THE JOURNAL OF PHILOSOPHY
VOLUME LXXIX, NO. 8, AUGUST 1982
.~ ~ ~ ~ ~ ~ . ..+I
MORAL SAINTS*
DON'T know whether there are any moral saints. But if there
are, I am glad that neither I nor those about whom I care most
are among them. By moral saint I mean a person whose every
action is as morally good as possible, a person, that is, who is as
morally worthy as can be. Though I shall in a moment acknowledge
the variety of types of person that might be thought to satisfy
this description, it seems to me that none of these types serve as unequivocally
compelling personal ideals. In other words, I believe
that moral perfection, in the sense of moral saintliness, does not
constitute a model of personal well-being toward which it would
be particularly rational or good or desirable for a human being to
strive.
Outside the context of moral discussion, this will strike many as
an obvious point. But, within that context, the point, if it be
granted, will be granted with some discomfort. For within that
context it is generally assumed that one ought to be as morally
good as possible and that what limits there are to morality's hold
on us are set by features of human nature of which we ought not to
be proud. If, as I believe, the ideals that are derivable from common
sense and philosophically popular moral theories do not support
these assumptions, then something has to change. Either we must
change our moral theories in ways that will make them yield more
palatable ideals, or, as I shall argue, we must change our conception
of what is involved in affirming a moral theory.
In this paper, I wish to examine the notion of a moral saint,
first, to understand what a moral saint would be like and why such
a being would be unattractive, and, second, to raise some questions
about the significance of this paradoxical figure for moral philo*I
have benefited from the comments of many people who have heard or read an
earlier draft of this paper. I wish particularly to thank Douglas MacLean, Robert
Nozick, Martha Nussbaum, and the Society for Ethics and Legal Philosophy.
0022-362X/82/7908/0419$02.lI0 ? 1982 The Journal of Philosophy, Inc.
419
This content downloaded from
147.251.6.77 on Sat, 03 Sep 2022 08:11:53 UTC
All use subject to https://about.jstor.org/terms
420 THE JOURNAL OF PHILOSOPHY
sophy. I shall look first at the model(s) of moral sainthood that
might be extrapolated from the morality or moralities of common
sense. Then I shall consider what relations these have to conclusions
that can be drawn from utilitarian and Kantian moral theories.
Finally, I shall speculate on the implications of these considerations
for moral philosophy.
MORAL SAINTS AND COMMON SENSE
Consider first what, pretheoretically, would count for us-contemporary
members of Western culture-as a moral saint. A necessary
condition of moral sainthood would be that one's life be dominated
by a commitment to improving the welfare of others or of society
as a whole. As to what role this commitment must play in the
individual's motivational system, two contrasting accounts suggest
themselves to me which might equally be thought to qualify a person
for moral sainthood.
First, a moral saint might be someone whose concern for others
plays the role that is played in most of our lives by more selfish, or,
at any rate, less morally worthy concerns. For the moral saint, the
promotion of the welfare of others might play the role that is
played for most of us by the enjoyment of material comforts, the
opportunity to engage in the intellectual and physical activities of
our choice, and the love, respect, and companionship of people
whom we love, respect, and enjoy. The happiness of the moral
saint, then, would truly lie in the happiness of others, and so he
would devote himself to others gladly, and with a whole and open
heart.
On the other hand, a moral saint might be someone for whom
the basic ingredients of happiness are not unlike those of most of
the rest of us. What makes him a moral saint is rather that he pays
little or no attention to his own happiness in light of the overriding
importance he gives to the wider concerns of morality. In other
words, this person sacrifices his own interests to the interests of
others, and feels the sacrifice as such.
Roughly, these two models may be distinguished according to
whether one thinks of the moral saint as being a saint out of love
or one thinks of the moral saint as being a saint out of duty (or
some other intellectual appreciation and recognition of moral
principles). We may refer to the first model as the model of the
Loving Saint; to the second, as the model of the Rational Saint.
The two models differ considerably with respect to the qualities
of the motives of the individuals who conform to them. But this
difference would have limited effect on the saints' respective public
This content downloaded from
147.251.6.77 on Sat, 03 Sep 2022 08:11:53 UTC
All use subject to https://about.jstor.org/terms
MORAL SAINTS 421
personalities. The shared content of what these individuals are
motivated to be-namely, as morally good as possible-would play
the dominant role in the determination of their characters. Of
course, just as a variety of large-scale projects, from tending the
sick to political campaigning, may be equally and maximally morally
worthy, so a variety of characters are compatible with the ideal
of moral sainthood. One moral saint may be more or less jovial,
more or less garrulous, more or less athletic than another. But,
above all, a moral saint must have and cultivate those qualities
which are apt to allow him to treat others as justly and kindly as
possible. He will have the standard moral virtues to a nonstandard
degree. He will be patient, considerate, even-tempered, hospitable,
charitable in thought as well as in deed. He will be very reluctant
to make negative judgments of other people. He will be careful not
to favor some people over others on the basis of properties they
could not help but have.
Perhaps what I have already said is enough to make some people
begin to regard the absence of moral saints in their lives as a blessing.
For there comes a point in the listing of virtues that a moral
saint is likely to have where one might naturally begin to wonder
whether the moral saint isn't, after all, too good-if not too good
for his own good, at least too good for his own well-being. For the
moral virtues, given that they are, by hypothesis, all present in the
same individual, and to an extreme degree, are apt to crowd out the
nonmoral virtues, as well as many of the interests and personal
characteristics that we generally think contribute to a healthy, wellrounded,
richly developed character.
In other words, if the moral saint is devoting all his time to feeding
the hungry or healing the sick or raising money for Oxfam,
then necessarily he is not reading Victorian novels, playing the
oboe, or improving his backhand. Although no one of the interests
or tastes in the category containing these latter activities could be
claimed to be a necessary element in a life well lived, a life in
which none of these possible aspects of character are developed
may seem to be a life strangely barren.
The reasons why a moral saint cannot, in general, encourage the
discovery and development of significant nonmoral interests and
skills are not logical but practical reasons. There are, in addition, a
class of nonmoral characteristics that a moral saint cannot encourage
in himself for reasons that are not just practical. There is a
more substantial tension between having any of these qualities unashamedly
and being a moral saint. These qualities might be deThis
content downloaded from
147.251.6.77 on Sat, 03 Sep 2022 08:11:53 UTC
All use subject to https://about.jstor.org/terms
422 THE JOURNAL OF PHILOSOPHY
scribed as going against the moral grain. For example, a cynical or
sarcastic wit, or a sense of humor that appreciates this kind of wit
in others, requires that one take an attitude of resignation and pessimism
toward the flaws and vices to be found in the world. A
moral saint, on the other hand, has reason to take an attitude in
opposition to this-he should try to look for the best in people,
give them the benefit of the doubt as long as possible, try to improve
regrettable situations as long as there is any hope of success.
This suggests that, although a moral saint might well enjoy a good
episode of Father Knows Best, he may not in good conscience be
able to laugh at a Marx Brothers movie or enjoy a play by George
Bernard Shaw.
An interest in something like gourmet cooking will be, for different
reasons, difficult for a moral saint to rest easy with. For it
seems to me that no plausible argument can justify the use of
human resources involved in producing a pate de canard en croute
against possible alternative beneficent ends to which these resources
might be put. If there is a justification for the institution of haute
cuisine, it is one which rests on the decision not to justify every activity
against morally beneficial alternatives, and this is a decision
a moral saint will never make. Presumably, an interest in high fashion
or interior design will fare much the same, as will, very possibly,
a cultivation of the finer arts as well.
A moral saint will have to be very, very nice. It is important that
he not be offensive. The worry is that, as a result, he will have to be
dull-witted or humorless or bland.
This worry is confirmed when we consider what sorts of characters,
taken and refined both from life and from fiction, typically
form our ideals. One would hope they would be figures who are
morally good-and by this I mean more than just not morally
bad-but one would hope, too, that they are not just morally good,
but talented or accomplished or attractive in nonmoral ways as
well. We may make ideals out of athletes, scholars, artists-more
frivolously, out of cowboys, private eyes, and rock stars. We may
strive for Katharine Hepburn's grace, Paul Newman's "cool"; we
are attracted to the high-spirited passionate nature of Natasha
Rostov; we admire the keen perceptiveness of Lambert Strether.
Though there is certainly nothing immoral about the ideal characters
or traits I have in mind, they cannot be superimposed upon the
ideal of a moral saint. For although it is a part of many of these
ideals that the characters set high, and not merely acceptable,
moral standards for themselves, it is also essential to their power
This content downloaded from
147.251.6.77 on Sat, 03 Sep 2022 08:11:53 UTC
All use subject to https://about.jstor.org/terms
MORAL SAINTS 423
and attractiveness that the moral strengths go, so to speak, alongside
of specific, independently admirable, nonmoral ground projects
and dominant personal traits.
When one does finally turn one's eyes toward lives that are dominated
by explicitly moral commitments, moreover, one finds oneself
relieved at the discovery of idiosyncrasies or eccentricities not quite
in line with the picture of moral perfection. One prefers the blunt,
tactless, and opinionated Betsy Trotwood to the unfailingly kind
and patient Agnes Copperfield; one prefers the mischievousness
and the sense of irony in Chesterton's Father Brown to the innocence
and undiscriminating love of St. Francis.
It seems that, as we look in our ideals for people who achieve
nonmoral varieties of personal excellence in conjunction with or
colored by some version of high moral tone, we look in our paragons
of moral excellence for people whose moral achievements
occur in conjunction with or colored by some interests or traits that
have low moral tone. In other words, there seems to be a limit to
how much morality we can stand.
One might suspect that the essence of the problem is simply that
there is a limit to how much of any single value, or any single type
of value, we can stand. Our objection then would not be specific to
a life in which one's dominant concern is morality, but would apply
to any life that can be so completely characterized by an extraordinarily
dominant concern. The objection in that case would
reduce to the recognition that such a life is incompatible with wellroundedness.
If that were the objection, one could fairly reply that
well-roundedness is no more supreme a virtue than the totality of
moral virtues embodied by the ideal it is being used to criticize. But
I think this misidentifies the objection. For the way in which a
concern for morality may dominate a life, or, more to the point, the
way in which it may dominate an ideal of life, is not easily imagined
by analogy to the dominance an aspiration to become an
Olympic swimmer or a concern pianist might have.
A person who is passionately committed to one of these latter
concerns might decide that her attachment to it is strong enough to
be worth the sacrifice of her ability to maintain and pursue a significant
portion of what else life might offer which a proper devotion
to her dominant passion would require. But a desire to be as
morally good as possible is not likely to take the form of one desire
among others which, because of its peculiar psychological
strength, requires one to forego the pursuit of other weaker and separately
less demanding desires. Rather, the desire to be as morally
This content downloaded from
147.251.6.77 on Sat, 03 Sep 2022 08:11:53 UTC
All use subject to https://about.jstor.org/terms
424 THE JOURNAL OF PHILOSOPHY
good as possible is apt to have the character not just of a stronger,
but of a higher desire, which does not merely successfully compete
with one's other desires but which rather subsumes or demotes
them. The sacrifice of other interests for the interest in morality,
then, will have the character, not of a choice, but of an imperative.
Moreover, there is something odd about the idea of morality itself,
or moral goodness, serving as the object of a dominant passion
in the way that a more concrete and specific vision of a goal (even a
concrete moral goal) might be imagined to serve. Morality itself
does not seem to be a suitable object of passion. Thus, when one
reflects, for example, on the Loving Saint easily and gladly giving
up his fishing trip or his stereo or his hot fudge sundae at the drop
of the moral hat, one is apt to wonder not at how much he loves
morality, but at how little he loves these other things. One thinks
that, if he can give these up so easily, he does not know what it is
to truly love them. There seems, in other words, to be a kind of joy
which the Loving Saint, either by nature or by practice, is incapable
of experiencing. The Rational Saint, on the other hand, might
retain strong nonmoral and concrete desires-he simply denies
himself the opportunity to act on them. But this is no less troubling.
The Loving Saint one might suspect of missing a piece of
perceptual machinery, of being blind to some of what the world
has to offer. The Rational Saint, who sees it but foregoes it, one
suspects of having a different problem-a pathological fear of
damnation, perhaps, or an extreme form of self-hatred that interferes
with his ability to enjoy the enjoyable in life.
In other words, the ideal of a life of moral sainthood disturbs not
simply because it is an ideal of a life in which morality unduly
dominates. The normal person's direct and specific desires for objects,
activities, and events that conflict with the attainment of
moral perfection are not simply sacrificed but removed, suppressed,
or subsumed. The way in which morality, unlike other possible
goals, is apt to dominate is particularly disturbing, for it seems to
require either the lack or the denial of the existence of an identifiable,
personal self.
This distinctively troubling feature is not, I think, absolutely
unique to the ideal of the moral saint, as I have been using that
phrase. It is shared by the conception of the pure aesthete, by a certain
kind of religious ideal, and, somewhat paradoxically, by the
model of the thorough-going, self-conscious egoist. It is not a coincidence
that the ways of comprehending the world of which these
ideals are the extreme embodiments are sometimes described as
This content downloaded from
147.251.6.77 on Sat, 03 Sep 2022 08:11:53 UTC
All use subject to https://about.jstor.org/terms
MORAL SAINTS 425
"moralities" themselves. At any rate, they compete with what we
ordinarily mean by 'morality'. Nor is it a coincidence that these
ideals are naturally described as fanatical. But it is easy to see that
these other types of perfection cannot serve as satisfactory personal
ideals; for the realization of these ideals would be straightforwardly
immoral. It may come as a surprise to some that there may in addition
be such a thing as a moral fanatic.
Some will object that I am being unfair to "common-sense morality"
-that it does not really require a moral saint to be either a disgusting
goody-goody or an obsessive ascetic. Admittedly, there is
no logical inconsistency between having any of the personal characteristics
I have mentioned and being a moral saint. It is not morally
wrong to notice the faults and shortcomings of others or to
recognize and appreciate nonmoral talents and skills. Nor is it immoral
to be an avid Celtics fan or to have a passion for caviar or to
be an excellent cellist. With enough imagination, we can always
contrive a suitable history and set of circumstances that will embrace
such characteristics in one or another specific fictional story
of a perfect moral saint.
If one turned onto the path of moral sainthood relatively late in
life, one may have already developed interests that can be turned to
moral purposes. It may be that a good golf game is just what is
needed to secure that big donation to Oxfam. Perhaps the cultivation
of one's exceptional artistic talent will turn out to be the way
one can make one's greatest contribution to society. Furthermore,
one might stumble upon joys and skills in the very service of morality.
If, because the children are short a ninth player for the team,
one's generous offer to serve reveals a natural fielding arm or if
one's part in the campaign against nuclear power requires accepting
a lobbyist's invitation to lunch at Le Lion d'Or, there is no
moral gain in denying the satisfaction one gets froni these activities.
The moral saint, then, may, by happy accident, find himself
with nonmoral virtues on which he can capitalize morally or
which make psychological demands to which he has no choice but
to attend. The point is that, for a moral saint, the existence of these
interests and skills can be given at best the status of happy accidents-they
cannot be encouraged for their own sakes as distinct,
independent aspects of the realization of human good.
It must be remembered that from the fact that there is a tension
between having any of these qualities and being a moral saint it
does not follow that having any of these qualities is immoral. For
it is not part of common-sense morality that one ought to be a
This content downloaded from
147.251.6.77 on Sat, 03 Sep 2022 08:11:53 UTC
All use subject to https://about.jstor.org/terms
426 THE JOURNAL OF PHILOSOPHY
moral saint. Still, if someone just happened to want to be a moral
saint, he or she would not have or encourage these qualities, and,
on the basis of our common-sense values, this counts as a reason
not to want to be a moral saint.
One might still wonder what kind of reason this is, and what
kind of conclusion this properly allows us to draw. For the fact
that the models of moral saints are unattractive does not necessarily
mean that they are unsuitable ideals. Perhaps they are unattractive
because they make us feel uncomfortable-they highlight our own
weaknesses, vices, and flaws. If so, the fault lies not in the characters
of the saints, but in those of our unsaintly selves.
To be sure, some of the reasons behind the disaffection we feel
for the model of moral sainthood have to do with a reluctance to
criticize ourselves and a reluctance to committing ourselves to trying
to give up activities and interests that we heartily enjoy. These
considerations might provide an excuse for the fact that we are not
moral saints, but they do not provide a basis for criticizing sainthood
as a possible ideal. Since these considerations rely on an appeal
to the egoistic, hedonistic side of our natures, to use them as a
basis for criticizing the ideal of the moral saint would be at best to
beg the question and at worst to glorify features of ourselves that
ought to be condemned.
The fact that the moral saint would be without qualities which
we have and which, indeed, we like to have, does not in itself provide
reason to condemn the ideal of the moral saint. The fact that
some of these qualities are good qualities, however, and that they
are qualities we ought to like, does provide reason to discourage
this ideal and to offer other ideals in its place. In other words, some
of the qualities the moral saint necessarily lacks are virtues, albeit
nonmoral virtues, in the unsaintly characters who have them. The
feats of Groucho Marx, Reggie Jackson, and the head chef at
Lutece are impressive accomplishments that it is not only permissible
but positively appropriate to recognize as such. In general, the
admiration of and striving toward achieving any of a great variety
of forms of personal excellence are character traits it is valuable
and desirable for people to have. In advocating the development of
these varieties of excellence, we advocate nonmoral reasons for acting,
and in thinking that it is good for a person to strive for an
ideal that gives a substantial role to the interests and values that
correspond to these virtues, we implicitly acknowledge the goodness
of ideals incompatible with that of the moral saint. Finally, if
we think that it is as good, or even better for a person to strive for
one of these ideals than it is for him or her to strive for and realize
This content downloaded from
147.251.6.77 on Sat, 03 Sep 2022 08:11:53 UTC
All use subject to https://about.jstor.org/terms
MORAL SAINTS 427
the ideal of the moral saint, we express a conviction that it is good
not to be a moral saint.
MORAL SAINTS AND MORAL THEORIES
I have tried so far to paint a picture-or, rather, two pictures-of
what a moral saint might be like, drawing on what I take to be the
attitudes and beliefs about morality prevalent in contemporary,
common-sense thought. To my suggestion that common-sense
morality generates conceptions of moral saints that are unattractive
or otherwise unacceptable, it is open to someone to reply, "so
much the worse for common-sense morality." After all, it is often
claimed that the goal of moral philosophy is to correct and improve
upon common-sense morality, and I have as yet given no attention
to the question of what conceptions of moral sainthood, if
any, are generated from the leading moral theories of our time.
A quick, breezy reading of utilitarian and Kantian writings will
suggest the images, respectively, of the Loving Saint and the Rational
Saint. A utilitarian, with his emphasis on happiness, will
certainly prefer the Loving Saint to the Rational one, since the Loving
Saint will himself be a happier person than the Rational Saint.
A Kantian, with his emphasis on reason, on the other hand, will
find at least as much to praise in the latter as in the former. Still,
both models, drawn as they are from common sense, appeal to an
impure mixture of utilitarian and Kantian intuitions. A more careful
examination of these moral theories raises questions about
whether either model of moral sainthood would really be advocated
by a believer in the explicit doctrines associated with either of these
views.
Certainly, the utilitarian in no way denies the value of self-realization.
He in no way disparages the development of interests, talents,
and other personally attractive traits that I have claimed the
moral saint would be without. Indeed, since just these features enhance
the happiness both of the individuals who possess them and
of those with whom they associate, the ability to promote these features
both in oneself and in others will have considerable positive
weight in utilitarian calculations.
This implies that the utilitarian would not support moral sainthood
as a universal ideal. A world in which everyone, or even a
large number of people, achieved moral sainthood-even a world
in which they strove to achieve it-would probably contain less
happiness than a world in which people realized a diversity of
ideals involving a variety of personal and perfectionist values.
More pragmatic considerations also suggest that, if the utilitarian
wants to influence more people to achieve more good, then he
This content downloaded from
147.251.6.77 on Sat, 03 Sep 2022 08:11:53 UTC
All use subject to https://about.jstor.org/terms
428 THE JOURNAL OF PHILOSOPHY
would do better to encourage them to pursue happiness-producing
goals that are more attractive and more within a normal person's
reach.
These considerations still leave open, however, the question of
what kind of an ideal the committed utilitarian should privately
aspire to himself. Utilitarianism requires him to want to achieve
the greatest general happiness, and this would seem to commit him
to the ideal of the moral saint.
One might try to use the claims I made earlier as a basis for an
argument that a utilitarian should choose to give up utilitarianism.
If, as I have said, a moral saint would be a less happy person
both to be and to be around than many other possible ideals, perhaps
one could create more total happiness by not trying too hard
to promote the total happiness. But this argument is simply unconvincing
in light of the empirical circumstances of our world.
The gain in happiness that would accrue to oneself and one's
neighbors by a more well-rounded, richer life than that of the
moral saint would be pathetically small in comparison to the
amount by which one could increase the general happiness if one
devoted oneself explicitly to the care of the sick, the downtrodden,
the starving, and the homeless. Of course, there may be psychological
limits to the extent to which a person can devote himself to
such things without going crazy. But the utilitarian's individual
limitations would not thereby become a positive feature of his personal
ideals.
The unattractiveness of the moral saint, then, ought not rationally
convince the utilitarian to abandon his utilitarianism. It may,
however, convince him to take efforts not to wear his saintly moral
aspirations on his sleeve. If it is not too difficult, the utilitarian
will try not to make those around him uncomfortable. He will not
want to appear "holier than thou"; he will not want to inhibit
others' ability to enjoy themselves. In practice, this might make the
perfect utilitarian a less nauseating companion than the moral
saint I earlier portrayed. But insofar as this kind of reasoning produces
a more bearable public personality, it is at the cost of giving
him a personality that must be evaluated as hypocritical and condescending
when his private thoughts and attitudes are taken into
account.
Still, the criticisms I have raised against the saint of commonsense
morality should make some difference to the utilitarian's
conception of an ideal which neither requires him to abandon his
utilitarian principles nor forces him to fake an interest he does not
This content downloaded from
147.251.6.77 on Sat, 03 Sep 2022 08:11:53 UTC
All use subject to https://about.jstor.org/terms
MORAL SAINTS 429
have or a judgment he does not make. For it may be that a limited
and carefully monitored allotment of time and energy to be devoted
to the pursuit of some nonmoral interests or to the development of
some nonmoral talents would make a person a better contributor to
the general welfare than he would be if he allowed himself no indulgences
of this sort. The enjoyment of such activities in no way
compromises a commitment to utilitarian principles as long as the
involvement with these activities is conditioned by a willingness to
give them up whenever it is recognized that they cease to be in the
general interest.
This will go some way in mitigating the picture of the loving
saint that an understanding of utilitarianism will on first impression
suggest. But I think it will not go very far. For the limitations
on time and energy will have to be rather severe, and the need to
monitor will restrict not only the extent but also the quality of
one's attachment to these interests and traits. They are only weak
and somewhat peculiar sorts of passions to which one can consciously
remain so conditionally committed. Moreover, the way in
which the utilitarian can enjoy these "extra-curricular" aspects of
his life is simply not the way in which these aspects are to be enjoyed
insofar as they figure into our less saintly ideals.
The problem is not exactly that the utilitarian values these aspects
of his life only as a means to an end, for the enjoyment he
and others get from these aspects are not a means to, but a part of,
the general happiness. Nonetheless, he values these things only because
of and insofar as they are a part of the general happiness. He
values them, as it were, under the description 'a contribution to the
general happiness'. This is to be contrasted with the various ways
in which these aspects of life may be valued by nonutilitarians. A
person might love literature because of the insights into human nature
literature affords. Another might love the cultivation of roses
because roses are things of great beauty and delicacy. It may be true
that these features of the respective activities also explain why these
activities are happiness-producing. But, to the nonutilitarian, this
may not be to the point. For if one values these activities in these
more direct ways, one may not be willing to exchange them for
others that produce an equal, or even a greater amount of happiness.
From that point of view, it is not because they produce happiness
that these activities are valuable; it is because these activities
are valuable in more direct and specific ways that they produce
happiness.
To adopt a phrase of Bernard Williams', the utilitarian's manner
This content downloaded from
147.251.6.77 on Sat, 03 Sep 2022 08:11:53 UTC
All use subject to https://about.jstor.org/terms
430 THE JOURNAL OF PHILOSOPHY
of valuing the not explicitly moral aspects of his life "provides
(him) with one thought too many".' The requirement that the u
tarian have this thought-periodically, at least-is indicative of not
only a weakness but a shallowness in his appreciation of the aspects
in question. Thus, the ideals toward which a utilitarian
could acceptably strive would remain too close to the model of the
common-sense moral saint to escape the criticisms of that model
which I earlier suggested. Whether a Kantian would be similarly
committed to so restrictive and unattractive a range of possible
ideals is a somewhat more difficult question.
The Kantian believes that being morally worthy consists in always
acting from maxims that one could will to be universal law,
and doing this not out of any pathological desire but out of reverence
for the moral law as such. Or, to take a different formulation
of the categorical imperative, the Kantian believes that moral action
consists in treating other persons always as ends and never as
means only. Presumably, and according to Kant himself, the Kantian
thereby commits himself to some degree of benevolence as well
as to the rules of fair play. But we surely would not will that every
person become a moral saint, and treating others as ends hardly requires
bending over backwards to protect and promote their interests.
On one interpretation of Kantian doctrine, then, moral perfection
would be achieved simply by unerring obedience to a limited
set of side-constraints. On this interpretation, Kantian theory
simply does not yield an ideal conception of a person of any fullness
comparable to that of the moral saints I have so far been
portraying.
On the other hand, Kant does say explicitly that we have a duty
of benevolence, a duty not only to allow others to pursue their
ends, but to take up their ends as our own. In addition, we have
positive duties to ourselves, duties to increase our natural as well as
our moral perfection. These duties are unlimited in the degree to
which they may dominate a life. If action in accordance with and
motivated by the thought of these duties is considered virtuous, it is
natural to assume that the more one performs such actions, the
more virtuous one is. Moreover, of virtue in general Kant says, "it
is an ideal which is unattainable while yet our duty is constantly to
approximate to it".2 On this interpretation, then, the Kantian
'"Persons, Character and Morality" in Amelie Rorty, ed., The Identities of Persons
(Berkeley: Univ. of California Press, 1976), p. 214.
2Immanuel Kant, The Doctrine of Virtue, Mary J. Gregor, trans. (New York:
Harper & Row, 1964), p. 71.
This content downloaded from
147.251.6.77 on Sat, 03 Sep 2022 08:11:53 UTC
All use subject to https://about.jstor.org/terms
MORAL SAINTS 431
moral saint, like the other moral saints I have been considering, is
dominated by the motivation to be moral.
Which of these interpretations of Kant one prefers will depend
on the interpretation and the importance one gives to the role of
the imperfect duties in Kant's over-all system. Rather than choose
between them here, I shall consider each briefly in turn.
On the second interpretation of Kant, the Kantian moral saint is,
not surprisingly, subject to many of the same objections I have
been raising against other versions of moral sainthood. Though
the Kantian saint may differ from the utilitarian saint as to which
actions he is bound to perform and which he is bound to refrain
from performing, I suspect that the range of activities acceptable to
the Kantian saint will remain objectionably restrictive. Moreover,
the manner in which the Kantian saint must think about and justify
the activities he pursues and the character traits he develops
will strike us, as it did with the utilitarian saint, as containing
"one thought too many." As the utilitarian could value his activities
and character traits only insofar as they fell under the description
of 'contributions to the general happiness', the Kantian would
have to value his activities and character traits insofar as they were
manifestations of respect for the moral law. If the development of
our powers to achieve physical, intellectual, or artistic excellence,
or the activities directed toward making others happy are to have
any moral worth, they must arise from a reverence for the dignity
that members of our species have as a result of being endowed with
pure practical reason. This is a good and noble motivation, to be
sure. But it is hardly what one expects to be dominantly behind a
person's aspirations to dance as well as Fred Astaire, to paint as
well as Picasso, or to solve some outstanding problem in abstract
algebra, and it is hardly what one hopes to find lying dominantly
behind a father's action on behalf of his son or a lover's on behalf
of her beloved.
Since the basic problem with any of the models of moral sainthood
we have been considering is that they are dominated by a single,
all-important value under which all other possible values must
be subsumed, it may seem that the alternative interpretation of
Kant, as providing a stringent but finite set of obligations and constraints,
might provide a more acceptable morality. According to
this interpretation of Kant, one is as morally good as can be so
long as one devotes some limited portion of one's energies toward
altruism and the maintenance of one's physical and spiritual
health, and otherwise pursues one's independently motivated interests
and values in such a way as to avoid overstepping certain
This content downloaded from
147.251.6.77 on Sat, 03 Sep 2022 08:11:53 UTC
All use subject to https://about.jstor.org/terms
432 THE JOURNAL OF PHILOSOPHY
bounds. Certainly, if it be a requirement of an acceptable moral
theory that perfect obedience to its laws and maximal devotion to
its interests and concerns be something we can wholeheartedly
strive for in ourselves and wish for in those around us, it will count
in favor of this brand of Kantianism that its commands can be fulfilled
without swallowing up the perfect moral agent's entire
personality.
Even this more limited understanding of morality, if its connection
to Kant's views is to be taken at all seriously, is not likely to
give an unqualified seal of approval to the nonmorally directed
ideals I have been advocating. For Kant is explicit about what he
calls "duties of apathy and self-mastery" (69/70)-duties to ensure
that our passions are never so strong as to interfere with calm,
practical deliberation, or so deep as to wrest control from the more
disinterested, rational part of ourselves. The tight and self-conscious
rein we are thus obliged to keep on our commitments to
specific individuals and causes will doubtless restrict our value in
these things, assigning them a necessarily attenuated place.
A more interesting objection to this brand of Kantianism, however,
comes when we consider the implications of placing the kind
of upper bound on moral worthiness which seemed to count in
favor of this conception of morality. For to put such a limit on
one's capacity to be moral is effectively to deny, not just the moral
necessity, but the moral goodness of a devotion to benevolence and
the maintenance of justice that passes beyond a certain, required
point. It is to deny the possibility of going morally above and
beyond the call of a restricted set of duties. Despite my claim that
all-consuming moral saintliness is not a particularly healthy and
desirable ideal, it seems perverse to insist that, were moral saints to
exist, they would not, in their way, be remarkably noble and admirable
figures. Despite my conviction that it is as rational and as
good for a person to take Katharine Hepburn or Jane Austen as her
role model instead of Mother Theresa, it would be absurd to deny
that Mother Theresa is a morally better person.
I can think of two ways of viewing morality as having an upper
bound. First, we can think that altruism and impartiality are indeed
positive moral interests, but that they are moral only if the
degree to which these interests are actively pursued remains within
certain fixed limits. Second, we can think that these positive interests
are only incidentally related to morality and that the essence of
morality lies elsewhere, in, say, an implicit social contract or in the
recognition of our own dignified rationality. According to the first
conception of morality, there is a cut-off line to the amount of alThis
content downloaded from
147.251.6.77 on Sat, 03 Sep 2022 08:11:53 UTC
All use subject to https://about.jstor.org/terms
MORAL SAINTS 433
truism or to the extent of devotion to justice and fairness that is
worthy of moral praise. But to draw this line earlier than the line
that brings the altruist in question into a worse-off position than
all those to whom he devotes himself seems unacceptably artificial
and gratuitous. According to the second conception, these positive
interests are not essentially related to morality at all. But then we
are unable to regard a more affectionate and generous expression of
good will toward others as a natural and reasonable extension of
morality, and we encourage a cold and unduly self-centered approach
to the development and evaluation of our motivations and
concerns.
A moral theory that does not contain the seeds of an all-consuming
ideal of moral sainthood thus seems to place false and unnatural
limits on our opportunity to do moral good and our potential
to deserve moral praise. Yet the main thrust of the arguments of
this paper has been leading to the conclusion that, when such
ideals are present, they are not ideals to which it is particularly
reasonable or healthy or desirable for human beings to aspire.
These claims, taken together, have the appearance of a dilemma
from which there is no obvious escape. In a moment, I shall argue
that, despite appearances, these claims should not be understood as
constituting a dilemma. But, before I do, let me briefly describe
another path which those who are convinced by my above remarks
may feel inclined to take.
If the above remarks are understood to be implicitly critical of
the views on the content of morality which seem most popular
today, an alternative that naturally suggests itself is that we revise
our views about the content of morality. More specifically, my remarks
may be taken to support a more Aristotelian, or even a more
Nietzschean, approach to moral philosophy. Such a change in approach
involves substantially broadening or replacing our contemporary
intuitions about which character traits constitute moral
virtues and vices and which interests constitute moral interests. If,
for example, we include personal bearing, or creativity, or sense of
style, as features that contribute to one's moral personality, then we
can create moral ideals which are incompatible with and probably
more attractive than the Kantian and utilitarian ideals I have discussed.
Given such an alteration of our conception of morality, the
figures with which I have been concerned above might, far from
being considered to be moral saints, be seen as morally inferior to
other more appealing or more interesting models of individuals.
This approach seems unlikely to succeed, if for no other reason,
because it is doubtful that any single, or even any reasonably small
This content downloaded from
147.251.6.77 on Sat, 03 Sep 2022 08:11:53 UTC
All use subject to https://about.jstor.org/terms
434 THE JOURNAL OF PHILOSOPHY
number of substantial personal ideals could capture the full range
of possible ways of realizing human potential or achieving human
good which deserve encouragement and praise. Even if we could
provide a sufficiently broad characterization of the range of positive
ways for human beings to live, however, I think there are
strong reasons not to want to incorporate such a characterization
more centrally into the framework of morality itself. For, in claiming
that a character trait or activity is morally good, one claims
that there is a certain kind of reason for developing that trait or engaging
in that activity. Yet, lying behind our criticism of more
conventional conceptions of moral sainthood, there seems to be a
recognition that among the immensely valuable traits and activities
that a human life might positively embrace are some of which we
hope that, if a person does embrace them, he does so not for moral
reasons. In other words, no matter how flexible we make the guide
to conduct which we choose to label "morality," no matter how
rich we make the life in which perfect obedience to this guide
would result, we will have reason to hope that a person does not
wholly rule and direct his life by the abstract and impersonal consideration
that such a life would be morally good.
Once it is recognized that morality itself should not serve as a
comprehensive guide to conduct, moreover, we can see reasons to
retain the admittedly vague contemporary intuitions about what
the classification of moral and nonmoral virtues, interests, and the
like should be. That is, there seem to be important differences between
the aspects of a person's life which are currently considered
appropriate objects of moral evaluation and the aspects that might
be included under the altered conception of morality we are now
considering, which the latter approach would tend wrongly to blur
or to neglect. Moral evaluation now is focused primarily on features
of a person's life over which that person has control; it is largely
restricted to aspects of his life which are likely to have considerable
effect on other people. These restrictions seem as they should be.
Even if responsible people could reach agreement as to what constituted
good taste or a healthy degree of well-roundedness, for example,
it seems wrong to insist that everyone try to achieve these
things or to blame someone who fails or refuses to conform.
If we are not to respond to the unattractiveness of the moral
ideals that contemporary theories yield either by offering alternative
theories with more palatable ideals or by understanding these
theories in such a way as to prevent them from yielding ideals at
all, how, then, are we to respond? Simply, I think, by admitting
This content downloaded from
147.251.6.77 on Sat, 03 Sep 2022 08:11:53 UTC
All use subject to https://about.jstor.org/terms
MORAL SAINTS 435
that moral ideals do not, and need not, make the best personal
ideals. Earlier, I mentioned one of the consequences of regarding as
a test of an adequate moral theory that perfect obedience to its laws
and maximal devotion to its interests be something we can wholeheartedly
strive for in ourselves and wish for in those around us.
Drawing out the consequences somewhat further should, I think,
make us more doubtful of the proposed test than of the theories
which, on this test, would fail. Given the empirical circumstances
of our world, it seems to be an ethical fact that we have unlimited
potential to be morally good, and endless opportunity to promote
moral interests. But this is not incompatible with the not-so-ethical
fact that we have sound, compelling, and not particularly selfish
reasons to choose not to devote ourselves univocally to realizing
this potential or to taking up this opportunity.
Thus, in one sense at least, I am not really criticizing either Kantianism
or utilitarianism. Insofar as the point of view I am offering
bears directly on recent work in moral philosophy, in fact, it bears
on critics of these theories who, in a spirit not unlike the spirit of
most of this paper, point out that the perfect utilitarian would be
flawed in this way or the perfect Kantian flawed in that.3 The assumption
lying behind these claims, implicitly or explicitly, has
been that the recognition of these flaws shows us something wrong
with utilitarianism as opposed to Kantianism, or something wrong
with Kantianism as opposed to utilitarianism, or something wrong
with both of these theories as opposed to some nameless third alternative.
The claims of this paper suggest, however, that this assumption
is unwarranted. The flaws of a perfect master of a moral
theory need not reflect flaws in the intramoral content of the theory
itself.
MORAL SAINTS AND MORAL PHILOSOPHY
In pointing out the regrettable features and the necessary absence
of some desirable features in a moral saint, I have not meant to
condemn the moral saint or the person who aspires to become one.
Rather, I have meant to insist that the ideal of moral sainthood
should not be held as a standard against which any other ideal
must be judged or justified, and that the posture we take in response
to the recognition that our lives are not as morally good as
3See, e.g., Williams, op. cit. and J. J. C. Smart and Bernard Williams, Utilitarianism:
For and Against (New York: Cambridge, 1973). Also, Michael Stocker, "The
Schizophrenia of Modern Ethical Theories," this JOURNAL, LXIII, 14 (Aug. 12, 1976):
453-466.
This content downloaded from
147.251.6.77 on Sat, 03 Sep 2022 08:11:53 UTC
All use subject to https://about.jstor.org/terms
436 THE JOURNAL OF PHILOSOPHY
they might be need not be defensive.4 It is misleading to insist that
one is permitted to live a life in which the goals, relationships, activities,
and interests that one pursues are not maximally morally
good. For our lives are not so comprehensively subject to the requirement
that we apply for permission, and our nonmoral reasons
for the goals we set ourselves are not excuses, but may rather be
positive, good reasons which do not exist despite any reasons that
might threaten to outweigh them. In other words, a person may be
perfectly wonderful without being perfectly moral.
Recognizing this requires a perspective which contemporary
moral philosophy has generally ignored. This perspective yields
judgments of a type that is neither moral nor egoistic. Like moral
judgments, judgments about what it would be good for a person to
be are made from a point of view outside the limits set by the
values, interests, and desires that the person might actually have.
And, like moral judgments, these judgments claim for themselves a
kind of objectivity or a grounding in a perspective which any rational
and perceptive being can take up. Unlike moral judgments,
however, the good with which these judgments are concerned is
not the good of anyone or any group other than the individual
himself.
Nonetheless, it would be equally misleading to say that these
judgments are made for the sake of the individual himself. For
these judgments are not concerned with what kind of life it is in a
person's interest to lead, but with what kind of interests it would be
good for a person to have, and it need not be in a person's interest
that he acquire or maintain objectively good interests. Indeed, the
model of the Loving Saint, whose interests are identified with the
interests of morality, is a model of a person for whom the dictates
of rational self-interest and the dictates of morality coincide. Yet, I
have urged that we have reason not to aspire to this ideal and that
some of us would have reason to be sorry if our children aspired to
and achieved it.
The moral point of view, we might say, is the point of view one
takes up insofar as one takes the recognition of the fact that one is
just one person among others equally real and deserving of the
4George Orwell makes a similar point in "Reflections on Gandhi," in A Collec
tion of Essays by George Orwell (New York: Harcourt Brace Jovanovich, 1945),
p. 176: "sainthood is . . . a thing that human beings must avoid. . . It is too readily
assumed that . . . the ordinary man only rejects it because it is too difficult; in
other words, that the average human being is a failed saint. It is doubtful whether
this is true. Many people genuinely do not wish to be saints, and it is probable that
some who achieve or aspire to sainthood have never felt much temptation to be
human beings."
This content downloaded from
147.251.6.77 on Sat, 03 Sep 2022 08:11:53 UTC
All use subject to https://about.jstor.org/terms
MORAL SAINTS 437
good things in life as a fact with practical consequences, a fact the
recognition of which demands expression in one's actions and in
the form of one's practical deliberations. Competing moral theories
offer alternative answers to the question of what the most correct or
the best way to express this fact is. In doing so, they offer alternative
ways to evaluate and to compare the variety of actions, states of
affairs, and so on that appear good and bad to agents from other,
nonmoral points of view. But it seems that alternative interpretations
of the moral point of view do not exhaust the ways in which
our actions, characters, and their consequences can be comprehensively
and objectively evaluated. Let us call the point of view from
which we consider what kinds of lives are good lives, and what
kinds of persons it would be good for ourselves and others to be,
the point of view of individual perfection.
Since either point of view provides a way of comprehensively
evaluating a person's life, each point of view takes account of, and,
in a sense, subsumes the other. From the moral point of view, the
perfection of an individual life will have some, but limited, valuefor
each individual remains, after all, just one person among others.
From the perfectionist point of view, the moral worth of an individual's
relation to his world will likewise have some, but limited,
value-for, as I have argued, the (perfectionist) goodness of an individual's
life does not vary proportionally with the degree to
which it exemplifies moral goodness.
It may not be the case that the perfectionist point of view is like
the moral point of view in being a point of view we are ever obliged
to take up and express in our actions. Nonetheless, it provides us
with reasons that are independent of moral reasons for wanting
ourselves and others to develop our characters and live our lives in
certain ways. When we take up this point of view and ask how
much it would be good for an individual to act from the moral
point of view, we do not find an obvious answer.5
The considerations of this paper suggest, at any rate, that the
answer is not "as much as possible." This has implications both
for the continued development of moral theories and for the development
of metamoral views and for our conception of moral phil-
5A similar view, which has strongly influenced mine, is expressed by Thomas
Nagel in "The Fragmentation of Value," in Mortal Questions (New York: Cambridge,
1979), pp. 128-141. Nagel focuses on the difficulties such apparently incommensurable
points of view create for specific, isolable practical decisions that
must be made both by individtuals and by societies. In focusing on the way in which
these points of view figure into the development of individual personal ideals, the
questions with which I am concerned are more likely to lurk in the background of
any individual's life.
This content downloaded from
147.251.6.77 on Sat, 03 Sep 2022 08:11:53 UTC
All use subject to https://about.jstor.org/terms
438 THE JOURNAL OF PHILOSOPHY
osophy more generally. From the moral point of view, we have
reasons to want people to live lives that seem good from outside
that point of view. If, as I have argued, this means that we have
reason to want people to live lives that are not morally perfect,
then any plausible moral theory must make use of some conception
of supererogation.6
If moral philosophers are to address themselves at the most basic
level to the question of how people should live, however, they must
do more than adjust the content of their moral theories in ways
that leave room for the affirmation of nonmoral values. They must
examine explicitly the range and nature of these nonmoral values,
and, in light of this examination, they must ask how the acceptance
of a moral theory is to be understood and acted upon. For the
claims of this paper do not so much conflict with the content of
any particular currently popular moral theory as they call into
question a metamoral assumption that implicitly surrounds discussions
of moral theory more generally. Specifically, they call into
question the assumption that it is always better to be morally better.
The role morality plays in the development of our characters and
the shape of our practical deliberations need be neither that of a
universal medium into which all other values must be translated
nor that of an ever-present filter through which all other values
must pass. This is not to say that moral value should not be an
important, even the most important, kind of value we attend to in
evaluating and improving ourselves and our world. It is to say that
our values cannot be fully comprehended on the model of a hierarchical
system with morality at the top.
The philosophical temperament will naturally incline, at this
point, toward asking, "What, then, is at the top-or, if there is no
top, how are we to decide when and how much to be moral?" In
other words, there is a temptation to seek a metamoral-though
not, in the standard sense, metaethical-theory that will give us
6The variety of forms that a conception of supererogation might take, however,
has not generally been noticed. Moral theories that make use of this notion typically
do so by identifying some specific set of principles as universal moral requirements
and supplement this list with a further set of directives which it is morally praiseworthy
but not required for an agent to follow. [See, e.g., Charles Fried, Right and
Wrong (Cambridge, Mass.: Harvard, 1979).] But it is possible that the ability to live
a morally blameless life cannot be so easily or definitely secured as this type of theory
would suggest. The fact that there are some situations in which an agent is
morally required to do something and other situations in which it would be good
but not required for an agent to do something does not imply that there are specific
principles such that, in any situation, an agent is required to act in accordance with
these principles and other specific principles such that, in any situation, it would be
good but not required for an agent to act in accordance with those principles.
This content downloaded from
147.251.6.77 on Sat, 03 Sep 2022 08:11:53 UTC
All use subject to https://about.jstor.org/terms
ON DOING GOOD 439
principles, or, at least, informal directives on the basis of which we
can develop and evaluate more comprehensive personal ideals.
Perhaps a theory that distinguishes among the various roles a person
is expected to play within a life-as professional, as citizen, as
friend, and so on-might give us some rules that would offer us, if
nothing else, a better framework in which to think about and discuss
these questions. I am pessimistic, however, about the chances
of such a theory to yield substantial and satisfying results. For I do
not see how a metamoral theory could be constructed which would
not be subject to considerations parallel to those which seem inherently
to limit the appropriateness of regarding moral theories as ultimate
comprehensive guides for action.
This suggests that, at some point, both in our philosophizing
and in our lives, we must be willing to raise normative questions
from a perspective that is unattached to a commitment to any particular
well-ordered system of values. It must be admitted that, in
doing so, we run the risk of finding normative answers that diverge
from the answers given by whatever moral theory one accepts.
This, I take it, is the grain of truth in G. E. Moore's "open question"
argument. In the background of this paper, then, there lurks
a commitment to what seems to me to be a healthy form of intuitionism.
It is a form of intuitionism which is not intended to take
the place of more rigorous, systematically developed, moral theories-rather,
it is intended to put these more rigorous and systematic
moral theories in their place.
SUSAN WOLF
University of Maryland
ON DOING GOOD: THE RIGHT AND THE WRONG WAY
N this paper I am interested in the relations governing three
fundamental moral injunctions: those bidding us to
(i) Do as much good as we can.
(ii) Save lives where we can.
(iii) Refrain from killing.
Recent ethicists have tended to focus on the relation between (ii)
and (iii), on the question of whether there is any intrinsic moral
difference between failing to save a life (letting someone die) and
taking a life (killing someone). But though I shall often have occasion
to refer to the literature on this subject, one of my main con-
0022-362X/82/7908/0439$01.60 ? 1982 The Journal of Philosophy, Inc.
This content downloaded from
147.251.6.77 on Sat, 03 Sep 2022 08:11:53 UTC
All use subject to https://about.jstor.org/terms