business. If the
gossip is injurious, she is an avid participant in disgraceful chatter. We are
imagining that she understands the problem and is disturbed by it. Having
seen herself in a very unflattering light, and having forced herself to pay
attention to what she saw, what does she need to do?
Although one normally discusses temperance in connection with control-
ling sense-appetite, appetite can take many forms, and an appetite for gossip
Natural Virtue and Proper Upbringing 155
is no less appetitive for feasting on bits of information rather than slices of
pie. The kind of guidance supplied by temperance is important here and will
counsel discretion. Our unhappy agent might rely upon her temperance to
help her, extending its scope from eating and such to conversation.
Why is it important to curb one’s appetite for salacious talk? Any of sev-
eral things might come into play. Although scintillating gossip tends to come
cloaked as a kind of generosity—a sharing of information with those who
take an interest—the mood of bad sorts of gossip is rarely generous. Nor-
mally, it is anything but generous to those whose lives are serving as fodder
for the gossip mill, which is why it tends to circulate outside the hearing of
those whose doings are under discussion. Our unhappy agent’s generosity
might be at work in identifying and addressing the problem. Being in on
the gossip may, at the very least, make it hard for her to feel comfortable
interacting with a co-worker because she knows things that she has no busi-
ness knowing, and this can lead to a well-grounded sense that she is being
dishonest or unjust in her dealings with her colleague. The unhappiness of
our unhappy agent, that is, could be grounded in her honesty or her justice.
It is not that we can’t imagine someone disengaging from gossip for any
number of reasons that have nothing to do with virtue. I could stop par-
ticipating because no one at my workplace ever does anything interesting.
However, in imagining somebody working to curb an appetite for gossip as
part of a self-improvement effort, we were imagining someone who takes
herself to have ethical reasons to change her ways. The problem isn’t bore-
dom. The problem is flawed character. To so much as identify that problem
correctly, and see its weight and scope in the right way, our agent requires
the strength that she has built in the course of cultivating ethical virtue.
She cannot do this on the basis of natural virtue alone, if Müller is right
about natural virtue. Natural virtues do not delimit each other. It could be a
natural tendency to generosity that is getting her into trouble, after all, and
her challenge might be the challenge of developing properly ethical generos-
ity given some insufficiently delimited natural generosity. What is called for
in order to change this aspect of her practical orientation is precisely the
kind of thing called for in cultivating any sort of virtue—taking the virtuous
action (in this case, disengaging from gossip as an obvious first step).
CONCLUDING REMARKS
Aristotelians hold that it belongs to mature human beings with their wits
about them to pursue good and avoid evil. Aristotelians hold that this bit of
practical direction belongs to our nature and is at work in vice and virtue
alike. To that extent virtuous and vicious action alike are informed by reason.
But right reason—practical wisdom—is charged with shaping and arranging
our pursuits, or motives, and our responses to the end of acting and living
well. In order to do this, practical wisdom works to realize the ends set by
156 Candace Vogler
specific virtues in light of an overarching concern with living a good human
life. The need to continue the work of virtue in light of such matters as having
a taste for gossip is the sort of need that can only be experienced as a need for
ethical self-improvement by someone who already has virtue in her. If Müller
is right, she will have to have ethical virtue, not just natural virtue, in her in
order to undertake ethical self-improvement as an adult.
Drawing on Müller’s work, I have urged that the way to understand the
advantage given by a proper upbringing is in terms of the ways that it guides
children to natural virtue. Developing natural virtue is genuinely helpful in
cultivating ethical virtue since, on Müller’s account, natural virtues are vir-
tues rather than just accidents of impulse and temperament. There is an ele-
ment of constraint in the feelings and acts of those with natural virtue, and
that constraint operates in the dimensions characteristic of the relevant cor-
responding ethical virtues. But a strong assortment of practical tendencies,
each of which provides one-dimensional constraint, does not add up to an
all-around good character. I took it that noticing this was the root of Foot’s
insistence that the virtues do not form a reciprocal unity but instead conflict
with one another. This may well be true of natural virtues as Müller teaches
us to understand them. It should not be true of ethical virtue, partly because
no one ethical virtue can operate well unless it is guided by other virtues.
In the course of discussing these matters, I also came out on the side of
the view that adult moral self-improvement is possible and that it requires
guidance by ethical virtue. This, of course, entails a commitment to a weaker
version of the reciprocity thesis than is common among Anglophone philos-
ophers who accept the reciprocity thesis but take it that I must have perfect
virtue in order to have any virtue at all. All I have to say in response to such
views is: I hope that they are wrong.
NOTES
REFERENCES
Aquinas, The Summa Theologica of St. Thomas Aquinas, Second and Revised Edition
(New York: Benziger Brothers, 1920).
Bradley, Denis J. M., Aquinas on the Twofold Human Good: Reason and Happiness
in Aquinas’s Moral Science (Washington, DC: Catholic University Press of America,
1997).
Broadie, Sarah, Ethics with Aristotle (Oxford: Oxford University Press, 1991).
Foot, Philippa, ‘Moral Realism and Moral Dilemmas’, Journal of Philosophy 80, no. 7
(1983), 379–398.
———, ‘Virtues and Vices’, in Virtues and Vices and Other Essays in Moral Philoso-
phy (Berkeley: University of California Press, 1978), 1–18.
Gottlieb, Paula, ‘Are the Virtues Remedial?’, The Journal of Value Inquiry 35 (2001),
342–354.
———, The Virtue of Aristotle’s Ethics (Cambridge: Cambridge University Press,
2009), Chapter 3, 52–72.
Irwin, T. H., ‘Practical Reason Divided’, in Ethics and Practical Reason, ed. Garrett
Cullity and Berys Gaut (Oxford: Oxford University Press, 1997), 189–214.
Müller, Anselm, ‘Aristotle’s Conception of Ethical and Natural Virtue: How the
Unity Thesis Sheds Light on the Doctrine of the Mean’, in Was ist das für den
Menschen Gute/What is Good for a Human Being?, ed. Jan Szaif and Matthias
Lutz-Bachmann (Berlin, NY: Walter de Gruyter, 2004), 18–53.
11 Kalou Heneka
Timothy Chappell
Think of the glass itself, with its five grand colours stained right through.
It was rougher than ours, thicker, fitted in smaller pieces. They loved
it with the same fury as they gave to their castles, and Villars de Hon-
necourt, struck by a particularly beautiful specimen, stopped to draw it
on his journeys, with the explanation that ‘I was on my way to obey a
call to the land of Hungary when I drew this window because it pleased
me best of all windows’.1
Say what you like, so long as it doesn’t stop you from seeing how things
are. (And if you see that, there are plenty of things that you won’t say.)3
This chapter is a meditation on the first two of these epigraphs, in the light
of the third. Continuing a project I have begun elsewhere,4 of suggesting
ways in which we might expand our repertoire of (recognized) ethical con-
cepts, it explores an idea present in both the first two passages. This is the
idea that tou kalou heneka, ‘for the sake of The Fine’, or ‘The Beautiful’, is a
name for more than one important kinds of practical-rational intelligibility.
One lesson of these explorations is that our practical reasons are much less
structured, much more piecemeal, particular, and bitty, than moral philoso-
phers generally like to think. Another lesson is a way to answer the familiar
question ‘Why be moral?’. It is with that question that I begin.
Typically, when people ask ‘Why be moral?’, they are asking for an explan-
atory reduction of The Moral to The Prudential, that is, for an argument that
we have reason to do what is moral that shows that it is in our prudential
interest. The question takes it as read that moral reasons are problematic
in some way, whereas prudential reasons are not. Moral reasons require
grounding, their force for us is somehow not obvious; whereas it is obvious
how prudential reasons are reasons for us. Their force and applicability is
self-evident, or something like self-evident.
Kalou Heneka 159
So far as I can see, we could just as well assume the inverse: that it is
moral reasons that are unproblematic and prudential reasons that require
explanatory reduction. Then the pressing question would not be ‘What (pru-
dential) reason do we have to be moral?’. It would be ‘What (moral) reason
do we have to be prudential?’.
A third possibility: we could insist that moral and prudential reasons
are both unproblematic so that no explanatory reduction is called for in
either direction. Or indeed, to add the fourth permutation in the table, that
both are problematic and unobvious—which is more or less the view I shall
defend here.
II
Suspicion about the moral as a category can take a number of forms. What
I have labeled the prudentialistic presumption is itself one of them: here the
suspicion is that moral reasons, to be made rationally intelligible, need to be
explanatorily reduced to practical reasons in the other category, the category
of the prudential. As will be clear already, I have my suspicions about this
sort of suspicion.
A different sort of suspicion about the moral is more to my present pur-
poses. This is the sort of suspicion you get in Anscombe, MacIntyre, and
Williams: the suspicion that ‘the moral’ just isn’t the name of any unitary
category that does much if any interesting work in justifying and explaining
what actual good agents characteristically do. At the deliberative level, there
may be some things that such agents do in which such agents are motivated
by the thought ‘Because it’s moral’. But such agents have lots of other moti-
vating thoughts—‘Because fairness requires it’, ‘Because I promised’, ‘Because
162 Timothy Chappell
she’s my wife’, ‘Because we are friends’, ‘Because you are sinking’, ‘Because
they are starving’, ‘Because otherwise he will be disgraced’, ‘Because it is suf-
fering’, ‘Because my last delivery was a no-ball’, and so on indefinitely; the
V-thoughts, as Rosalind Hursthouse calls them. Clearly none of these is just
another way of thinking the ‘Because it’s moral’ thought; arguably many of
them are not even consistent with that thought, and/or with the conscious
entertainment of that thought. So it isn’t clear why the ‘Because it’s moral’
thought should be supposed to be a specially deep or important form of moti-
vation for good agents.
Similarly at the criterion of rightness level, there may be a few things that
good agents characteristically do in which such agents are justified simply
and directly by the consideration ‘Because it’s moral’. But such agents can
have lots of other justifications, many of them very different from this jus-
tification and nearly all of them more informative. As before, what they do
may be justified by considerations about fairness, or promises, or someone’s
being my wife, or a friendship, or shipwreck, or famine, or disgrace, or
pain—and so on indefinitely. Hence it is no clearer why the consideration
‘Because it’s moral’ should be a particularly special or basic justification
than it is why the thought ‘Because it’s moral’ should be a particularly spe-
cial or basic motivation for good agents. (Or should we take the domain
of practical justification to be theoretically unified in a way the domain of
practical deliberation conspicuously isn’t? It is hard to see why we would;
more about that in section III.)
At both the deliberative and the justificatory levels, to insist on the prior-
ity or basicity of appeals to The Moral looks like mere stipulation. There is
no obvious explanatory gain in this redescription. If it is taken, as often, to
be the right way to marshal the phenomena, to the exclusion of other ways
of looking at them that may prove equally or even more fruitful, then we risk
explanatory loss, too: this insistence may well obfuscate the real structure of
our deliberative and justificatory practices.
These phenomena about the multifariousness of the moral have been
extensively studied by ethicists and moral psychologists. The thought is
familiar—and I think true—that ‘the moral’ is not really the name of any
wide-ranging and sharply defined category of practical reasons at all. What
may be less familiar, but I think is equally true, is that something analogous
applies to the prudential.
‘The prudential’ is supposed to be about what is advantageous, or benefi-
cial, or in the agent’s own interest, or what furthers the agent’s well-being.
There are serious ambiguities in all of these notions. ‘Advantage’ and ‘bene-
fit’ and ‘interest’ and ‘well-being’ are concepts of which there are indefinitely
many competing accounts, both philosophical and informal. These words
can mean many different things, and there is no particular reason—aside
from theory—to expect their extensions to converge at all neatly. In perfectly
ordinary senses of the words, forgoing a pay-rise to impress my boss may
be to my advantage but does not benefit me, while health-threatening and
Kalou Heneka 163
anxious dedication to my work as a novelist or a famine-relief worker may
(in a way) be in my interest but does not further my well-being.
‘But that means I have an implicit notion of benefit, or advantage, or
interest, or well-being, such that I value one of these more than the other
according to it. It means I think that one of these scores higher on some scale
than the other’. There is a strong inclination to say something like this. Yet
there is no good reason why we should say something like this rather than
saying, simply, that the agent prefers one of these options to the other. Phi-
losophers are convinced that there must be some scale of well-being against
which to calibrate such preferences. The fruitless, but not yet abandoned,
search for that scale is not a search for something objectively ‘out there’
that, once found, could settle questions about the nature of ‘prudential well-
being’ in favor of a unitary account of it. On the contrary, it depends on the
prior assumption that prudential well-being has a unitary nature—the very
assumption that I am questioning here.
Nor is there any reason—again, aside from theory—to think that these
‘prudential’ notions can easily or conveniently be kept clear of what the
proponent of the moral/prudential contrast would like to call moral con-
notations. This is especially obvious with ‘well-being’, but a similar ‘mixing
of the prudential and the moral’ can be imaginatively effected just by adding
real on the front of the other terms—‘real advantage’, ‘real benefit’, ‘real
interest’. If the prudential is to be a category clearly distinct from the moral,
having to do with the agent’s individual well-being, and fit to serve as an
explanatory foundation for the moral, then this cross-infection of the moral
and the prudential cannot be allowed. But suppose well-being turns out to
be a notion that cannot be properly understood except when it is given a
moral loading. This is likely to be how things turn out if, as seems plausible,
inquiring what to count as well-being is not a value-neutral anthropological
enterprise but a key part of constituting one’s own moral character.8 In this
case the cross-infection of the moral and the prudential is inevitable. And so,
to switch metaphors, the idea of an exclusive moral/prudential distinction is
already holed below the waterline.
Alongside these two problems, there is a third. The very idea of the pru-
dential, as most commonly understood, appears to rest on an obvious false-
hood. In its most typical form, the category of the prudential is meant to fit
both of two criteria: (a) It is supposed to be about people acting in pursuit of
their own interest (or welfare or advantage or whatever). (b) It is supposed to
be definable by exclusion from the category of the moral: the moral and the
prudential are supposed to be an exhaustive and exclusive pair of categories
that between them cover every case of having a practical reason. But (a) and
(b) together imply that whenever someone acts on a non-moral reason, he
acts in pursuit of his own interest. Since ‘non-moral reason’ and ‘in pursuit
of his own interest’ are both extremely vague phrases, it is difficult to conclu-
sively refute this. But on any commonsensical understanding of the words, it
is false. The man who works himself half to death to please his beloved, or to
164 Timothy Chappell
perfect his conducting of the Eroica Symphony, acts on a non-moral reason:
his reason is romantic, or musical, or what you will. Does that mean that
he acts ‘self-interestedly’? A natural thing to say is that what he is doing is
against his own interest; he is making a great and dangerous sacrifice of his
own interests for those of his lover or his art. In this sense it is obvious that
not every non-moral reason is a reason of self-interest. So evidently not all
practical reasons are either moral or else prudential—unless the prudential
becomes so wide a category that it is not really a category at all.
III
The moral has no clear structural unity; nor does the prudential, which is
also very hard to articulate without the importation of moral elements that
threaten the supposed exclusiveness of the moral/prudential distinction; the
supposed exhaustiveness of that distinction seems highly questionable, too.
These are the basic problems about ‘the moral’ and ‘the prudential’.
These are the reasons that motivate me in adopting what according to the
previous Table we may call skeptical dualism, though really the position is
so much a skepticism about the moral and the prudential that it is not really
a dualism at all—the whole point is that we can’t divide the phenomena at
all neatly into just these two classes. As Plato in effect pointed out in the
Republic, these problems are not well finessed by a Thrasymachean cyni-
cism about the moral, precisely because of the instability of the notion of the
prudential—especially when that notion is not allowed to ‘cross-infect’ with
the moral in the way described previously.
Nor are they well finessed by the kind of moralism we get in Kant, who
will happily accept that the prudential is a shapeless mess, provided he is
allowed to say that there is something that unifies the moral—namely, the
move to universalizability. Familiarly, the trouble with this, as I have argued
elsewhere,9 is that it isn’t at all clear that the universalizable and (what we
might intuitively call) the moral coincide.
Some other accounts of how to characterize ‘the moral’ succumb even
faster to even more obvious problems. William Frankena, for example, stip-
ulates in one well-known discussion that one is not ‘taking the moral point
of view’ unless, inter alia, ‘one is willing to universalize’, and one’s ‘reasons
for one’s judgments consists of facts about what the things judged do to
the lives of sentient beings in terms of promoting or distributing non-moral
good and evil’.10 Of course (see my third epigraph) we can use the word
‘moral’ however we like, but it is hard to see the point of using it in such a
way that non-universalizable moral judgments, and moral judgments about
the environment, are ruled out by definition. A similar objection can made
to Catherine Wilson’s more recent suggestion that ‘There is an anonymity
requirement on moral theorising’. If she wants to use the words this way,
she is free to propose that we should see any endorsement of a partial norm
Kalou Heneka 165
as ‘ideology’, not as ‘moral theory proper’. The trouble with her proposal
is that it just seems like a stretching of the sense of ‘moral’, and indeed a
case of ideology.11 Wilson’s anonymity constraint entails that there can be
no such thing as an individual’s moral style—no such thing as approaching
the problems that arise for practical decision in line with any particularities
of character. But this is no trivial loss to our ethical thought; it is a disabling
deprivation.12 In reality, I suspect we barely know what it would mean to be
her sort of moral reasoner.
My point is not that, since neither notion can be cleanly and exclusively
defined, we should just junk the notions of the moral and the prudential. On
the contrary, I think we should rehabilitate them. There are some actions—
not many, but some; perhaps handing in a lost tenner to the police sta-
tion—that really are motivated and justified by nothing else but the thought
‘Because it’s moral’. With these actions we find the legitimate scope of the
notion of the moral. Similarly, there are some things that people do—not
everything, not perhaps even all that many things, but some: perhaps apply-
ing for a stop-gap job to pay the mortgage ‘just till something better comes
up’—in which what motivates them, and perhaps even justifies them, really
is solely and simply a concern with the agent’s own advantage or interest
in some clearly non-moral sense. With these actions we find the legitimate
scope of the notion of the prudential. The notions of the moral and the
prudential work fine in these, their home territories. The question to the
systematizing moral theorist is why we should feel any impulse to insist that
these are the only two basic-level notions that we can deploy to think about
our practical reasons in any territory.
IV
Recall the sheer variousness of the things that can appear explanatorily basic
to people whose rationality is, we would normally say, indisputable. The
more we understand this variety, the less we will be tempted by the thought
that, in ethics, we face a large and pressing task set for us by the opening
question, ‘Why be moral?’: the task of providing an explanatory reduction
of the moral to the prudential. As if the prudential were somehow the uni-
versal currency of practical reasoning; as if the prudential were an exchange
for and a measure of every other sort of practical reason. If my argument so
far is right, nothing of this sort can possibly be true. What we should pursue
in the theory of practical reasons is not chimerical unificatory projects like
this; it is an exploration and assessment of the diversity and disunity of our
actual practical reasons.
One example, the example I discuss in ‘Glory as an Ethical Idea’, is practi-
cal reasons having to do with glory. Another example is humor. A third—the
one that I want to discuss in the rest of this chapter—is, What if someone
were to say ‘I did it because it was beautiful’?
166 Timothy Chappell
One form of this is in effect what Villars de Honnecourt is saying in my
first epigraph. The possibility of saying it, and saying it in many different
ways, is arguably what Aristotle affirms in my second.
‘In many different ways’: of course there is a difference between Villars
de Honnecourt, who does what he does because some object (the window)
is beautiful, and Aristotle’s andreios, with whom the point is, I take it, that
what he does is to-be-done because it, his action, is beautiful or fine, kalos.
Different again is a case that Simone Weil describes:
It is the beauty of the world which compels the man who is drained
empty, the man who has spent all his inheritance, all his energy, to
remember that his father’s slaves have more of a share in the good than
he who is the son. The share that things have in the good, the wages of
the slaves of the Father—it is beauty.13
Here someone changes his whole way of living (Weil is thinking of the gospel
parable of the Prodigal Son, Luke 15.11–32) because he comes to see his
present life in a new light that is cast over it by the illumination of something
like the Platonic ideal of Beauty. Further cases, coming shortly, introduce
still further differences.
I do not want to discount these differences but, on the contrary, to empha-
size them. I am not proposing that, alongside the two familiar monolithic
categories of practical reasons, The Moral and The Prudential, we recognize
a third monolithic, capitalized category, Reasons To Do With Beauty. Still
less, if less is possible, do I mean to suggest that this category should be
called The Aesthetic. I am not saying that if any practical reason is neither
Moral nor Prudential, then it must be Aesthetic. As far as I am concerned,
the more other categories of practical reason there are—and the less mono-
lithic and indeed capitalized they are—the better.
These caveats aside, the idea that the beautiful can give us reasons is—as my
first and second epigraphs demonstrate—of course not new. In a classical
Greek context, it is not even, so far forth, controversial. Aristotle, Pericles,
and Plato disagree about many things; not about this. With Aristotle’s the-
sis, in my second epigraph, that the brave man does his brave deeds kalou
heneka, ‘for the sake of the fine’ (or ‘the beautiful’ or ‘the noble’, as it is also
sometimes translated), compare Socrates’s claim at Gorgias 477c8 that what
is worst (aiskhiston) is also what is ugliest, and Pericles’s famous words in
the Funeral Oration:
They resigned to hope their unknown chance of happiness; but in the face of
death they resolved to rely upon themselves alone. And when the moment
Kalou Heneka 167
came they thought it more beautiful (kallion hêgêsamenoi) to stand firm
and die, rather than to fly and save their lives; they ran away from the
word of dishonour (to men aiskhron tou logou), but on the battlefield
their feet stood fast, and in an instant, at the height of their fortune, they
passed away from the scene, not of their fear, but of their glory (doxês).14
The ideas that ‘I did it because it was beautiful’ can be a rationally intel-
ligible motivation, and that the beauty of ways of acting can be an important
aspect of their goodness are not exclusively pagan Greek ones either. If Plato
speaks of the beauty of good people’s actions (Symposion, 210c), so too does
Aquinas:
When the train began to move, and I walked away, I felt as though I
were leaving behind me all the beauty and all the joy in the universe.
Something of God’s universal love has rubbed off on Mother Teresa,
giving her homely features a noticeable luminosity; a shining quality.
She has lived so closely with her Lord that the same enchantment clings
about her that sent the crowds chasing after him in Jerusalem and Gali-
lee, and made his mere presence seem a harbinger of healing. Outside,
the streets were beginning to stir; sleepers awakening, stretching, and
yawning; some raking over the piles of garbage in search of something
edible. It was a scene of desolation, yet it, too, seemed somehow irradi-
ated. This love, this Christian love, which shines down on the misery
we make, and into our dark hearts that make it; irradiating all, uniting
all, making of all one stupendous harmony. Momentarily I understood;
then, leaning back in my American limousine, was carried off to break-
fast, to pick over my own particular garbage-heap.
I want to suggest that the answer to ‘Why be moral?’ is quite often ‘Because
that is the beautiful thing to do’. It’s not that the moral act is itself pruden-
tially disastrous, but just happens to be, unfortunately enough, one of a class
168 Timothy Chappell
to the whole of which we are somehow committed, if we are committed to
any part of it—as theories like rule-consequentialism and Gauthier’s con-
tractualism sometimes suggest. Nor is it that the moral act is prudentially
advantageous in some way—just a very obscure way, one which is consis-
tent with the fact that the moral act is attended with terrible penalties like
those that Hans and Sophie Scholl faced, or those described by Callicles at
Gorgias 486b. We need not think that there is any prudential advantage, in
any sense, in the gravely sacrificial moral act. At least sometimes, advantage
simply isn’t the point. It is rather that the moral act demands to be done even
if it does involve a grave sacrifice—just because it is beautiful.
Perhaps this appeal to to kalon, The Beautiful, is the answer to the puzzle-
ment expressed by the person who said of Sophie Scholl and those who
suffered with her that ‘the fact that five little kids, in the mouth of the wolf
where it really counted, had the tremendous courage to do what they did,
is spectacular to me. I know that the world is better for them having been
there, but I do not know why’.17
Perhaps it is also the best answer to the difficulty that Philippa Foot is
struggling with in the following two paragraphs:
One may think that there was a sense in which the Letter-Writers18
did, but also a sense in which they did not, sacrifice their happiness in
refusing to go along with the Nazis. In the abstract what they so longed
for—to get back to their families—was of course wholly good. But as
they were placed [facing imminent execution for involvement in the
German resistance to Hitler] it was impossible to pursue this end by just
and honorable means. And this, I suggest, explains the sense in which
they did not see as their happiness what they could have got by giving
in. Happiness in life, they might have said, was not something possible
for them [. . .]
Yet this is not the heart of the matter. For supposing that they had
been offered a ‘Lethe-drug’ that would have taken from them all future
knowledge of the action [of giving in to the Nazis]? They would not have
accepted. And there would have been a way in which they would not
have felt that happiness lay in acceptance. ((To see it as happiness they
would have to have changed, and would not have accepted the prospect
of such a change [. . .] one would not wish for the sake of friends one
loved that ‘in the tight corner’ they would be able to forsake their virtue
in time.))19 [. . .] Happiness isolated from virtue is not the only way in
which the concept is to be found in our thoughts. The suggestion is,
then, that humanity’s good can be thought of as happiness, and yet in
such a way that combining it with wickedness is a priori ruled out.20
VI
And being in Bethany in the house of Simon the leper, as he sat at meat,
there came a woman having an alabaster box of ointment of spikenard
very precious; and she brake the box, and poured it on his head. And
there were some that had indignation within themselves, and said, Why
was this waste of the ointment made? For it might have been sold for
Kalou Heneka 171
more than three hundred pence, and have been given to the poor. And
they murmured against her. And Jesus said, Let her alone; why trouble
ye her? she hath wrought a beautiful work on me. For ye have the poor
with you always, and whensoever ye will ye may do them good: but me
ye have not always. She hath done what she could: she is come afore-
hand to anoint my body to the burying. Verily I say unto you, Where-
soever this gospel shall be preached throughout the whole world, this
also that she hath done shall be spoken of for a memorial of her. And
Judas Iscariot, one of the twelve, went unto the chief priests, to betray
him unto them.22
According to Jesus the justification of the woman’s action lies simply and
straightforwardly in its beauty. (Kalon ergon êrgasato en emoi, influentially
mistranslated in the King James Version as ‘she hath wrought a good work
on me’. That the deed’s beauty is the point, as we would expect from the
word kalon, is underlined by a detail that St John adds in his version of the
story: ‘the house was filled with the fragrance of the perfume’, Jn 12.3.) The
objection to the woman’s action is a form of the objection that its cost is too
high; an objection that we will probably hear a lot against actions motivated
by beauty. The objection is that the action—in which the woman was mak-
ing a huge financial sacrifice—is wasteful and inefficient from the point of
view of public utility. Jesus responds that there are cases where beauty over-
rides public utility. This is one of them because of the grace (both theologi-
cal and colloquial) and the exactly apposite symbolic value of the woman’s
action. The passage gives us a concise, but rather rich, example of how a
debate about the relative justificatory powers of beauty and utility might
be intelligible: for what Jesus says to justify the woman’s deed is perfectly
intelligible, even if we do not accept it. (To many today Jesus’s justification
of the action as a symbolic preparation for the arrest, torture, mock trial,
and mob-justice execution that were about to happen to him will no doubt
be a scandalous one, just as it seems to have scandalized some of his hearers
at the time: Judas, for instance, for whom ‘the poor you have always with
you’ seems to have been the last straw that provoked that good utilitarian
into betrayal.)
VII
NOTES
REFERENCES
Beyond (Neo-)Aristotelian
Virtue Ethics
12 A New Metaphysics for Virtue Ethics
Hume Meets Heidegger
Christine Swanton
Call (I) the Intelligibility Thesis. The Intelligibility Thesis is about the very
existence of ethical properties as ethical properties. A similar thesis applied
to redness is a thesis about the existence of redness as a phenomenological
A New Metaphysics for Virtue Ethics 179
‘secondary quality’: it is not a thesis about the existence of redness as a
‘microscopic textural property’ of the surface of an object.5
(I) has generally been construed as a claim that ethical properties are
relational, rather than monadic, as in (R):
(R) A person’s being virtuous consists in her virtuousness evoking some
relevant response.
Call (R) the Relational Thesis. (R) conceives of response dependent proper-
ties as relational rather than monadic. To answer the question what and
whose responses are relevant the concept of qualified agent is invoked. So
we have:
Call (Q) the Qualified Agent Thesis. The Relational and Qualified Agent
theses have been subjected to criticism. However, invoking the thought of
Hume and Heidegger, I argue that Thesis (I) does not entail either (R) or (Q).
The first problem concerns the Relational Thesis. Wiggins notes after all
that ‘redness is an external, monadic property of a [British] postbox’.6 The
second arises when we appreciate that understanding normative properties
as response dependent is not sufficient for understanding them as norma-
tive. We want to say, for example, that a trait V is a virtue if and only if V
merits the relevant response (and does not, for example, just cause it). The
Qualified Agent Thesis is then invoked to give an account of what counts as
a merited response.
However, when we think of a property as meriting a certain type of
response, it seems natural to think of it as a monadic property that makes
certain responses justified. What makes it the case that a trait merits status as
a virtue does not seem to depend on the responses of an agent, even a quali-
fied one, but on certain facts, such as the trait actually systematically pro-
ducing good consequences. Secondly, competent agents may be wrong about
those key facts. Competence is not the same as omniscience. Competent,
indeed virtuous, agents have access to at least some of the more important
and relevant background theories of human nature and have wise, relatively
informed views about long-term consequences, but some of these theories
and beliefs may be false or incomplete.
These objections to (Q) on my view are decisive. The solution is not how-
ever to reject response dependence but to reject (R) and (Q) as an analysis
of the response dependent nature of ethical properties. ‘Virtue’ on the view
outlined below is a response dependent concept in sense (I), but (I) is deemed
to entail neither (R) nor (Q), which are false. These claims rely on three central
180 Christine Swanton
theses of Heidegger that I believe are shared by both Hume and McDowell.
These theses are as follows:
If we cannot find vice ‘in the object’ understood in either of the two
ways described in (a) and (b), where else should we look? Hume answers
that not ‘till you turn your reflexion into your own breast’ will you find
what you are looking for (T 468). Alas, it is easy to miss the point. It is
tempting to read Hume as saying this: looking at the blood, etc., causes
distress, disgust, horror, and so forth, so moral judgment is nothing but
the following matter of fact: I am distressed at the willful killing; or if not
a fact of this kind, then an expression of our emotion or a projection of
them. But this kind of analysis misses the point because it is still trying to
use reason in the narrow sense (in the form of causal reasoning) coming
up with forms of subjectivism or skepticism about the genuine objectiv-
ity of ethics. In short at play here is ‘empiricistic naturalism’, one of the
variants of the pretensions of reason in the narrow sense. Such an analysis
has failed to grasp the central point of Hume’s philosophy: reason in the
narrow sense should know its limits.
What then should replace this notion of reason in the ethical domain?
What does Hume mean by the need to turn your reflection into your own
breast? In very general terms we need to turn our reflection to our passions
in order to understand the nature of moral properties as response depen-
dent.14 This is indeed how I have interpreted him,15 with the following major
proviso. On my view he should be interpreted as a response dependence
virtue ethicist: a form of virtue ethics not so far salient in the virtue ethical
literature. This is important to secure the normativity of Hume’s response
dependence as a form of substantive moral philosophy in a suitably objec-
tivist tradition. However, in this chapter I wish to tie his response depen-
dence not to his normative theory understood as a type of virtue ethics, but
to a Heideggerian understanding of the very nature of Hume’s response
dependence at a metaphysical level. Most importantly, we need to see how
Hume’s response dependence, so interpreted, can be seen as realist (in Pig-
den’s sense)16 without being a species of ‘empiricistic naturalism’, where
rationality in ethics has to be understood through the operations of reason
in the narrow sense.
Understanding the limits of reason in Hume’s narrow sense is the essen-
tial first step to understanding (I). To make ethics both intelligible and truth
apt we need to understand the distinctive mode of intelligibility of ethics.
Heidegger’s metaphysics of engagement, elaborated in his hermeneutic phe-
nomenology, gives us explicit resources for securing that understanding.
A New Metaphysics for Virtue Ethics 183
ALETHEIA: UNCONCEALMENT
It is well known that at the end of Book I of the Treatise Hume apparently
lapses into skeptical despair being ready to ‘reject all belief and reasoning’
(T 268) as capable of furnishing answers to questions ranging from ‘from
what causes do I derive my existence?’ to ‘whose anger must I dread?’ (T 269).
He concludes that ‘since reason is incapable of dispelling these clouds, nature
herself suffices to that purpose, and cures me of this philosophical melancholy
and delirium’ (T 269). Hume was onto something here: not only is reason (in
the narrow sense) entirely insufficient for practical competent engagement
in the real world, but our very understanding of what it is to inhabit such a
world must be furnished by nature much more broadly understood. Indeed
Hume himself claims that ‘the understanding [reason in the narrow sense],
when it acts alone, and according to its most general principles, entirely
subverts itself’ (T 267). But it is Heidegger who provides at the very heart
of his philosophy the metaphysical justification for such a view. Hume sim-
ply proceeds to act on the cure and writes Book II and III as if no skeptical
despair had ever occurred.
Hume’s problem is that he implicitly (at the end of Book I) privileges scien-
tific modes of understanding understood through reason in the narrow sense
so that other forms of practical engagement with the world do not have as
much claim as the scientific to make reality intelligible. Heidegger resolves
that problem by not privileging the scientific mode. On this view as McDowell
suggests there are multiple forms of basic construals of the world as a whole,
yielding multiple types of standard for ‘objective correctness’.
Our task in this section is to offer a Heideggerian understanding of the
metaphysics of (I). How can we understand Hume’s point that scientific
understandings alone cannot bring the world of ethics into being, and yet
claim that that world is both a world of nature and has objective normative
reality? In Heideggerian terms, to determine what he calls the ‘worldhood’
or ‘world’ of ethics we must understand that particular mode of comport-
ment to the world that brings its ethical aspect into being. This idea is at the
core of Heidegger’s notion of aletheia, which we might call (following normal
translations of Heidegger) openness, uncoveredness, unhiddenness, or uncon-
cealment. Fundamental orientations or attunements are ways of what Hei-
degger calls unconcealing: opening or revealing the world as a whole in one
or other of its aspects. These constitute backgrounds of understanding (com-
portments), including emotional understanding: forms of ‘disclosure’. Such
a revealing is what Heidegger calls a ‘clearing’ (Lichtung), which is a way of
securing intentional access to the world as a place of a certain type. Wrathall
puts the point this way: ‘Unconcealment, when understood as a clearing, does
not name a thing, or a property or characteristic of things, or a kind of action
we perform on things, or even the being of things. It names, instead, a domain
or structure that allows there to be things with properties or characteristics, or
modes of being . . . It is something like the space of possibilities’.17
184 Christine Swanton
Such a domain or structure Heidegger calls a ‘worldhood’, even ‘world’
(Welt). Truth as correspondence, a property of judgments or assertions,
being a relation of ‘agreement of the judgment with its object’, presupposes
that there has been a ‘Lichtung’ within which we have a mode of intelligi-
bility.18 This conception of truth Heidegger also calls truth as correctness.
Crucially, truth as correspondence presupposes aletheia: ‘The correctness
of seeing and viewing things, and thus of definition and assertion [truth as
correspondence] is grounded in the particular manner of orientation and
proximity to beings, i.e. in the way in which beings are in each case unhid-
den. Truth as correctness is grounded in truth as unhiddenness’.19
We are now in a position to see what is wrong with orthodox interpreta-
tions of the passage from Hume concerning ‘seeing the vice in the object’.
Hume’s claim that it is not until you turn your reflection into your own
breast that you will see the vice in the object is a claim concerning (I). Emo-
tional construal is essential for the world of ethics to be intelligible. There is
no ethical object that can be the object of our representation and epistemol-
ogy if we only construe the object through the disclosures appropriate to
forensics (the knife in the back, the heart stopped, the blood on the floor).
There is no ‘space of possibilities’ for ethics at all. The Qualified Agent
Thesis, in any version, cannot even apply till the ethical object is brought
into being. If it is not, we are left only with our feelings toward the object
forensically conceived; we are on the road to ethical subjectivism.
It turns out that the process of adequate uncovering has many aspects. This
complexity is explicit in Heidegger’s depiction of aletheia as a multiple unity
having four aspects called by Heidegger ‘The space of the four-fold unitary
openness’.20 In the remainder of this chapter I specify the four aspects or modes
of the four-fold of openness and apply each of them to Hume. In so doing, I
show the manner in which Hume’s moral sentimentalism can be understood as
not in the least subjectivist, private, immune to critical analysis, and anti-realist.
With the help of Heidegger’s analysis, I show how Hume’s moral sentimental-
ism provides a very rich account of (I). The four-fold of aletheia ‘uncovers’
ethics: the ‘mode’ or ‘character of being’, which is the ‘worldhood’ of ethics.
For example, we might say, more or less controversially, if there were no sen-
sibilities capable of experiencing color, there would be no color properties; if
A New Metaphysics for Virtue Ethics 185
there was no benevolence, love, or empathy, there would be no moral sense
and thus no ethical properties; if there were no sensations and reason proper
to the understanding (in Hume’s sense), there would be no scientific proper-
ties; if there were no aesthetic sensibilities, there would be no beauty. With-
out sensibilities and ‘Verstehen’ (understanding in a broad sense) responding
to powers and creating frameworks of disclosure of aspects of the world,
and thereby securing an intentional relation to them, there would only be,
in Heidegger’s language, the mystery of Being. This is a reality conceived
of prior to or independently of its powers to affect sensibilities and thereby
allow for the possibility of being disclosed.
Concealment, hiddenness, of the world of ethics is not to be confused with
falsity or illusion, which come on board only with truth as correspondence.
Rather, reference to the worldhood of ethics has not even been secured. Hume
presupposes this distinction in his discussion of such beings as the ‘fancied
monster’ of the Enquiries and the ‘perverse’ individual of the Skeptic: ‘where
one is born of so perverse a frame of mind, of so callous and insensible a dis-
position, as to have no relish for virtue and humanity, no sympathy with his
fellow-creatures, no desire of esteem or applause; such a one must be allowed
entirely incurable, nor is there any remedy in philosophy’.21 Were the world
to be entirely filled with such persons we would have a world of moral blind-
ness analogous to a world of color blindness where no one had color vision.
What is it not to be blind to the ethical realm? For Hume, the disclo-
sure of the world of ethics requires a moral sense, a sentiment of approval;
more particularly, a passion of ‘fainter and more imperceptible love’ (T 614;
T 3.3.5.1), constituting our emotional sense of virtue properties. The condi-
tions of possibility of such a sense (if it is to be understood as a moral sense
disclosing that aspect of reality) are certain emotional responses and psy-
chological capacities. The relevant responses and capacities are benevolence,
and benevolent empathy and sympathy constrained by self-love. Without
benevolence, we could not have an orientation to the world fundamental
to ethics on Hume’s view: desire for another’s good. Without sympathy, we
could not transmit this desire sufficiently widely: humanity toward strang-
ers and those distant from us is also an orientation fundamental to ethics.
Furthermore, if there were in addition no passion of self-love or ‘love of
life’ (that is, one’s own life) we could not make sense of our approval of the
self-regarding virtues (virtues that render people ‘serviceable to themselves,
and enable them to promote their own interests’ (T 587)), and the world of
ethics disclosed by the moral sense would then take on an entirely different
cast. We would see ourselves wholly as instruments in the service of others’
needs. This would be a world without proper agency and not a world of
morality as we would understand it for Hume. This is not to say that these
are the only emotions relevant to ethics, but they constitute the background
(horizon of disclosure) within which ethics makes sense.
It is Hume’s fundamental contention that the sensibilities requisite for
understanding the scientific nature of the world, namely sensations and the
186 Christine Swanton
reason proper to the understanding, are not sufficient to allow ethical prop-
erties to ‘be out in the open’ or unconcealed. What such sensibilities and
reason aim to do is discover causal relations and ‘eternal’, ‘immutable’ fit-
nesses and unfitnesses of things that would impose obligations that would
be ‘the same to every rational being that considers them’ (T 456), regardless
of the kinds of sentiments (if any) that form their constitution. However,
for Hume, there are no such eternal fitnesses in the moral, prudential, or
aesthetic domains. This is the point of Hume’s claims that ‘ ’Tis not contrary
to reason to prefer the destruction of the whole world to the scratching of
my finger’ (T 416) and ‘ ’Tis not contrary to reason for me to chuse my total
ruin, to prevent the least uneasiness of an Indian or person totally unknown
to me’ (T 416).
Here Hume means by ‘reason’ pure emotionless reason of the kind pos-
sessed by all rational beings regardless of their emotional constitutions.
Human beings do possess such reason, provided we understand it as the
operations of the understanding, a faculty ‘considered apart from any pas-
sions and any feelings of pleasure and pain’. However, reason in this narrow
sense is simply not able to deconceal what Hume calls ‘natural fitnessess’ in
the practical domains. Indeed the previous claims about what is not ‘con-
trary to reason’ are a reductio of the idea that reason in the narrow sense is
fit to disclose such natural fitnesses.22
(C2) is subdivisible into several criteria on the assumption that several types
of feature, not reducible to consequences for the happiness of mankind, make
it fitting that ‘immediate taste or sentiment’ be produced.24 Both these criteria
can be applied in a warranted way by those who are ‘open to the thing’: that
is, for Hume, those with doxastic virtue and an authoritative moral sense.
We have seen that, for both Hume and Heidegger, different sensibilities open
up or disclose different aspects of the world: it must not be assumed that the
reason of the understanding (in Hume’s sense) is competent to disclose all
aspects of reality. This raises the question: what is the relation between the
‘region’ of ethics and others?
Hume’s broad conception of ethics has the consequence that the worlds
of natural fitnesses in the practical domains (aesthetic, ethical, prudential)
should not be seen as sharply disjointed from each other. The moral and the
prudential come together with the virtues useful or agreeable to self, virtues
that inevitably have effects also on others. This is unsurprising since a condi-
tion of a moral sense proper to humans is benevolence constrained by self-
love. Included amongst such virtues are ‘prudence, temperance, frugality,
assiduity, enterprise, dexterity’ (T 587).
Most importantly what is the relation between the region of ethics and
that of science? I claimed previously that for Hume, emotional disclosure is
necessary but not sufficient for the disclosure (unconcealment) of ethics. It
is not sufficient, for as Hume makes clear, the cooperation of reason proper
to the understanding is necessary. Such cooperation is necessary since the
criteria of virtue cannot be ‘judiciously’ applied without it. Science pro-
vides background theories that help explain important moral practices and
enrich our understanding of a good life for a human being.25 So, though
moral truth depends on the possibility of a moral sense, and thereby on
its conditions of possibility (notably benevolence and benevolent empathy),
188 Christine Swanton
its operations do not guarantee the discernment of moral truth. Even the
warranted judgments of those with an authoritative moral sense may thus
fail to be true, for status as a virtue or a vice depends on ‘matters of fact’,
potentially discoverable by the understanding, which may not be picked up,
even by wise and sensitive judges. Hence, Hume’s response dependent view
of (virtue) ethics should be understood as a version of (I) (see section ‘Virtue
Ethics and Response Dependence’), but not (Q).
This fourth aspect of truth as aletheia brings into focus an essential aspect
of Heidegger’s metaphysics—a rejection of a private, spectator metaphysics
in favor of one constituted by an essentially social engagement. It may seem
that Hume belongs to the former camp, but I hope to show that this is debat-
able, certainly where his ethics is concerned.
We have seen that the fundamental idea of truth as aletheia is the intelligi-
bility of a thing that, for Heidegger, presupposes engagement with and com-
petence in a social world. The emotional background (attunement) essential
for the disclosure of ethics, that is practical competence in ethics (as Hume
also saw), thus has an essentially social dimension. Competent emotional
engagement presupposes that we understand the significance of emotions
within cultural and historical contexts: emotions are not to be seen as pri-
vate or inner. What counts as naturally good and what counts as a defect is
often a matter of interpretation.26 In a world of social meanings, the claim
that something is a defect might be taken to be an insult.
Hume recognized this feature of emotion in two ways:
a. Hume understood the public nature of emotion at a very basic level;
indeed he can be seen as a philosophical precursor of modern mirroring
theory. He claims in the Treatise that emotion is transmitted to another as
a ‘contagion’ or ‘infusion’ of sentiments. For example: ‘A chearful counte-
nance infuses a sensible complacency and serenity into my mind; as an angry
or sorrowful one throws a sudden damp upon me’ (T 317). ‘Empathy and
sympathy are thus capacities enabling us to ‘receive by communication [oth-
ers’] inclinations and sentiments’ (T 316).
Now we may think that Hume’s talk of ‘infusion’ is committed to an
asocial view of emotions of the sort criticized by Heidegger. For the latter,
although ‘[emotional] attunements are feelings’, such an attunement is not a
private emotional experience that is then transmitted to others, in the man-
ner of ‘infectious germs’.27 An attunement, including an emotional one, is a
way of being in the world and being with others. We say, for example, that
persons of ‘good humour bring a lively atmosphere with them’.28 But this
is exactly Hume’s view. The melancholy person of Hume’s description in a
similar way brings a melancholy atmosphere with him: he ‘throws a damp’
A New Metaphysics for Virtue Ethics 189
on a person, or on the conversation in general. Compare Heidegger: ‘Or
another person is in a group that in its manner of being dampens and
depresses everything; no one is outgoing . . . Moods are not accompany-
ing phenomena; rather, they are the sort of thing that determines being-
with-one-another in advance’.29 The four-fold of aletheia, which includes an
irreducible social dimension, thus applies to emotional disclosure as to any
other form of disclosure.
b. Hume understood that emotional responses are inculcated within
educational and social practices. This is clear in his discussion of justice.
Although the natural materials for emotional response must be within our
‘frame and constitution’, it is education and upbringing that turns these into
emotionally laden norms. Disgust, distaste, and hate are transmuted through
education and custom into sentiments of seeing as disgraceful and ignoble;
admiration into sentiments of seeing as noble and admirable, and so on.
CONCLUSION
This characteristic of virtue relates to the first feature of the first aspect of
the four-fold of aletheia—the thing being ‘out in the open’. This feature is:
things must have powers to affect sensibilities if they are to be disclosed.
(c) The moral sense most relevant to the disclosure of virtue is the author-
itative moral sense of qualified persons.
This characteristic of virtue is associated with the second aspect of the four-
fold, being ‘open to the thing’. Not all moral senses are equally ‘open to the
thing’. Notice, however, that (c) does not imply (Q) (section ‘Virtue Ethics
and Response Dependence’).
(d) Virtues are traits that tend to the good of mankind or are naturally
fitted to produce immediate pleasure from ‘species or appearance’.
(e) Virtues are traits that are approved from the common point of view
in social contexts.
This feature of virtue relates to the fourth aspect of the four-fold, the essen-
tially public nature of aletheia.
Applying the account of the four-fold of aletheia to Hume shows that for
him vice is not just a property like knives sticking out of backs, or a feel-
ing of disgust in the contemplator of it; nor is it a queer property. Virtues
and vices are powers but not queer powers. For talk of powers is simply
a way of indicating the reality of virtue and vice: there is no suggestion of
representational adequacy secured through scientific modes of understand-
ing. An agent cannot just see such a power, or simply reason causally about
it. Talk of virtue and vice is intelligible only in a background context of
emotional comportment in a world of engaged agency, conducted by beings
of a fundamentally benevolent empathetic disposition, steeped in contexts
of education, politics and policy, friendship, family, justice, respect within
(legitimate) social hierarchies, and so on. On these practices Hume has quite
a lot to say, and much work by Hume scholars such as Annette Baier and
Jackie Taylor has shown the practical, engaged, socially embedded nature of
Hume’s moral philosophy.
A New Metaphysics for Virtue Ethics 191
NOTES
Baier, Annette C., Death and Character: Further Reflections on Hume (Cambridge,
MA: Harvard University Press, 2008).
———, Progress of Sentiments (Cambridge, MA: Harvard University Press, 1991).
Cohon, Rachel, Hume’s Morality: Feeling and Fabrication (Oxford: Oxford University
Press, 2008).
———, ‘Reply to Radcliffe and Garrett’, Hume Studies 34 (2008), 277–288.
de Gaynesford, Maximilian, John McDowell (Cambridge: Polity Press, 2004).
Dreyfus, Hubert L., Being-in-the-World: A Commentary on Heidegger’s Being and
Time, Division I (Cambridge, MA: The MIT Press, 1991).
Dupre, John, The Disorder of Things (Cambridge, MA: Harvard University Press, 1993).
Hacker-Wright, John, ‘What is Natural about Foot’s Ethical Naturalism?’, Ratio 22
(2009), 308–321.
Heidegger, Martin, Basic Questions of Philosophy, trans. Richard Rojcewicz and
Andre Schuwer (Bloomington: Indiana University Press, 1994).
———, Being and Time, trans. John Macquarrie and Edward Robinson (New York:
Harper, 1962).
———, The Essence of Truth: On Plato’s Cave Allegory and Theaetetus, trans. Ted
Stadler (New York: Continuum, 2002).
———, The Fundamental Concepts of Metaphysics: World, Finitude, Solitude,
trans. William McNeill and Nicholas Walker (Bloomington: Indiana University
Press, 1995).
Hume, David, Enquiries Concerning the Principles of Morals, 3rd ed., ed. P. H. Nid-
ditch. (Oxford: Clarendon Press, 1995).
———, ‘The Standard of Taste’, in Essays ,Moral, Political, and Literary, vol. 1, ed.
T. H. Green and T. H. Grose (London: Longmans, Green, and Co., 1912).
———, A Treatise of Human Nature, ed. L. A. Selby-Bigge (Oxford: Clarendon Press,
1978).
Malpas, Jeff, ‘The Twofold Character of Truth: Heidegger, Davidson, Tugendhat’,
Divinatio (forthcoming).
McDowell, John, ‘Two Sorts of Naturalism’, in Virtues and Reasons, ed. Rosalind
Hursthouse, Gavin Lawrence and Warren Quinn (Oxford: Clarendon Press,
1995), 149–179.
———, ‘Values and Secondary Qualities’ in Morality and Objectivity, ed. T. Hond-
erich (London: Routledge and Kegan Paul, 1985).
Pigden, Charles R., ‘If Not Non-Cognitivism, Then What?’ in Hume on Motiva-
tion and Virtue, ed. Charles R. Pigden (Basingstoke: Palgrave McMillan, 2010),
80–104.
Railton, Peter, ‘Subjective and Objective’, in Truth in Ethics, ed. Brad Hooker (Oxford:
Blackwell, 1996), 51–68.
Roberts, Robert C., Emotions: An Essay in Moral Psychology (Cambridge: Cambridge
University Press, 2003).
Rorty, Richard, Philosophy and the Mirror of Nature (Princeton: Princeton University
Press, 1979).
Swanton, Christine, ‘Can Hume be Read as Virtue Ethicist?’, Hume Studies 33
(2007), 91–113.
———, ‘What Kind of Virtue Theorist is Hume?’, in Hume on Motivation and Vir-
tue, ed. Charles R. Pigden (Basingstoke: Palgrave McMillan, 2010), 226–248.
Taylor, J., ‘Hume on the Standard of Virtue’, Journal of Ethics 6 (2002), 43–62.
———, ‘Virtue and the Evaluation of Character’, in The Blackwell Guide to Hume’s
Treatise, ed. Saul Traiger (Oxford: Blackwell, 2006), 276–295.
194 Christine Swanton
Tugendhat, Ernst, ‘Heidegger’s Idea of Truth’, in Hermeneutics and Truth, ed. Brice
R. Wachterhauser (Evanston, IL: Northwestern University Press, 1994).
Wiggins, David, ‘A Sensible Subjectivism’, in Needs, Values, and Truth, ed. David
Wiggins (Oxford: Clarendon Press, 1987).
Williams, Bernard, ‘Evolution, Ethics, and the Representation Problem’, in Making
Sense of Humanity and Other Philosophical Papers 1982–1993, ed. Bernard Wil-
liams (Cambridge: Cambridge University Press, 1995).
Wrathall, Mark A., Heidegger and Unconcealment: Truth Language and History
(Cambridge: Cambridge University Press, 2011).
13 A Kantian Plea for Virtues?
Erasmus Mayr
II
III
There are different attempts to fill the gap in Kant’s account of moral right-
ness in the Groundwork, which differ with regard to the degree to which
they correspond—or claim to correspond—to Kant’s own (supposed) views
on how to apply the CI to particular cases. I’ll briefly look at some of these
attempts before turning to the question of how an appeal to virtues could
contribute to a solution.
(a) First, there is what one might call the ‘Groundwork/Critique inter-
nal’ strategy, which attempts to fill the gap by only using resources from
the Groundwork and the Critique of Practical Reason. The (unconscious)
adoption of some such a strategy may, partly, explain the popularity of the
third formula of the CI—the respect-for-humanity formula—in the debate
since this formula seems to have a material content that the two other, law-
formulas, lack. But, given that Kant thought the third formula to be equiva-
lent to the other two, it would be very surprising if applying this formula
A Kantian Plea for Virtues? 199
should, ipso facto, yield substantially stronger results. Furthermore, while
it appears plausible that we can reject some conceptualizations of action-
situations on the basis of the third formula, it seems implausible that this
should work generally, for all illicit conceptualizations. While, for example,
a discriminatory ‘racist’ conceptualization of an action-situation in which
a person’s race is per se considered as relevant for how he is to be treated
would fail to show the respect due to this person in virtue of his standing as
a moral person, not all cases of ignoring morally salient features are illegiti-
mate in precisely this way. For example, when the mistake in the conceptu-
alization isn’t that fundamental, but ‘only’ involves neglecting some salient
aspect of a person, such as her shyness, while otherwise acknowledging her
moral standing.
(b) Second, there is the ‘impure ethics’ strategy,14 according to which the
CI test alone cannot tell us what to do in particular action situations—but
neither did Kant ever intend it to. After all, the CI is only the expression of
the moral law for finite rational beings in general, while in the MM Kant
goes on to discuss species-specific duties for human beings as such, becom-
ing even more specific—to differences of gender, race, etc.—in the Anthro-
pology.15 And these latter specifications are not mere afterthoughts but had
already been alluded to by Kant in the Groundwork when he stressed that
ethics had an empirical part, ‘praktische Anthropologie’.16 So, it would
seem, the worry about applying the CI to particular cases rests on a misun-
derstanding of Kant’s own conception of the CI.
The impure ethics strategy can claim great exegetical plausibility, and it
is very probable that Kant himself would have subscribed to one version of
it. However, there remain grave doubts as to whether the ‘human-specific’
or ’empirical’ parts of Kant’s ethics can, by themselves, answer the constitu-
tive question of what makes certain aspects of a situation morally salient.
It is certainly true that these parts of Kant’s ethics will spell out some of the
rules of moral salience, but the problem is not merely that the ‘impure’ part
of Kant’s ethics has never been fully worked out by himself, but is, at best,
fragmentary, as even philosophers who have followed this track, admit.17
The more basic problem is that describing features that are considered as
salient for the application of moral laws in human life doesn’t by itself tell
us why these features are salient and whence their normative significance
for determining what is morally required in a particular situation derives.
This is particularly difficult to see with regard to Kant’s primarily descriptive
work, for example, in the Anthropology: while the factual information Kant
presents here can, plausibly, give rise to further hypothetical imperatives
once we know what our moral duties are, and can show us how agents natu-
rally conceive of themselves and their action-situations, it remains unclear
why this information should have a foundational role for ethics.18
Nor does an appeal to the Faculty of Judgment (‘Urteilskraft’), to which
Kant ascribes the task of applying abstract principles to particular cases—
also, explicitly, in the case of moral principles19—by itself fill the gap, or,
200 Erasmus Mayr
at least, not straightforwardly so. For the problem is not simply the gen-
eral problem of subsuming particular cases under abstract principles—this
general problem faces us even when we apply the CI to already formed
particular maxims. The problem is rather to determine—or to ‘construct’—
the particular objects to which the abstract principle is to be applied in
the first place. What we need is therefore something like a ‘schematization’
of general moral principles, which will have to fulfill a similar task to the
schematization of the categories in the Critique of Pure Reason.20 But while
Kant, in the latter case, takes great care to argue how this schematization is
supposed to work, in his ethics, he does not do so, nor does he tell us what
makes one particular way of schematizing the right one.
IV
These problems make it attractive to turn to virtues in order to fill the gap in
Kant’s account of moral decision-making. The aspect of virtues that makes
them appear particularly apt for playing this role is that having a specific
virtue is, at least on the traditional, Aristotelian, conception, typically con-
nected to a certain way of seeing situations in which this virtue is relevant.
The virtuous man ‘sees’ the situation he faces in the required way, which
allows him to emotionally and volitionally respond to it as it is appropri-
ate. For example, the courageous man sees the attacking enemy in battle
primarily as a danger to his own hometown, which must be averted, rather
than as a menace to his own life, which must be escaped. If possessing a
virtue is indeed connected to a special way of seeing action-situations and
conceptualizing them, a virtuous person will, as such, be able to pick out
those features of a situation that are morally salient and must be included
within the content of his maxim.
The insight that possession of a virtue has this cognitive component of
perception of morally salient features has been defended, at some length, by
John McDowell.21 In ‘Virtue and Reason’, McDowell argued that, in moral
action, there is a complex two-stage interaction between general knowledge
of how to live and particular knowledge about the action-situation in order
for the general knowledge to be applied. His characterization is based on the
model of the practical syllogism, and not of the CI, but I’ll quote it at some
length here: ‘It is at the first stage . . . that knowledge of how to live interacts
with particular knowledge: knowledge, namely, of all the particular facts
capable of engaging with concerns whose fulfillment would, on occasion, be
virtuous. This interaction yields, in a way essentially dependent on apprecia-
tion of the particular case, a view of the situation with one such fact, as it
were, in the foreground. Seen as salient, that fact serves, at the second stage,
as minor premiss in the core explanation’.22 So, in order for general knowl-
edge of how to live to be applied to the action-situation, first some features
of the situation are identified as salient in the light of this general knowledge,
A Kantian Plea for Virtues? 201
for example, ‘This food is healthy’. If one adds to this premise the general
principle that healthy food should be eaten, one can, in the second step,
construct a practical syllogism with the conclusion that this food should be
eaten. It’s crucial to notice, though, that the real work is not constructing
this practical syllogism, but identifying the relevant features of the situation
that might be used in such a syllogism. And, as McDowell rightly insists, this
work is not done by applying codified principles of moral relevance, but is a
procedure that is best—perhaps, only—describable in terms of what features
a virtuous man would pick out.
Some analogous process of singling out morally salient features, it seems,
must take place in the process of deciding what is morally right to do when
conceived on the Kantian model. But this turns out to be much trickier than
on the practical syllogism model used by McDowell. For, on the latter model,
the major premise of the syllogism contains some substantial, not purely for-
mal, ethical principle (like ‘healthy food should be eaten’), either constitut-
ing or being derived from a conception of how to live, whose content offers
(some) guidance in picking out the relevant features of the situation. This is
not the case with the CI, which is a purely formal principle. Furthermore,
remember, our problem is not merely an epistemological one, but a consti-
tutive one, about what determines which aspects are morally salient in this
situation. This cannot be the CI test itself—for, as we have seen, applying the
CI test already presupposes a maxim that picks out the aspects considered as
salient, which means that making the CI test itself generally determine the
morally salient aspects would lead to an infinite regress. So, do we have to
accept additional and independent normative facts about which aspects are
morally salient? Once we admit such facts, we seem to have given up on the
central claim of the Kantian view that the imperatives of morality are obliga-
tory for us qua rational beings simply in virtue of the structure of the will
we have. For once we start to accept facts about moral relevance that are
not themselves anchored in the structure of a rational will, it seems that we
can, just as well, go along all the way to accepting moral facts about what
to do that are thus independent. (And for McDowell himself, perception of
morally salient features does also involve a perception of what these features
demand of us.)23 But then, what role will be left for the CI test to determine
whether an action is right or wrong? The test will have become dispensable,
both in determining the rightness of actions and in explaining our grasp of
it, given that we have to accept independent moral facts about the rightness
and wrongness of actions, anyway, and that we must be aware of them in
order to apply the test.
So, while filling the gap between action-situation and applying the CI test
requires some account of which features are morally relevant and of what
makes them relevant, the Kantian account would be abandoned rather than
completed if the sensitivity to morally relevant features was so comprehen-
sive as to already include sensitivity to what was demanded of us. What
the Kantian account needs to be supplemented by is only something that
202 Erasmus Mayr
excludes cases of moral ‘aspect-blindness’, such as the failure to see that the
shyness of a person is a morally relevant factor for situations.24
VI
So here is, finally, the answer to our constitutive question about what makes
it the case that a certain feature of an action-situation is morally salient: It is
that a virtuous agent would take this feature to be morally salient, where the
figure of the ‘virtuous agent’ has been constructed by the recursive process
we have described.
It is important to note four things about this answer: (i) First, since we
have explained moral salience via the construction of the figure of the ‘vir-
tuous agent’, we have not been forced to introduce normative facts about
moral salience that are independent from the structure of a rational agent’s
will, which, for a Kantian account, is the sole genuine source of normativ-
ity. (ii) Second, at the same time, there is a genuine and irreducible role for
‘virtue’ and the ‘virtuous agent’ in this account of moral salience. On the one
hand, the notion of the ‘virtuous agent’ plays a genuine role in deciding what
the agent should do in a particular situation—by determining which features
of the situation are salient. On the other hand, since, during the recursive
process, we have gradually enlarged the set of character-traits and intellec-
tual capacities to be counted among the virtues, by exploiting the fact that
virtues are character-traits that do not only manifest themselves in behavior
of a certain kind, we cannot eliminate the notion of virtues from our account
of moral salience in favor of other notions, such as a disposition to act on
principles of right. So, the notion of virtue that we need in order to explain
A Kantian Plea for Virtues? 207
moral salience turns out to be tied to, but not reducible to, the notion of
moral rightness. (And, indeed, had it turned out to be so reducible, the notion
could not have helped us to explain moral salience at all—because, as we
have seen, the CI Test alone was insufficient to determine which features of
an action-situation are morally salient.) (iii) Third, during the procedure we
have gradually fed more and more human-specific facts into our account of a
good life and a virtuous agent, by exploiting knowledge about the conditions
of agency and the constitution of character-traits that is specific for humans.
For example, we have exploited the knowledge that courage does not only
manifest itself in morally relevant situations, which is, presumably, a specific
fact about courage in humans. Thus, we can escape the objection that the
notion of the ‘virtuous agent’ is too general and abstract to determine what
is morally salient for humans. (iv) While we have worked with one notion
of a virtuous agent, it is crucial to notice that we haven’t presupposed that
all virtuous agents must be the same—nor is this a result of the argument
presented here. On the contrary, virtuous agents can be different and can, to
a degree, consider different things as salient, for example, due to their diverg-
ing aims or due to different cultural settings. What we have been developing
is only an account of which features all virtuous human agents will consider
as salient—for these are the features that must be taken into account by an
agent in a particular action-situation when he is forming his maxim.
I therefore submit that appealing to virtues in the way described is indeed
an attractive way to close a crucial lacuna in Kantian moral theory because
it offers us a viable account of moral salience that is compatible with the key
Kantian contention that morality is obligatory to us qua rational beings in
virtue of the structure of our will. If this is correct, then, curiously enough,
appealing to irreducible virtues might well be a Kantian’s best bet to save the
‘purity’ of the moral law, while ensuring its applicability.
NOTES
I am grateful to Andree Hahmann, Franz Knappik, Julia Peters and Wilhelm Vossen-
kuhl for very helpful comments on earlier drafts of this paper.
1. Cf. Watson, ‘On the Primacy of Character’, 229.
2. Rawls, A Theory of Justice, 436.
3. MM Akademie-Ausgabe (AA) 6: 394.
4. Critique of Pure Reason (CPrR) AA 5: 19: ‘contain a general determination
of the will’.
5. The situation-type description covers both external circumstances and the
agent’s own self-conception at the time of his action. For different views on
the content of maxims see Nell, Acting on Principle, ch. 1.
6. Some of Kant’s formulations in the Critique of Practical Reason suggest that
only specification of action and end is needed; e.g., AA 6: 20: ‘prescribes
action as a means to an effect’. But Kant’s treatment of the examples in
Groundwork of the Metaphysics of Morals (GMM), AA 4: 421 ff., clearly
indicates that also some specification of the situation is required.
208 Erasmus Mayr
7. See O’Neill, Constructions of Reason. Explorations of Kant’s Practical Phi-
losophy, 87. O’Neill, however, believes that this difficulty can be resolved.
8. The same kind of problem arises for the required degree of specificity of
maxims, with regard to the description of the circumstances and the action.
9. Pace Herman, The Practice of Moral Judgment, ch. 7.
10. Unless he acts in pursuance of one of the obligatory ends discussed in the
Doctrine of Virtue of the MM.
11. Anscombe, ‘Modern Moral Philosophy’, 27.
12. Pace Esser, Eine Ethik für Endliche: Kants Tugendlehre in der Gegenwart, who
argues that we always start off our deliberation either from a particular perspec-
tive or from an overarching moral perspective, not from a neutral description
of the action-situation. ‘Moralisch relevante Erfordernisse der Situation erlangt
man in Beschreibungen der Situation, die bereits unter Voraussetzung dieses
umfassenden moralischen Anspruchs angestellt werden’ (271).
13. Herman, The Practice of Moral Judgment, 77 ff.
14. For this strategy in general see Louden, Kant’s Impure Ethics.
15. For an overview over the different degrees of ‘impurity’ in Kant’s ethics see
Louden, Kant’s Impure Ethics, 10 ff.
16. GMM AA 4: 387.
17. Louden, Kant’s Impure Ethics, 6 ff.
18. This kind of worry about the role of ‘moral anthropology’ is raised by Gregor,
Laws of Freedom, 8.
19. GMM AA 4: 389.
20. In MM 6: 468, Kant himself uses this comparison.
21. The locus classicus being McDowell, ‘Virtue and Reason’, 65 ff. Also Herman,
Moral Literacy.
22. McDowell, ‘Virtue and Reason’, 69.
23. Cf. McDowell, ‘Are Moral Requirements Hypothetical Imperatives’, 80 ff.
24. Example from McDowell, ‘Are Moral Requirements Hypothetical Impera-
tives’, 85.
25. See, e.g., MM AA 6: 377.
26. For an attempt to develop a notion of human flourishing along these lines see
Denis, ‘A Kantian Conception of Human Flourishing’.
27. CPrR AA 5: 110 ff.
28. GMM AA 4: 397.
29. Nor is it a life that necessarily exhibits more virtue, as Kant himself stresses:
The temptations that are overcome merely allow us to (subjectively) ‘calcu-
late’ the greatness of moral fortitude, but do not objectively determine this
greatness, MM AA 6: 397.
30. MM AA 6: 385 ff.
31. MM AA 6: 399 ff.
32. GMM AA 4: 393. Also MM AA 6: 408. As Kant makes clear in the latter pas-
sage, there is a corresponding moral duty to cultivate self-control, but pursu-
ing this duty is not the same as already possessing self-control.
33. See, e.g., MM AA 6: 388.
34. CPrR AA 5: 155 f.
35. Plausibly, there will be such further characteristics, and our procedure will
not stop with step (ii). For example, leading a life that combines both cour-
age and prudence will require intellectual capacities for assessing danger and
possible gain that were not yet included in the set spelled out in step (i).
36. The ideal figure of the ‘virtuous agent’ has a parallel in the figure of Jesus in
the Religionsschrift, who incorporates, for Kant, the ideal of moral perfection
such as it is possible for human beings, AA 6: 61. (I owe this point to Andree
Hahmann.)
A Kantian Plea for Virtues? 209
REFERENCES
Page references to Immanuel Kant’s works are according to the pagination of the
Akademieausgabe of Kant’s Gesammelte Schriften. English translations are from
Gregor, Mary (transl. end ed.), and Allen Wood (general introd.), The Cambridge
Edition of the Works of Immanuel Kant: Practical Philosophy (Cambridge: Cam-
bridge University Press 1996).
Anscombe, Elizabeth, ‘Modern Moral Philosophy’, in Virtue Ethics, ed. R. Crisp and
M. Slote (Oxford: Oxford University Press, 1997), 26 ff.
Denis, Lara, ‘A Kantian Conception of Human Flourishing’, in Perfecting Virtue.
New Essays on Kantian Ethics and Virtue Ethics, ed. L. Jost and J. Wuerth (Cam-
bridge: Cambridge University Press, 2011), 164 ff.
Esser, Andrea, Eine Ethik für Endliche: Kants Tugendlehre in der Gegenwart (Stutt-
gart: Frommann-Holzboog, 2004).
Gregor, Mary, Laws of Freedom (Oxford: Basil Blackwell, 1963).
Herman, Barbara, Moral Literacy (Cambridge, MA: Harvard University Press, 2007).
———, The Practice of Moral Judgment (Cambridge, MA: Harvard University
Press, 1993).
Louden, Robert, Kant’s Impure Ethics. From Rational Beings to Human Beings
(New York: Oxford University Press, 2000).
McDowell, John, ‘Are Moral Requirements Hypothetical Imperatives’, in Mind,
Value, and Reality, ed. J. McDowell (Cambridge, MA: Harvard University Press,
1998), 77ff.
———, ‘Virtue and Reason’, in Mind, Value, and Reality, ed. J. McDowell (Cam-
bridge, MA: Harvard University Press, 1998), 50 ff.
Nell, Onora, Acting on Principle (New York: Columbia University Press, 1975).
O’Neill, Onora, Constructions of Reason. Explorations of Kant’s Practical Philoso-
phy, (Cambridge: Cambridge University Press, 1989).
Rawls, John, A Theory of Justice (Oxford: Oxford University Press, 1971).
Watson, Gary, ‘On the Primacy of Character’, in Virtue Ethics, ed. S. Darwall
(Oxford: Blackwell, 2003), 229 ff.
14 Toward a Humean Virtue Ethics
Lorenzo Greco
More and more scholars, for various and often contrasting reasons, have
recently put Hume’s moral philosophy under the heading ‘virtue ethics’. Also,
many contemporary philosophers are trying to elaborate a specific form of
Humean virtue ethics to be contrasted with the more famous neo-Aristotelian
alternatives. Hence, as occurred with the renaissance of Aristotelian virtue
ethics, it appears that there is space to develop a full-fledged Humean ver-
sion of it as well. My scope here, however, is more limited. After having
presented the main reasons in favor of a classification of Hume among virtue
ethicists, what I would like to do is to take into account some recent attempts
at presenting a virtue ethical interpretation of Hume, with the aim of shed-
ding some light on the theoretical direction I believe a project of a systematic
Humean virtue ethics should take. I shall proceed by addressing some specific
issues raised by the favorable reading of Hume provided by Christine Swan-
ton1 and by the criticism moved against Hume by Rosalind Hursthouse.2
By doing that I’ll argue that Hume offers the philosophical tools to redefine
some basic notions of virtue ethics in a more efficacious way compared to
the opposing neo-Aristotelian model and that the strength of Hume’s version
of virtue ethics is that he aims at the unity of character instead of the unity
of the virtues. This makes it possible to develop a pluralistic and secularized
morality that denies any supposed final cause or télos for human beings con-
ceived as a species and instead upholds the individuality of the person as the
fundamental value that should be pursued and promoted.
II
To begin with, is Hume’s ethics a form of virtue ethics in all ways? What
cannot be denied is that Hume himself, in his examination of morality, rec-
ognizes a crucial role to the notions of virtue and vice (EPM 1.10; SBN
173–174).3 Hume’s intent is to give a list of virtues and vices in accordance
with the way human nature develops within particular contexts.4 Moreover,
Toward a Humean Virtue Ethics 211
Hume tells us that the objects of moral judgments are not people’s actions,
but the motives that lie behind them; human actions may well be regarded in
a positive or in a negative light, but only insofar as the motives that activated
them are valued positively or negatively (T 3.2.1.2; SBN 477).5 In turn, these
motives have to be related to the characters of people, and people are mor-
ally evaluated because they display characters of certain kinds (T 3.3.1.4–5;
SBN 575 and T 3.3.1.19; SBN 584).
This progress from actions to motives, and from motives to characters
that make persons virtuous or vicious agents, brings Hume’s conception
of morality very close to a virtue ethical model. Moreover, since agents are
morally evaluated because of their characters, the way these characters are
formed becomes an issue of the greatest importance for Hume (T 3.2.2.26;
SBN 500–501).6 Hume appears to be concerned with that ‘ethical forma-
tion’7 that again occupies so much space in many virtue ethics discussions.
It is important, however, to stress the original way in which Hume explains
how characters develop, an explanation that is in line with his sentimental-
ism. Hume says that, by ‘custom and education’ (T 3.2.2.26; SBN 500) peo-
ple can build up ‘calm’ passions, whereby it is possible to lead lives guided by
stable principles of action. Often confused with reason because of their lack-
ing of emotive violence, calm passions are in fact for Hume strong passions
that organize one’s existence according to goals that in the end become firm
and coherent. Thanks to calm passions, agents acquire ‘strength of mind’
by which they are able to persist in the realization of their projects, without
being tempted by false ends—maybe more appealing in the short period, but
in fact pernicious to their lives considered in their totality (T 2.3.3.8 and 10;
SBN 417–418).8 Only those who are properly educated and have adopted
the correct habits will curb their passions and fortify those characters that
will make them virtuous agents. But it is worth repeating that this moral
learning, for Hume, works purely and solely at a sentimental level. Virtuous
agents are those who come to be moved by calm passions, which correspond
to traits of character regarded as virtues.
This marks a difference between the Humean conception of ethics
and other virtue ethical approaches—in particular some kinds of neo-
Aristotelianism9—according to which being properly educated means being
able to respond correctly to the moral features of a given situation. Accord-
ing to this neo-Aristotelian model, virtue should foremost be considered as
a form of knowledge, and the virtuous person as someone who first of all
gets things right and then acts accordingly. The phronimos is gifted with
a perceptual capacity, usually explained in intellectual terms as a form of
moral wisdom, by which he or she becomes sensible to the suitable require-
ments that the situation imposes on behavior. Conversely, Hume makes no
reference to any intellectual faculty of any kind when he has to explain
how a person becomes a virtuous agent; the Humean virtuous person does
not act on the strength of such a faculty as ‘either desire-related intellect or
thought-related desire’,10 which guarantees at the same time the right look
212 Lorenzo Greco
on things and the motivational force to move consequently. Besides, for
Hume ‘morality [. . .] consists not in any matter of fact, which can be dis-
covered by the understanding’ (T 3.1.1.26; SBN 468). Rather, values seem
to work rather like secondary qualities (T 3.1.1.26; SBN 469). Whether the
secondary quality comparison is the best way to explain Hume’s conception
of the nature of values is still a much debated question, and I will not address
it here. However, what can be observed is that, though for Hume the dimen-
sion of values is presented as a sort of projection onto the world due to the
sentimental framework of human nature, this dimension does not require
anything beyond this very sentimental framework to be stated. Taste, Hume
affirms, moral and aesthetic, ‘has a productive faculty, and gilding or stain-
ing all natural objects with the colours, borrowed from internal sentiment,
raises in a manner a new creation’ (EPM appendix 1.21; SBN 294). ‘To
have the sense of virtue, is nothing but to feel a satisfaction of a particular
kind from the contemplation of a character. The very feeling constitutes our
praise or admiration. We go no farther; nor do we enquire into the cause
of the satisfaction’ (T 3.1.2.3; SBN 471). This is because ‘there is just so
much vice or virtue in any character, as every one places in it, and [ . . . ] ’tis
impossible in this particular we can ever be mistaken’ (T 3.2.8.8; SBN 547;
see also ‘The Sceptic’, 168).
These passages in Hume’s texts seem to justify the conclusion that for
Hume the evaluative dimension is a sentimental representation—not an
intellectual one—that human beings cast on things as a result of the activity
of their passions—not a state of affairs that is perceived, and with which
the virtuous person becomes attuned. True, he says that in morality ‘reason
and sentiment concur in almost all moral determinations and conclusions’
(EPM 1.9; SBN 172); but this ‘reason’ Hume refers to here is to be translated
in terms of calm passions (T 3.3.1.18; SBN 583). In a sense human beings
can sometimes be said to be ‘morally blind’ to the relevant ethical features
of situations.11 If what has been said so far is correct though, reference to
moral blindness (and, conversely, to moral vision) is to be taken figuratively.
Human beings may be morally blind for Hume because they are primarily
morally insensible, that is, because they are endowed with a poor sentimen-
tal equipment, incapable of being affected by sympathetic exchanges among
people. There is not really anything to be seen out there; ‘seeing’ makes sense
if taken as a metaphor for ‘feeling’ in a proper way, given a human nature
described in sentimental terms, which presents itself as the benchmark for
stating what the virtues and vices are.
III
This interpretation of the way Hume conceives the sphere of value, and the
role played by sentiment in it, does not go without criticism. Sentimentalism
may be defined very broadly as ‘the thesis that evaluation is to be understood
Toward a Humean Virtue Ethics 213
by way of emotional response’,12 and some scholars have recently argued that
Hume presents a kind of sentimentalist virtue ethics very close in its results
to the neo-Aristotelian one that has been presented as non-Humean so far.
For example, Christine Swanton considers Hume ‘as being part of both the
sentimentalist and the virtue ethical traditions’13 in maintaining a response-
dependent theory whereby ‘a virtue or a vice is a power in an object to elicit
relevant responses in qualified actors’.14 In turn, a qualified actor is someone
in possession of certain emotional dispositions that make him or her sensible
and reactive to the powers in the object, which are the virtues. In this sense,
in Swanton’s interpretation of Hume, ‘morality is not a matter of fact about
our sentiments, it is a matter of fact about virtue and vice, which are in
objects’;15 ‘virtues are response-dependent properties, and are therefore not
projections as some commentators claim’.16 By appropriately exercising their
moral sense, human beings can thus track the moral truths that allegedly
compose the ethical reality.17 A partly similar conclusion has been recently
given also by Michael Slote in his sentimentalist ethics of care.18
This response-dependent reading of Hume is a fascinating way of assess-
ing his ethical sentimentalism in the light of virtue ethics, but doubts can
legitimately be raised both about whether it corresponds to Hume’s own
intentions and also, more generally, about whether this is the correct way to
frame a Humean virtue ethics.19 It is indicative, for example, that both Swan-
ton and Slote make reference to the work of David Wiggins. Wiggins presents
a ‘sensible subjectivism’, according to which moral properties and appropri-
ate human sentiments are mutually correlated in <property, response> asso-
ciations, so that ‘x is good/right/beautiful if and only if x is such as to make
a certain sentiment of approbation appropriate’.20 By appealing to nothing
more fundamental than human sentiments, Wiggins aims at giving a cogni-
tivist account of the sphere of morality in which the claim to objectivity that
appears to be deeply rooted in the very concept of morality finds its proper
vindication.21 In developing his sensibilist model, Wiggins mentions Hume as
one of the authors with whom he has a close affinity. But whereas Swanton
says that Hume’s virtue ethics corresponds to a response-dependent theory
matching Wiggins’ sensibilism, Wiggins, on his part, admits instead that his
sensible subjectivism diverges from Hume’s ‘official theory’.22 Wiggins says
that we can (and indeed we should) progressively move from ‘[The real]
David Hume’ (who roughly corresponds to the projectivist description given
above) to ‘[A possible] David Hume: x is good if and only if x is such as to
arouse approbation’, and eventually end up with a ‘Refined Humean subjec-
tivism: x is good if and only if x is such as to deserve (N.B.) or merit approba-
tion’.23 But this is not what the real Hume does. So why should Hume (and
those of us who want to develop a Humean virtue ethics) make this move?
A revealing answer is given by Swanton herself:
For Hume,
So human beings recognize as virtues those character traits that are useful
to their possessors or to others, or immediately agreeable to their possessors
or to others. Vices are the opposite. In turn, thanks to sympathy, which is
considered by Hume as the principle of sentimental communication among
human beings, we can approve those traits of character that produce plea-
sure or advantage for other people or for their possessors themselves and
disapprove those traits of character that give pain or prove to be disadvanta-
geous for other people or their possessors themselves. Specifically, we have
a properly moral approval (or disapproval) when these sympathetic judg-
ments on traits of character are given from what Hume calls a ‘steady and
general’, or ‘common’, point of view (T 3.3.1.15–16 and 30; SBN 581–582
and 591. EPM 9.6; SBN 272), from which it is possible to determine a stable
and as much as possible impartial perspective on virtues and vices.
Now, like Swanton and Slote, Rosalind Hursthouse, too, takes into con-
sideration Hume as a possible representative of virtue ethics, but she discards
his moral theory as defective at the very root. In particular, she criticizes
Hume’s four sources of pleasure and pain as a correct standard for defin-
ing which character traits should be appreciated and which not, since these
four sources would correspond to a disjunctive claim, whose upshot is the
impossibility of defending a single measure of virtue and vice. Justice and
injustice, courage and cowardice, generosity and meanness would all turn
out to be virtues.30 Moreover, the steady and general point of view cannot
be a correct standard for moral judgments because it would be defined by
Hume as ‘uninfluenced by distances in time: it can respond to the virtues
of the ancient Greeks as competently as it can respond to those of its pos-
sessor’s contemporaries’.31 This would make the Humean steady and general
point of view too abstract and distant from those who must endorse it for
it to become a reliable standard in ethics. According to Hursthouse, to save
Hume’s theory from collapsing, it has to be, so to speak, ‘Aristotelized;’ the
steady and general point of view should be discarded as a reliable ethical
yardstick and replaced with the good ‘critic’ in morals as it is expressed by
Hume in his essay Of the Standard of Taste.32 Such a good critic is interpreted
by Hursthouse as the well-trained person, who is immersed in a particular
216 Lorenzo Greco
reality of which he or she is able to recognize the relevant moral aspects, thus
representing the closest approximation to the phronimos we can arrive at
within a Humean framework.
Yet what should be noted is that Hursthouse moves her objections against
Hume while taking for granted from the very beginning the Aristotelian per-
fectionist conception of human nature she endorses as normatively sound.
For Hursthouse, in fact, ‘the standard neo-Aristotelian completion claims
that a virtue is a character trait a human being needs for eudaimonia, to
flourish or live well’.33 And she defines what it means to ‘live well’ by mak-
ing reference to those distinctive functions characteristic of human beings
whose fulfillment allow human beings to live in the right way, as they are
required qua human beings, and thus to obtain the real happiness, or the
sort of happiness worth having.34 By presupposing such a unit of measure-
ment—human nature as she conceives it—Hursthouse can present a notion
of the phronimos that corresponds to somebody who shows practical wis-
dom, gathering coherently in himself or herself all the virtues at once, hence
embodying in himself or herself the criterion for objectivity that has been
looked for so far.
However, Hume has never professed the need to single out a criterion of
good and right that has to be valid in advance and that guarantees some-
thing like the unity of the virtues. Nor does the Humean steady and general
point of view correspond, as Hursthouse seems to believe, to a timeless
‘point of view of the universe’, or a ‘view from nowhere’. It is, rather, a point
of view that develops within human history as the result of people’s sympa-
thetic exchanges, that is, of a moral sentiment where ‘is displayed the force
of many sympathies’ (EPM 9.11; SBN 276). It is a reflective stance resulting
from that moral conversation human beings entertain because of their senti-
mental constitution that assures a convergence in moral judgments, but does
not provide that single, definitive measure of objectivity neo-Aristotelians
are looking for. On the contrary, Hume’s steady and general point of view
evolves through time and space, leaning on a fixed human nature whose
constancy is nothing but the product of a generalization (EHU 8.7; SBN
83–84).35 Hume’s way of establishing what constitutes human flourishing
is always an a posteriori operation, the consequence of empirical ascertain-
ment. Which character traits happen to be agreeable or useful to their pos-
sessors or to others can be derived from ‘a cautious observation of human
life’, and the list of virtues we will come up with is the outcome of ‘experi-
ments [. . .] as they appear in the common course of the world, by men’s
behaviour in company, in affairs, and in their pleasures’ (T Intro. 10; SBN
xix). His quadripartite standard works contingently in the course of human
history by considering how human nature expresses itself in the multiplicity
of situations in which people find themselves. So Hume is far from presup-
posing a finalistic notion of human nature and then stating which character
traits fully realize human nature’s peculiar ends. In turn, the Humean good
critic is precisely someone who puts himself or herself, and reflects, from
Toward a Humean Virtue Ethics 217
the perspective of the steady and general point of view. That is to say, the
good critic can be seen as the criterion for judging what is virtuous or vicious
only insofar as he or she has embraced that very point of view. His or her
practical wisdom does not reflect any phronesis whatsoever, but is the con-
sequence of having adopted that contingent position that is the product of
the continuous corrections human beings progressively bring to their moral
assessments, thanks to their uninterrupted passional exchanges, and to their
imaginative efforts to get in touch with their fellow men. In this sense the
Humean good critic is a human being like anyone else, but one who has
educated himself or herself to be more sentimentally attentive, less prone
to prejudice, more willing to engage in specifically moral arguments, and
thus to recognize moral distinctions established from the steady and general
point of view as sound and to be pursued.
VI
The picture of the virtuous person as the proud person allows Hume to
present his own peculiar notion of human excellence—a notion that com-
petes with the one belonging to the classical, that is, ancient Greek tradition
of virtue ethics.39 This Humean conception of human excellence takes the
form of ‘greatness of mind’—which for Hume is nothing but a steady and
well-established pride and self esteem—which displays traits of character
such as courage, ambition, love of glory, magnanimity, explicitly presented
by Hume as closely related to the classical world, and in opposition to the
distorted values of the Christian tradition (T 3.3.2.13; SBN 599–600). Now,
greatness of mind may well reveal itself also in the form of heroism and
military glory. And even though ‘men of cool reflexion are not so sanguine
in their praise of it’, because of the great damages it may cause to society,
when we fix our view on the person himself, who is the author of all this
mischief, there is something so dazling in his character, the mere con-
templation of it so elevates the mind, that we cannot refuse it our admi-
ration. The pain, which we receive from its tendency to the prejudice of
society, is over-power’d by a stronger and more immediate sympathy.
(T 3.3.2.15; SBN 601)
Here Hume touches a point that has been acknowledged and accepted by
present-day virtue ethicists such as Slote and Swanton, namely, the idea that
there may exist an ‘admirable immorality’40 and that we frequently esteem
virtues that do not bring any benefit to humankind.41 In doing that, Hume
develops a virtue ethics that could be defined as ‘pluralistic’, to use Swan-
ton’s expression.42 However, Hume’s ethics can be said to be pluralistic in
a different way from Swanton’s. She conceives her pluralistic virtue ethics
along with a response-dependent line, and the interpretation of Hume’s
Toward a Humean Virtue Ethics 219
ethics presented so far goes in another direction. Nonetheless, describing
Hume’s virtue ethics in pluralistic terms makes sense if we take Hume as
having as his core moral interest not so much an objective criterion to dis-
tinguish virtues and vices, but rather the individuality of persons. Individu-
ality stands out as a value that should be pursued and promoted precisely
because Hume has a pluralistic conception concerning virtues and vices,
which does not look for the unity of the virtues, but instead for the unity of
character.43 Hume does not have a problem of consistency among the vir-
tues; consistency becomes a problem only if we decide to embrace an ‘abso-
lute’ conception such as the neo-Aristotelian one. Rather, from Hume’s a
posteriori perspective, we may well admit the existence of people whose
characters are mixtures of a plurality of traits,44 some of which are virtues
when seen from the steady and general point of view, while others turn out
to be vices.45 What counts is character in its totality, as reflecting the indi-
viduality of a given person, not the determination of an objective perspec-
tive from which to label virtues and vices—a perspective that, Humeanly,
runs the risk of being nothing but a philosophical chimera. In a sense, this
allows Humeans to regain that notion of an end of human beings that the
neo-Aristotelians are so fond of. But in a Humean perspective this notion—
as with all the other fundamental notions of the virtue ethics vocabulary—
acquires a new meaning. It ceases to stand for a télos of humanity taken
as a species, but instead is always used in the plural form, to refer to the
most different ends individuals pursue. This is not to be understood as an
approximation of the ideal of the phronimos, but instead as the realization
of a unified character in the light of the steady and general point of view.
Finally, what should also be emphasized is that Hume mentions as an
integral part of human excellence the virtue of benevolence (T 3.3.3; EPM 2).
This is one of the differences between the Humean conception of a virtuous
life and the classical one. What is peculiar to the alternative offered by Hume
is that a life can be virtuous only if it is open to others, considered as differ-
ent persons who deserve our respect. Such moral relevance of benevolence
has nothing to do with Christian piety, but again is explained by Hume with
reference to the sentimental mechanisms of human psychology. Greatness
of mind and benevolence weigh each other out and are virtues insofar as
they reveal the social nature of human beings, defining the virtuous person
as someone who stands as a morally laudable individual because of his or
her connections with other people (T 3.3.3.9; SBN 606). So it turns out that
even though greatness of mind is indeed a virtue for Hume, it may not, in his
own terms, be appropriately ascribed to common people. Greatness of mind
suits soldiers or noblemen well; it represents an aristocratic way of being
virtuous that is certainly accepted by Hume but that he does not consider to
be the only or the best way of behaving virtuously. The peculiarity of Hume’s
conception of the virtuous person is that it appears to be, as it were, ‘democ-
ratized;’ his virtue ethics is not addressed to heroes, even less to saints, but
to people as they are commonly found in the world. That is, we do not need
220 Lorenzo Greco
to be heroes nor saints to be justly proud of ourselves since the steady and
general point of view is set on that ‘middle station of life’ that, according to
Hume, qualifies the condition of the greatest part of human beings, ‘afford-
ing the fullest security for virtue’, and giving opportunity ‘for the most ample
exercise of it’.46 Hence according to Hume, virtue emerges as a process of
continuous self-improvement in which people develop stable characters they
can be proud of, thus conceiving themselves as unitary individuals, without
having to presuppose an end-state of ideal or absolute perfection. In the end,
it may well happen that, when regarded a posteriori, a certain virtuous per-
son turns out to possess all the virtues. If so, this cannot be but a contingent
result. But, from a Humean perspective, this is more than enough.47
NOTES
REFERENCES
Christoph Halbig has held the Chair of Practical Philosophy at the University
of Gießen since 2011; before that, he held the Chair of Practical Philosophy
at the University of Jena from 2006–2010. In 2007 he received the ‘Award
for outstanding research in the social sciences and the humanities’ from the
Berlin-Brandenburg (former Prussian) Academy of Sciences and Humani-
ties. His books include Objektives Denken. Erkenntnistheorie und Philoso-
phy of Mind in Hegels System (Frommann-Holzboog, 2002) and Praktische
Gründe und die Realität der Moral (Vittorio Klostermann, 2007).
226 List of Contributors
Edward Harcourt is Fellow & Tutor in Philosophy at Keble College, Oxford.
His research interests include ethics (in particular moral psychology, neo-
Aristotelianism and child development, ethical dimensions of psycho-
analysis, meta-ethics, and Nietzsche’s ethics), literature and philosophy,
and Wittgenstein, areas in which he has published a number of articles in
journals and collections.
Candace Vogler is the David B. and Clara E. Stern Professor of Philosophy and
Professor in the College at the University of Chicago. Her research interests
are in practical philosophy, practical reason, Kant’s ethics, Marx, and neo-
Aristotelian naturalism. She is the author of Reasonably Vicious (Harvard
University Press, 2002).
Index