Egoism in Ethics

International Phenomenological Society
Philosophy and Phenomenological Research
Egoism in Ethics
Author(s): Kai Nielsen
Source: Philosophy and Phenomenological Research, Vol. 19, No. 4 (Jun., 1959), pp. 502-510
Published by: International Phenomenological Society
Stable URL: http://www.jstor.org/stable/2105117
Accessed: 02-12-2015 15:03 UTC
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at http://www.jstor.org/page/
info/about/policies/terms.jsp
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content
in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship.
For more information about JSTOR, please contact support@jstor.org.
Wiley, International Phenomenological Society and Philosophy and Phenomenological Research are collaborating with
JSTOR to digitize, preserve and extend access to Philosophy and Phenomenological Research.
http://www.jstor.org
This content downloaded from 193.0.65.67 on Wed, 02 Dec 2015 15:03:47 UTC
All use subject to JSTOR Terms and Conditions
EGOISM IN ETHICS
Egoism in ethics may be divided into psychological egoism and ethical

egoism. The former,which must be clearly distinguished from ethical egoism,
is a theory of human motivation. Stated most simply, the psychological
egoist argues that man always in fact seeks his own good. An ethical egoism,
in any form, argues that man ought always to seek his own good. Ethical
egoism may take two logically independent forms: it may be developed as
an ethical egoism of ends or an ethical egoism of means.
The initial plausibility of egoism stems from taking it as a psychological
doctrine of human motivation. If psychological egoism is false, this
certainly does not entail that ethical egoism is also false, but it undermines
the rather superficial persuasiveness of the doctrine. In fact, the truth of
ethical egoism implies the falsity of psychological egoism.1 This can be
seen to be the case by examining the very use of terms like 'ethical egoism'
and 'psychological egoism.' The ethical egoist wishes to say that man ought
always to seek his own good. But 'ought' - at least when used in moral
discourse - is taken to imply 'can.' This means that for any act about
which it is appropriate to say that it ought to be done, it follows that
one can choose either to do it or not to do it. This is usually taken to mean
that man in fact is physically and psychologically able to choose between
alternative courses of action.2 The psychological egoist is asserting that as
a matter of fact man always does seek his own good. But if this is the case,
then the truth of psychological egoism is incompatible with the truth of
ethical egoism. This is true for the following reasons: ethical egoism asserts
that man ought always to seek his own good. 'Ought' is taken to imply
'can' in the quite unexceptional sense in which the ethical egoist wishes to
use it, but if psychological egoism, in asserting that man always in fact
does seek his own good, denies 'can' in the sense of 'factually can,' then it
trivially follows as merely a matter of logic that psychological and ethical
egoism cannot both be true.
1 At this point, as well as several other places in this paper, I am indebted to Pro-
fessor Charles A. Baylis although I would not like to hold him responsible for any
shortcomings that my argument may exhibit.
2 I do not wish this to be confused with any decision for or against the doctrine of
"free will." A deterministic view could well be developed that would not conflict with
anything said above. For one such formulation see Charles A. Baylis, "Rational Prefer-
ence, Determinism and Moral Obligation," The Jaurn-al of Philosophy, Vol. XLVII,
No. 3, February 2, 1950, pp. 57-63.
502
EGOISMIN ETHICS 503
II
However, in saying that the initial plausibility of all forms of ethical

egoism stems from psychological egoism a much simpler point was intend-
ed. The point is simply this: the initial "convincing" character of egoism
stems from the rather obvious psychological observation that an extra-
ordinary amount of our behavior, even that which is "officially" nonego-
istic, is often revealed, in even a relatively superficial introspection, as
egoistic.
The truth or falsity of a doctrine of psychological egoism is, of course, a
complicated psychological question, but it is apparent that man, in the
vast majority of his dealings, acts, as Spinoza put it, so as to inhere in his
own being. To meet the many obvious examples of at least apparently
altruistic behavior, psychological egoism is driven, in attempting to make
its theory plausible at all, to the doctrine that man is not always consciously
egoistic. There is no question, as psychoanalytic studies have made
increasingly apparent, that the "self-sacrifice" of many a mother for her
children is at bottom an effort of a rather crushed personality to gain or
regain lost or wavering self-esteem; or, the "self-sacrifice" may be a
maneuver on the part of a woman, disappointed in her relationships with
her husband, to bind her children to her through acts of "self-sacrifice." 3
This contention of the psychological egoist is ostensibly, at least, an
empirical claim. It is important to realize that it is not just a claim that
people sometimesact this way, but rather the much stronger claim that all
people always act this way. But how could the psychological egoist prove
this? It is not, on his own admission, analytically true. How would we
ever know whether all men always act in that way ? If we mean by 'all men'
all men who are now living, the statement is verifiable in principle, that is,
it is logically possible to verify it. It may even be physically or technically
possible to verify that all men always consciously seek their own good even
in the above contexts.4 But to say in any of the three mentioned senses
that the egoist's claim in verifiable is only to say that it is a meaningful
empirical statement, not that it is a true or even a verified empirical
statement. It only means that it might be true. It does not entail that it is
true or that there is any evidence for it at all. We have only to ask the
psychological egoist what the evidence for his claim is, or what he would take
as a falsification of his claim, to bring out the "metaphysical" nature of his
3 The mother in D. H. Lawrence's Sons and Lovers is a fictional example of this

unfortunately not infrequent character trait.
4 For the distinction between physical, technical and logically possible verification
see Hans Reichenbach, "The Verifiability Theory of Meaning," in Readings in the
Philosophy of Science (Feigl and Brodbeck eds.), p. 97, and Hans Reichenbach, Elements
of Symbolic Logic, Chapter VIII.
504 PHILOSOPHY AND PHENOMENOLOGICAL RESEARCH
contention. True enough, people often do seek their own good and neglect
what is called "the common good." These two things do not prove, how-
ever, that all people always do so or even that they ever do what common-
sensically would be regarded as paradigm cases of altruistic behavior for
essentially selfish reasons. The egoist must supply some evidence that some
people do in fact go to their death willingly because the prior anticipatory
good is so great that it outweighs the resultant evil of death and violent
destruction. But such a proof even if it were forthcoming is not enough.
The psychological egoist must prove that it is at least probable that all
men always act on such egoistic motives in such contexts. I do not see
that there is the slightest evidence for this contention of the egoist.
Without this proof the psychological egoist's contention is mere dogma.
There is a Freudian tack the psychological egoist might take. It might
be possible for a psychological egoist to argue that unconsciously each man
is seeking his own death. There are in all human beings, according to one
theory, "two fundamentally different kinds of instincts, the sexual
instincts, in the widest sense of the word (Eros) and the aggressive instincts,
whose aim is destruction." 5 The so-called death instinct (Thanatos) as it is
fused (as on Freud's theory it always is) with Eros is exhibited in sadism
and masochism. Man's altruistic behavior is really, on this theory, a form
of masochism (pleasure in self-torture) erotically bound.
While the phenomenon of masochism is too well established in psycho-
logical theory to be reasonably denied, it is wel1to distinguish this frequent
clinical phenomenon from a metapsychologicaltheory which would argue
that there is this form of self-aggression in all human beings which,
together with sadistic impulses, constitute an all-pervasive impulse
toward death. If this theory is used to bolster psychological egoism, it is
not conclusively refutable, but it is also without a crucial test that
would establish its truth. Overtly, egoistic behavior can be explained by
the pleasure principle, and all exceptions as cases of either sadism or
masochism. But if this "theory" is to be more than a dogma, we must be
able to say what it would be like for it to be wrong.
The skeptic might say at this point that although my reasoning seems
plausible I must have missed the point of psychological egoism. Psycho-
logical egoism has occurred and reoccurred throughout the history of
ethics from Democritus to Gardner Williams. And in addition to the
philosophers, it is a "theory" that frequently recommends itself to
reflective common sense. If it is so patently absurd, why has it been such
a hardy perennial and why has it seemed so plausible to so many "tough-
minded" people? While neither longevity nor common consent is a good
test for truth, it is the better part of wisdom to pay close attention to such
5 Sigmund Freud, ThreeCotribution4 to the Theoryof Sex, p. 77.
EGOISM IN ETHICS 505
a recurrent theory as psychological egoism. Now, if my above argument is

correct, the plausibility of psychological egoism is very slight indeed: but,
I still need to show why people have come to think it so plausible, and
to lay bare the philosophical malaise that motivates psychological egoism.
It is an analytic truth that we would not do any of the acts we commonly
regard as voluntary acts unless we were motivatedto do them. In ordinary
language, we might restate this so: We only do what we prefer doing or (in
some contexts) dislike least. Let us call this (1). But we easily confuse (1)
with (2), i.e., we only do that which satisfies our own personal (selfish)
desires, or with (3), i.e., we only do that which promotes our own good.
The psychological egoist mistakenly takes (1), (2), and (3) to be equivalent;
but they are not equivalent in any of the usual senses of that word.
I will now try to make good my claim that (1) and (2) are not equivalent.
After the explication of the differences between (1) and (2), the reader
will be able to see readily that a similar explication would apply to (1) and
(3).
It is, of course, a truism that all my preferences must be my own
personal preferences, for only I can have them; but this does not mean
(in any ordinary sense) that my preferences must be selfish preferences.
If we concentrate our attention on the subject of the preferences rather
than on the objects of preference we are apt to think (1) and (2) have the
same meaning. But we must recall that our preferences also are for
something, that is, they have some object. Sometimes we prefer our own
pleasure, happiness, or our own maximum satisfaction of desire (another
vague phrase); but food, sex, companionship, recognition, the welfare of
family and friends and sometimes even mankind as a whole are also the
objects of our preferences.
Williams in defending egoism (reflecting in all probability Butler's
classic critique of psychological egoism) is willing to acknowledge this. He
remarks: "Love is absolutely unselfish in the ordinary sense that it aims
wholly at the welfare of others. But it is absolutely selfish in a Pickwickian
sense; since it is a part of a self, its expression is that self's self-expression,
and its satisfaction is that self's satisfaction." 6 In an ordinary sense we
may say that at least some preferences are unselfish or not egoistic, though
in a Pickwickian sense of 'selfish' all our preferences are "selfish prefer-
ences," i.e., they spring from a person or are located in a given organism.
All preferences are located in some organism (human or infrahuman), but
this is not to say in any ordinary sense that all preferences are "selfish
6 GardnerWilliams,"The Inevitability of Egoism in Ethics," Paper8of the Michigan
Academyof Science and Letters,Vol. XXXVII (1951), p. 494. E. W. Hall's review of
Williams'HumanisticEthics is worth noting with respect to this issue and with respect
to the issue noted in Footnote 9. See E. W. Hall, Ethics, Vol. LXIII, October, 1952,
pp. 69-70.
506 ANDPHENOMENOLOGICAL
PHILOSOPHY RESEARCH
preferences." Yet it is the ordinary sense of 'selfish preference' that causes

worry when we say that all human acts are rooted in preferences. If the
psychological egoist says that egoistic aims (i.e., selfish antisocial aims) are
the only ultimate aims people have he must give some evidence for this.
And it will not help him in the slightest to say that no one would ever do
anything if he didn't want to do it, for this only shows that a voluntary
act, to be a 'voluntary act,' must be a preferred act or an act toward which
the actor has a pro-attitude. It says nothing at all about the goals or objects
of our preferences.
Alternatively, the psychological egoist may take a different tack. If he
says that egoistic aims are the only aims a rational and tough-minded man
ought to have he has left psychological egoism for ethical egoism; and, as
we have seen in Part I, psychological egoism and ethical egoism are
logically incompatible.
Thus (1) is radically different from (2). It is obviously also radically
different from (3). The truth of (1) does not entail the truth of (2) and/or
(3). (1) does not even serve as good evidence for (2) or (3). But if we do not
pay careful heed to the uses or functions of our language it is very easy to
confuse truisms like (1) with statements like (2) and (3). We misunderstand
the logic of our language. Perplexed by this, we then recall how often it is
that people rationalize and how often it is that intentions which parade as
altruistic and high-minded are actually egoistic and selfish in the extreme.
We remember our Ibsen and Cellini as well as our Mill and Epicurus,
though we forget our Tolstoy and Dickins. We combine our odd linguistic
beliefs, which result from a faulty understanding of the functions of our
language, with confirmed empirical beliefs, and conclude, plausibly
(though falsely), that psychological egoism is true. From both the vantage
point of language and from factual observation our vision is blurred by a
one-sided diet.
III
A purely ethical egoism divorced from a psychological egoism is also

difficult to render plausible. Taken as a doctrine of ends the ethical egoist
might argue that all value or good terminates in immediate prizings and
disprizings. The final test of the valuable is what I like in the way of
experience. This experience in its most primitive state is irreducibly
personal, and only later in a state of partly settled inquiry does it become
"shared." Thus, when A says "X is valuable," he is really only saying in a
confusing and elliptical way, "I value X." To be meaningful all valuing
must be identified with a valuer. Good, then, is always "good for me."
That there are "shared goods" and that occasionally self-sacrifice is
deemed fitting, only reflects that some men have gone beyond "human
bondage" and have an "adequate idea" of their own needs and the means
by which they are realized. In short, they have become "enlightened"
egoists and realize that their personal good can be insured in the long run
only if at least occasionally they do perform "altruistic" and "non-
egoistic" acts and urge others to do likewise.7
An ethical egoism would not have to explain away instances of what
would ordinarily be called altruistic behavior as disguised instances of
egoistic behavior, but could simply assert altruistic ideals are mistaken
moral ideals and that we oughtto seek only our own good.
It is plausible to contend that an ethical egoism of ends could justify a
goodly number of our obviously nonegoistic acts as instrumental to an
enlightened self-interest, but again the difficulty returns of explaining all
nonegoistic ethical action on this basis. It is most paradoxical on such a
theory to say that a man who volunteers for a rear-guard action that will
most probably result in his death considers this as a good instrumental to
his welfare. This might be plausible if the egoist in question were also a
supernaturalist and felt that this action would be instrumental to his
reward in heaven. But it is at least thinkable that the egoist might be an
atheist and how, if he were an atheist, this act could then be viewed as
personally instrumentally good is far from clear.
When Professor X (an ethical egoist) claims that the end of all moral
action ought to be the self-interest of the individual involved, what is he
claiming? He may be taken to mean that all men ought to seek their own
self-interest as an end, but how then will the fact - if it is a fact - that all
valuings terminate in the immediately prized help him. The latter is a
factual statement, but Professor X's statement is a normative statement
stating a moral criterion. The fact that all valuings are "for me" in the
sense that all value is derivative from direct prizings does not entail, in
any usual sense of 'entail,' that "all men ought to seek their own interest
as an end." From the factual premise about valuings we could with as
much (or as little) justification derive the normative premise, "the only
worthy moral end is the cultivation of a Good Will," or any other normative
premise that might be desired.
Alternatively, Professor X might be taken, though with less plausibility,
to be asserting not the universal statement that all men ought to seek their
own good, but only that he, Professor X, ought to seek solely his own good.
But then his statement would only tell us what Professor X ought to do
and not what all men, a group of men, or even any other men ought to do.
A moral criterion is public. Thus his statement could not possibly serve as
a moral criterion for any morality no matter how iconoclastic. The very
7 This type of view has been briefly but baldly stated by Gardner Williams in
"Ethics for Scientific Humanists" and "The Relativity of Right" in Humanist World
Digest, November 1956 and February 1957, respectively.
508 PHILOSOPHY
ANDPHENOMENOLOGICAL
RESEARCH
uses of both 'moral' and 'criterion' make the above egoistic use of 'moral
criterion' completely unintelligible. The egoist's use of 'moral criterion'
does not even bear a family resemblance to the ordinary uses of 'moral
criterion.' We have only similar tokens (sign vehicles) with completely
different uses.
The final alternative for egoism in ethics is ethical egoism as a doctrine
of means. As a doctrine of means, ethical egoism could be completely
compatible with an altruistic doctrine of ends or with an objective ethical
theory. This last variety of ethical egoist (Professor Y may serve as its
protagonist) need merely assert that good-on-the-whole or what is
commonly called "the common good" can best be achieved by each man
consciously seeking his own good. Yet in the face of the above-mentioned
exceptions to what would normally be considered egoistic behavior, an
ethical egoism of means is also difficult to hold. It is usually felt that
certain acts involving self-sacrifice ought, under certain conditions, to be
performed to insure the common good. An ethical egoist of this type might
argue that a common good that is nobody's individual good would not be a
suitable ideal and would not be in fact what is usually meant by 'good' at
all. This seems to be obviously (and tritely) true. Certainly, it is important
to take proper account, in constructing criteria for good-on-the-whole, of
the goods of particular individuals. Yet if psychological egoism is given up
and an ethical egoism of ends is given up, an ethical egoism as a doctrine of
means seems at a loss to explain the obvious "fact" of everyday moral
experience that self-sacrifice - and sometimes even extreme self-sacrifice
- is upon occasion necessary for the common good. Ethical egoism must
contend that it is never right for a soldier to volunteer for a rear-guard
action that will almost certainly mean his own death. This must be so, no
matter how just the war or how great the needs of his comrades and
country. An ethical egoism of this type makes the factual claim that
the exclusive seeking one's own good always best contributes to good-
on-the-whole. We are not dealing here with purely moral principles but
with whether certain moral principles will be more casually effective
in achieving certain goals, e.g., good-on-the-whole. Yet clearly sometimes
in cases like that of a man or group of men fighting a rear-guard action
more total good is realized in the universe than if the rear-guard fighters
had followed through on egoistic principles.
The ethical egoist can save himself here only by offering some bizarre
persuasive definition of 'good-on-the-whole.' Yet, at least within Western
Culture, there is, generally speaking, a fairly stable conception of the
common good. If this last extra-ordinary kind of egoist denied this con-
ception and redefined 'good-on-the-whole' in some other fashion to make
it consonant with his theory,his critic might finally have to smile tolerantly
and grant him that this, in some "Alice-In-Wonderland world," is a "con-
sistently thinkable" view. If 'good' is so defined, then he cannot be re-

futed conclusively; but it can be held against his view that it does not fit
the usual use of moral words and that if he will use moral language as it
is usually used his view cannot be sustained.
The ethical egoist might reply that his opponent is here making only a
rather interesting sociological and philological observation, but that the
objection does not even tend to disprove his theory. To follow this re-
buttal through would be to launch into a discussion of the role of con-
ceptual analysis or linguistic analysis in ethics. This, of course, would
require a paper of some considerable length. Here I can only make a few
brief comments.
Whether we start out talking about common moral experience or about
the actual language of moral discourse, our subjectmatterin ethics remains
the moral assessments and moral arguments we actually make.8 Egoists, by
and large, have recognized this in practice. Hobbes had a different
rationale for accepting many of our common-sense moral appraisals but he
accepted them nonetheless. Gardner Williams (a latter-day egoist) does
much the same thing.9 They do not make actual moral recommendations
of an iconolastic or nihilistic kind. Finally we may, as Nietszche, wish to go
beyond moral good and evil altogether. Perhaps (if we do not put too much
weight on the 'ethical' in 'ethical egoism') that is what the "ethical egoist"
wishes to account for. He may want to point out that finally we can give
no further reason for being moral than that being moral is a good for us. If
another person does not find that being moral is a good for him, there is,
in the last analysis, nothing of an intellectual sort that we can say to him.
A good, finally, is not a good unless it is a good for the individual. We will
only take a moral point of view if taking a moral point of view meets our
long-range satisfaction. In that case, though the egoist'sremarks may be of
great interest and though they may be relevant to ethics, his remarks are
not 'moral remarks' or remarks about or explicative of moral remarks,
that is to say, they are not made from within the mode of moral reasoning
8 This has been well put by A. I. Melden, "Two Comments on Utilitarianism,"
PhilosophicalReview, October, 1951, Vol. LX, pp. 512-13. In the above broad sense
both Edwards and Mandelbaumagree about the subject matter of ethics, though the
formerexplicates actual moral discoursevia a Wittgenstein linguistic method and the
latter turns to actual moral experience with a modified phenomenologicalmethod.
See Paul Edwards, The Logic of Moral Discourse,pp. 19-21 and 77-81, and Maurice
Mandelbaum,The Phenomenologyof Moral Experience,Chapter1.
9 Note how Hobbes restates the traditional virtues of the natural law and common-
sense tradition while giving them a completely different basis. Thomas Hobbes,
Leviathan,ChaptersXIV, XV. See also GardnerWilliams, "Commentsand Criticism:
Universalistic Hedonism vs. Hedonic Individual Relativism," The Journal of Phio-
phy, Vol. LII, February 3, 1955, pp. 72-7, and most particularly GardnerWilliams,
"The Inevitability of Egoism in Ethics," Papers of the Michigan Academyof Science,
Arts and Letters,Vol. XXXVII; 1951, pp. 494-5.
510 PHILOSOPHYAND PHENOMENOLOGICAL
RESEARCH
nor are they metaethical explications of moral reasoning.10 His value

judgments may be maxims of prudence or of self love, but they are not,
literally speaking, moral judgments.
I have in my critique proceeded on the assumption that the ethical
egoist was sticking to the subject matter and was not asking the question,
or "alleged question," "Why be moral?" I have assumed in criticizing
ethical egoism that when the ethical egoist refers to his position as an
'ethical position' or speaks of 'moral standards' he uses these phrases in the
same way they are used in ordinary language. Within these limitations, I
have tried to show that a psychological egoism and an ethical egoism of
ends or means are either false or unintelligible. If the "ethical egoist"
replies that he is not using language as it is ordinarily used, I will reply
that he has then changed the subject. If he replies that he is not trying to
justify egoism in ethics but is asking, as a candid amoral egoist, "Why
should I be moral?" I will again say he has changed the subject. I do not
deny his right to do either and I do not regard the question "Why should I
be moral?" as self-contradictory or patently absurd; 11 but I would insist
that none of these moves by the "ethical egoist" will at all weaken my
argument that there is no good reason to accept the view of egoism in
ethics.12
KAI NIELSEN.
AMHERSTCOLLEGE.
10 For the conception of the mode of ethical reasoning relevant here see Stephen
Toulmin, An Examinationof the Place of Reamonin Ethics, pp. 102-4 and Chapters 9,
10, and 11.
11 I have argued for this as againstToulmin and Meldenelsewhere.See my "Is 'Why
Should I be Moral?'an Absurdity?" Australa8ianJournal of Philosophy,Vol. 36, No. 1
(1958), pp. 25-31.
12 For my remarks about the relation of philosophy to language I am, of course,
deeply indebted to LudwigWittgenstein'sPhilosophicalInvestigation8.For an elaborate
and brilliant defense of a "linguistic method in ethics" see Everett W. Hall, What is
Value?,Chapter7.

Egoism in Ethics

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Egoism in Ethics

Uploaded by

Copyright:

Available Formats

International Phenomenological Society

Philosophy and Phenomenological Research

Egoism in ethics may be divided into psychological egoism and ethical

However, in saying that the initial plausibility of all forms of ethical

3 The mother in D. H. Lawrence's Sons and Lovers is a fictional example of this

5 Sigmund Freud, ThreeCotribution4 to the Theoryof Sex, p. 77.

a recurrent theory as psychological egoism. Now, if my above argument is

preferences." Yet it is the ordinary sense of 'selfish preference' that causes

A purely ethical egoism divorced from a psychological egoism is also

sistently thinkable" view. If 'good' is so defined, then he cannot be re-

nor are they metaethical explications of moral reasoning.10 His value

You might also like