M O RA L

OBLIGATION
ESSATSAND LECTURES
BY

H. A. PR IC H AR D

O XFORD
A T T HE C L A R E N D O N P R E S S

1(11

I
C ON T E N T S
I.

DoES MORAL PHILOSOPHY REST ON A MISTAXT? (l9lz)

2. DUTv AND IGNoRANcE oF recr

3. THE MEANTNG Ot dya06v

(1932)

IB

rN THE ETHTCS OF ARIS-

rorr-E (tggS)

40

pRrNcIpLEs oF poLITIcAL oBLIGATIoN

4. cRnnN's

('ggs-z)

5. MoRAL oBLIGATIoN (tggZ)

54
B7

6. rHn oBJECT oF A DEsIRn (r94o)

t64

pRoMrsE (c. r94o)

7. THE oBLrcATroN To KEEr A

r69

8. nxcseNcrxc (rg4o)

rBo

9. THE TrME OF AN OBLIGATION

rBe

IO. THE PSYCHOLOGY OF WILLING

I84

rr.

AcrrNG,

wILLING,

12. 'oucHt' (tg+l)

INDEX

DEsIRING (,g+S)

I87
r99
20r

I) OES MORAL
PH ILOSOPH Y
ON A M IST AKE?I

RE S T

l)r.()BABLy to most students of Moral Philosophy there comes a

I lirne when they feel a vague sense of dissatisfaction with the
wlrolc subject. And the senseof dissatisfaction tends to grow rather
tlr:rrr to diminish. It is not so much that the positions, and still
nrorr: the arguments, of particular thinkers seem unconvincing,
llrorrgh this is true. It is rather that the aim of the subject becomes
irrcrr::rsingly obscure. 'What', it is asked, 'are we really going to
lr;rrn by Moral Philosophy?' 'What are books on Moral Philosophy
r r':rlly trying to show, and when their aim is clear, why are they so
rrrrr:onvincingand artificial?' And again: 'Why is it so difficult to
rrrlrstitute anything better?' Personally, I have been led by growirrg <lissatisfactionof this kind to wonder whether the reason may
rrrrl bc that the subject, at any rate as usually understood, consists
itr llrc attempt to answer an improper question. And in this article
I slr:rll venture to contend that the existence of the whole subject,
,rs rrsrrallyunderstood, rests on a mistake, and on a mistake parallel
to tlurt on which rests, as I think, the subject usually called the
' l'lrr:ory of Knowledge.
ll'wc reflect on our own mental history or on the history of the
rrrlrir:t:t,we feel no doubt about the nature of the demand which
oliginates the subject. Any one who, stimulated by education, has
r orrrr: to feel the force of the various obligations in life, at some
tirrrr: or other comes to feel the irksomeness of carrying them out,
irrrrl to recognize the sacrifice of interest involved; and, if thoughtlrrl, lr<:inevitably puts to himself the question:'Is there really a
r(':rsorr why I should act in the ways in which hitherto I have
tlrorrght I ought to act? May I not have been all the time under an
rllrrsionin so thinking? Should not I really be justified in simply
tr ying to have a good time?' Yet, like Glaucon, feeling that somelr,,w lrc ought after all to act in these ways, he asks for a proof that
tlris fi:cling is justified. In other words, he asks 'Wh2 should I do
ilrcsr:things?', and his and otherpeople's moral philosophizing is
,lr :rttcmpt to supply the answer, i.e. to supply by a process of
rr'll'r tion a proof of the truth of what he and they have prior to
I From Mi nd, v ol . x x i , no. B r,
J an. rgrz .

DO E S M O RAL P H IL OS O P H Y R E ST O N A MIS TA K E ?

reflection believed immediately or without proof. This frame of

mind seems to present a close parallel to the frame of mind which
originates the Theory of Knowledge. Just as the recognition that
the doing of our duty often vitally interferes with the satisfaction
of our inclinations leads us to wonder whether we really ought to
do what we usually call our duty, so the recognition that we and
others are liable to mistakes in knowledge generally leads us, as it
did Descartes, to wonder whether hitherto we may not have been
always mistaken. And just as we try to find a proof, based on the
general consideration of action and of human life, that we ought
to act in the ways usually called moral, so we, like Descartes, propose by a process of reflection on our thinking to find a test of
knowledge, i.e. a principle by applying which we can show that
a certain condition of mind was really knowledge, a condition
which ex h2pothesiexisted independently of the process of reflection.
Now, how has the moral question been answered? So far as I
can see, the answers all fall, and fall from the necessitiesof the
case, into one of two species. Eitherthey state that we ought to do
so and so, because, as we see when we fully apprehend the facts,
doing so will be for our good, i.e. really, as I would rather say, for
our advantage, or, better still, for our happiness; or they state that
we ought to do so and so, because something realized either in or
by the action is good. In other words, the reason 'why' is stated
in terms either of the agent's happiness or of the goodness of something involved in the action.
To see the prevalence of the former species of answer, we have
onl! to consider the history of Moral Philosophy. To take obvious
instances, Plato, Butler, Hutcheson, Paley, Mill, each in his own
way seeks at bottom to convince the individual that he ought to
act in so-called moral ways by showing that to do so will really be
for his happiness. Plato is perhaps the most significant instance,
because of all philosophers he is the one to whom we are least
willing to ascribe a mistake on such matters, and a mistake on his
part would be evidepce of the deep-rootedness of the tendency to
make it. To show that Plato really justifies morality by its profitableness,it is only necessaryto point out (r ) that the very formulation of the thesis to be met, viz. that justice is d.M6rpnv dya06v,
implies that any refutation must consist in showing that justice is
oil<etov&ya06v, i.e. rreally, as the context shows, one's own advantage, and (z) that the term )uorzeleiu supplies thg key not only
to the problem but also to its solution.

D OE S MOR A L P HI LO SO PHY REST O N A M I STAKE?

g
'l'hc tendency to justify acting on moral rules in this way is
rr:rtural.For if, as often happens, we put to ourselvesthe question
'Wlry should we do so and so?', we are satisfied by being convirrc:cd either that the doing so will lead to something which we
w:rrrt (e.g. that taking certain medicine will heal our disease), or
tlrat the doing so itself, as we see when we appreciate its nature, is
somcthing that we want or should like, e.g. playing golf. The
lirrrnulation of the question implies a state of unwillingness or
irrrlifference towards the action, and we are brought into a conditiorr of willingness by the answer. And this process seems to be
prccisely what we desire when we ask, e.g., 'Why should we keep
()rrr engagementsto our own loss?'; for it is just the fact thatthe
kr:qring of our engagements runs counter to the satisfaction of our
rlcsires which produced the question.
'I'he answer is, of course, not an answer, for it fails to convince
rrs that we ought to keep our engagements; even if successfulon its
rrwn lines, it only makes us want to keep them. And Kant was really
orrly pointing out this fact when he distinguished hypothetical and
crrtcgorical imperatives, even though he obscured the nature of the
lirr:t by wrongly describing his so-called 'hypothetical imperatives'
lrs imperatives. But if this answer be no answer, what other can be
oll'cred? Only, it seems, an answer which bases the obligation to
rlo something on the goodness
either of something to which the act
k:ads or of the act itself. Suppose, when wondering whether we
rr:itlly ought to act in the ways usually called moral, we are told
lls a means of resolving our doubt that those acts are right which
produce happiness.We at once ask: 'Whose happiness?'If we are
tolcl 'Our own happiness', then, though we shall lose our hesitation
lo act in these ways, we shall not recover our sense that we ought
to do so. But how can this result be avoided? Apparently, only by
f rr:ing told one of two things; either that anyone's happiness is a
t lrirrg good in itself, and that thereforewe ought to do whatever will
;rr<rduce it, or that working for happiness is itself good, and that
tlrr: intrinsic goodnessof such an action is the reason why we ought
to do it. The advantage of this appeal to the goodnessof something
corrsistsin the fact that it avoids reference to desire, and, instead,
rr:li:rs to something impersonal and objective. In this way it seems
possible to avoid the resolution of obligation into inclination. But
jrrst for this reason it is of the essenceof the answer, that to be
cllt:ttive it must neither include nor involve the view that the
.rPPlchension of the goodness of anything necessarily arouses the

DO E S M O RA L p H IL O S O p H y R ES T O N A MTS TA K E ?
4
desire for it. Otherwise the answer resolves itself into a form of the
former answer by substituting desire or inclination for the senseof
obligation, and in this way it loseswhat seemsits special advantage.
Now it seems to me that both forms of this answer break down.
though each for a different reason.
Consider the first form. It is what may be called Utilitarianism
in the generic sense,in which what is good is not limited to pleasure.
It takes its stand upon the distinction between something which
is not itself an action, but which can be produced by an action, and
the action which will produce it, and contends that if something
which is not an action is good, then we ought to undertake the
action which will, directly or indirectly, originate it.I
But this argument, if it is to restore the senseof obligation to act,
must presuppose an intermediate link, viz. the further thesis that
what is good ought to be.2 The necessity of this link is obvious. An
'ought,' if it is to be derived at all, can only be derived from
another 'ought'. Moreover, this link tacitly presupposes another,
viz. that the apprehension that something good which is not an
action ought to be involves just the feeling of imperativeness or
obligation which is to be aroused by the thought of the action
which will originate it. Otherwise the argument will not lead us to
feel the obligation to produce it by the action. And, surely, both
this link and its implication are false.3 The word 'ought' refers to
actions and to actions alone. The proper language is never 'So and
so ought to be', but'I ought to do so and so'. Even if we are sometimes moved to say that the world or something in it is not what it
ought to be, what we really mean is that God or some human being
has not made something what he ought to have made it. And it is
merely stating another side of this fact to urge that we can only feel
the imperativeness upon us of something which is in our power ;
for it is actions and actions alone which, directly at least, are in our
power.
Perhaps, however, the best way to see the failure of this view is
to see its failure to c-orrespond to our actual moral convictions.
Suppose we ask ourselves whether our sensethat we ought to pay
our debts or to tell the truth arises from our recognition that in
doing so we should be originating something good, e.g. material
I Cf. Dr. Rashdall'sTIuorTof GoodandEuil,vol. i, p. r3B.

2 Dr. Rashdall, if I understand him rightly, supplies this link (cf. ibid., pp. r35-6).
3 When we speak of anything, e.g, of some emotion or of some quality
of a human
being, as good, we never dream in our ordinary consciousnessofgoing on to say that
therefore it ought to be.

D OE S MOR A L P H I LO SO PHY REST O N A M I STAKE?

r:omfort in A or true belief in B, i.e. suppose we ask ourselves

wlrcther it is this aspect of the action which leads to our recognition
tlrrrt we ought to do it. We at once and without hesitation answer
'No'. Again, if we take as our illustration our sensethat we ought
to:rct justly as between two parties, we have, if possible,even less
lrr:sitationin giving a similar answer; for the balance of resulting
good may be, and often is, not on the side ofjustice.
At best it can only be maintained that there is this element of
trrrth in the Utilitarian view, that unlesswe recognized that-sometlrirrg which an act will originate is good, we should not recognize
tlurt wc ought to do the action. Unless we thought knowledge a
good thing, it may be urged, we should not think that we ought to
tr:l[ the truth; unless we thought pain a bad thing, we should not
tlrink the infliction of it, without special reason' wrong. But this
is not to imply that the badness of error is the reason why it is
\vrons to lie, or the badness of pain the reason why we ought not to
irrllict it without special cause.r
It is, I think, just because this form of the view is so plainly
irt variance with our moral consciousnessthat we are driven to
rrrlopt the other form of the view, viz. that the act is good in itself
;rnd that its intrinsic goodnessis the reason why it ought to be done.
tt is this form which has always made the most serious appeal; for
thc goodness of the act itself seems more closely related to the
ollligation to do it than that of its mere consequences or results,
ltnd therefore, if obligation is to be based on the goodness of sometlring, it would seem that this goodness should be that of the act
itsclf. Moreover, the view gains plausibility from the fact that
rnrlral actions are most conspicuously those to which the term
'intrinsically good' is applicable.
Nevertheless this view, though perhaps lesssuperficial, is equally
rrntcnable. For it leads to precisely the dilemma which faces everyorre who tries to solve the problem raised by Kant's theory of the
good will. To see this, we need only consider the nature of the acts
to which we apply the term 'intrinsically good'.
'l'here is, of course, no doubt,that we aPProve and even admire
cr:r'tain actions, and also that we should describe them as good, and
rrs good in themselves. But it is, I think, equally unquestionable
' It may be noted that if the badness of pain were the reason why we ought not to
irrlli<:tpain on another, it would equally be a reason why we ought not to inflict pain
,rrr otrroelves;yet, though we should allow the wanton infliction of pain on ourselves
to lrc foolish, we should not think ofdescribing it as wrong.

DO E S M O RA L P H IL OS O P H Y R ES T ON A MIS TA K E ?

that our approval and our use of the term 'good' is always in
respect of the motive and refers to actions which have been actually
done and of which wc think we know the motive. Further, the
actions of which we approve and which we should describe as
intrinsically good are of two and only two kinds. They are either
actions in which the agent did what he dtd because he thought he
ought to do it, or actions of which the motive was a desire prompted
by some good emotion, such as gratitude, affection, family feeling,
or public spirit, the most prominent of such desires in books on
Moral Philosophy bcing that ascribed to what is vaguely called
benevolence. For the sake ol'simplicity I omit the casc of actions
done partly from some such desire and partly from a senseof duty;
for evcn if all good actions are done from a combination of these
motives, the argument will not be affected. The dilemma is this. If
the motivc in respect of which we think an action good is the sense
of obligation, then so far from the sensethat we ought to do it being
derived from our apprehension of its goodness, our apprehension
of its goodncss will presuppose the sensethat we ought to do it. In
other words, in this case the recognition that the act is good will
plainly presupposethe recognition that the act is right, where as the
view under consideration is that the recognition of the goodnessof
the act giues rise to the recognition of its rightness. On the other
hand, if the motive in respect of which we think an action good is
some intrinsically good desire, such as the desire to help a friend,
the recognition of the goodness of the act will equally fail to give
rise to the senseof obligation to do it. For we cannot feel that we
oughf to do that the doing of which is ex hypothesfprompted solely
by the desire to do it.t
The fallacy underlying the view is that while to base the rightnessof an act upon its intrinsic goodnessimplies that the goodness
in question is that of the motive, in reality the rightness or wrongnessof an act has nothing to do with any question of motives at all.
For, as any instance will show, the rightness of an action concerns
an action not in the fulfpr senseof the term in which we include the
motive in the action, but in the narrower and commoner sensein
which we distinguish an action from its motive and mean by an
action merely the conscious origination of something, an origination which on different occasions or in different people may be
prompted by different motives. The question 'Ought I to pay my
' It is, I think, on this latter horn of the dilemma that Martineau'e view falls: cf.
Tlpes of Ethical Theory, patt ii, book i.

DOES MORAL

P H IL OSOPH Y

R EST ON A M ISTAKE?

lrills?' really means simply 'Ought I to bring about my tradesof what by my previous acts I explicitly or imrnen'spossession
plicitly promised them?' There is, and can be, no question of
whcther I ought to pay my debts from a particular motive. No
rkrubt we know that if we pay our bills we shall pay them with
;r motive, but in consideringwhether we ought to pay them we
irrcvitably think of the act in abstraction from the motive. Even
il'we knew what our motive would be if we did the act, we should
n<ltbe any nearer an answerto the question.
Moreover, if we eventually pay our bills from fear of the county
r:ourt,we shall still have done whatwe ought, even though we shall
not have done it.as we ought. The attempt to bring in the motive
i rrvolvesa mistakesimilar to that involved in supposingthat we can
will to will. To feel that I ought to pay my bills is to be moaed
paying them. But what I can be moved towards must
I.outards
;rlwaysbe an action and not an action in which I am moved in a
particular way, i.e. an action from a particular motive; otherwise
I should be moved towards being moved, which is impossible.Yet
the view under consideration involves this impossibility, for it
rcally resolvesthe sensethat I ought to do so and so, into the sense
that I,ought to be moved to do it in a particular way.I
So iar my contentions have been mainly negative, but they
lirrm, I think, a useful,if not a necessary,introduction to what I
taketo be the truth. This I will now endeavourto state,first formulirting what, as I think, is the real nature of our apprehensionor
itppreciationof moral obligations, and then applying the result
to elucidatethe questionof the existenceof Moral Philosophy.
The senseof obligation to do, or of the rightnessof, an action of
rr particular kind is absolutely underivative or immediate. The
rightnessof an action consistsin its being the origination of something of a certain kind I in a situation of a certain kind, a situation
r:onsistingin a certain relation B of the agent to. others or to his
own nature. To appreciateits rightnesstwo preliminaries may be
rrccessary.We may have to follow out the consequencesof the
proposedaction more fully than we have hitherto done, in order
to realize that in the action we should originate l. Thus we may
rrot appreciate the wrongnessof telling a certain story until we
rr::rlizethat we should thereby be hurting the feelingsof one of our
r I t is of course not denied here that an action done from a particular motive may
l"' good; it is only denied that the righhuss ofan action depends on its being done with
,r rr:rrticular motive.

DO E S M O R AL P H IL OS O P H Y R E ST ON A MIS TA K E ?

audience. Again, we may have to take into account the relation,B

involved in the situation, which we had hitherto failed to notice.
For instance, we may not appreciate the obligation to give X a
present, until we remember that he has done us an act of kindness.
But, given that by a process which is, of course, merely a process of
general and not of moral thinking we come to recognize that the
proposed act is one by which we shall originate A in a relation,B,
then wc appreciatc the obligation immediately or directly, the
appreciation being an activity of moral thinking. We recogni ze, for
instance, that this performance of a service to X, who has done us
a service, just in virtue of its being the performance of a service to
one who has rendered a service to the would-be agent, ought to be
done by us. This apprehension is immediate, in precisely the sense
in which a mathematical apprehension is immediate, e.g. the
apprehension that this three-sided figure, in virtue of its being
three-sided, must have three angles. Both apprehensions are immediate in the sense that in both insight into the nature of the
subject directly leads us to recognize its possessionofthe predicate;
and it is only stating this fact from the other side to say that in
both casesthe fact apprehended is self-evident.
The plausibility of the view that obligations are not self-evident
but need proof lies in the fact that an act which is referred to as an
obligation may be incompletely stated, what I have called the
preliminaries to appreciating the obligation being incomplete. If,
e.9., we refer to the act of repaying Xby a present merely as giving
X a present, it appears, and indeed is, necessary to give a reason.
In other words, wherever a moral act is regarded in this incomplete way the question 'Wh2 should I do it?' is perfectly legitimate. This fact suggests, but suggests wrongly, that even if the
nature of the act is completely stated, it is still necessary to give a
reason, or, in other words, to supp]y a proof.
The relations involved in obligations of various kinds are, of
course, very different. The relation in certain casesis a relation to
others due to a past act of theirs or ours. The obligation to repay
a benefit involves a relation due to. a past act of the benefactor.
The obligation to pay a bill involves a relation due to a past act of
ours in which we have either said or implied that we would make
a certain return for something which we have asked for and received. On the other hand, the obligation to speak the truth implies
no such definite act; it involves a relation consisting in the fact that
others are trusting us to speak the truth, a relation the apprehen-

I)I)I']S

MOI{N I, P H ILOS OP H Y

REST

ON

A MIS TA K E ?

ilrrrr ol wlri<:lrgives rise to the sense that communication of the

Ir rrtlr rr sorrrr:thingowing by us to them. Again, the obligation not
ll lrrrrI tlrr: l'cclingsof another involves no special relation of us to
llrirt .llrcr', i.e. no relation other than that involved in our both
lr irrp,rrrcrr,lrnd men in one and the same world. Moreover, it
rllnlr tlr;rt thc rclation involved in an obligation need not be a
lr.l,rtr'rnto irnothcr at all. Thus we should admit that there is an
llrlrli,rtiorr to overcome our natural timidity or greediness,and that
llrl,rrrrvolvcsno relations to others. Still there is a relation involved,
rl r rr rr' l i rti onto our own disposit ion.I t is sim ply becausewe can
rrrrrllrcr;rrrscothers cannot directly modify our disposition that it
to im pr ove it , and t hat it is not t heir s, or , at least ,
l r,,rrr l rrrsi rrcss
trl l l l rri l s to the sam e ext ent .
'l'lrc rrcgirtiveside of all this is, of course, that we do not come
rr
f rrl rl rrcri :ttc an obt igat ion by anar gum ent , i. e.by a pr ocessof nonl t.rr,rl tl ri rrki ng, a nd t hat , in par t icular , we do not do so by an
ril llil ril('ilt o{'which a premissis the ethical but not moral activity of
rrl )l l r| r i :rti rrgtheg oodnesseit her of t he act or of aconsequenceof t he
{rrt; i.(:. tlrat our senseof the rightnessof an act is not a conclusion
| | rrrrr .r rr nppreciation of the goodnesseither of it or of anything else.
It rvill probably be urged that on this view our various obligalrnrr,rlirrnr, like Aristotle's categories,an unrelated chaos in which
ll r,,rrnlrossiblcto acquiesce.For, according to it, the obligation to
I r'lr,ry ;r lrt'rrcfit,or to pay a debt, or to keep a promise, presupposes
rr lrr('vious act of anothel; whereas the obligation to speak the
Irrrtl r or n()t to ha r m anot her does not ; and, again, t he obligat ion
ln rcrnovc our timidity involves no relations to others at all. Yet,
ad hominemis at hand in the fact
rrf ,rrryr:rtL-,an effective argumentum
tlr,rt tlrt: virrious qualities which we recognize as good are equally
rrrrrr'l.rtccl
; c.g. courage, humility, and interest in knowledge. If, as
I'r ;rl.rirrfy thc case,dya1d.differ fi dya1d.,why should not obligations
lr;rr.rlly <lill'<:r
qua their obligatoriness?Moreover, if this were notso
tl rl r. torrl d i n th e end be only one obligat ion, which is palpably
| | r rt| ,u'y to fact.r
' ls'o ollrt:r objections may be anticipated: (r) that obligations cannot be selfr r trl rrt, rirrcc tnany actions regarded as obligations by some are not so regarded by
,,rlrr rr, ,rrrrl (..r)that if obligations are self-cvident, the problem of how we ought to act
lr il'' pr(s('rr((' of conflicting obligations is insoluble.
| ,r tl rr' l i l st I shoul d repl y :
(,r1 | lr;rt tlrc appreciation ofan obligation is, ofcourse, only possiblefor a developed
l r,,r,rl l x'i rrr.,:rrrrlthat di fferent degreesof dev el opmentare pos s i bl e.
1/,1 llr;rt thc lailure to recognize some particular obligation is usually due to the

IO

D O E S M O RA L P H IL OS O P H Y R ES T ON A MIS TA K E ?
Certain observations will help to make the view clearer.
In the first place, it may seem that the view, being-as it isavowedly put forward in opposition to the view that what is right
is derived from what is good, must itself involve the opposite of this,
viz. the K:rnti:rn position that what is good is based upon what is
right, i.e. that an act, if it be good, is good becauseit is right. But
this is not so. For, on thc view put forward, the rightnessof a right
action lies solcly in the origination in which the act consists,
whcreas the intrinsic pJoodness
of an action lies solely in its motive I
and this implies that a morally good action is morally good not
simply because it is a right action but because it is a right
action done becausc it is right, i.e. from a sense of obligation.
And this implication, it may be remarked incidentally, seems
plainly true.
In the second place, the view involves that when, or rather
so far as, we act from a senseof obligation, we have no purpose or
end. By a 'purpose' or 'end' we really mean something the existence of which we desire, and desire of the existence of which leads
us to act. Usually our purpose is something which the act will
originate, as when we turn round in order to look at a picture. But
it may be the action itself, i.e. the origination of something, as
when we hit a golf-ball into a hole or kill someone out of revenge.r
Now if by a purpose we mean something the existence of which we
desire and desire for which leads us to act, then plainly, so far as we
act from a sense of obligation, we have no purpose, consisting
either in the action or in anything which it will produce. This is so
obvious that it scarcely seems worth pointing out. But I do so for
two reasons. (r) If we fail to scrutinize the meaning of the terms
'end' and 'purpose', we are apt to assume uncritically that all
deliberate action, i.e. action proper, must have a purpose; we then
become puzzled both when we look for the purpose of an action
fact that, owing to a lack of thoughtfulness,what I have called the preliminaries to this
recognition are incomplete.
(c) That the view put forward is consistentwith the admission that, owing to a lack
of thoughtfulness,even tbc best men are blind to man)' of dteir obligations, and that in
the end our obligations are seen to be co-extcnsivewith almost the whole ofour life.
To the second objection I should reply that obligation admits of degrees,and that
where obligations conflict, the decision of what we ought to do turns not on the
question'Which of the alternative coursesof action will originate the greater good?'
but on the question 'Which is thcgreater obligation?'
t ft is no objection to urge that an
action cannot be its own purpose, since the purpose ofsomething cannot be the thing itself, For, speaking strictly, the purpose is not
the action'spurpose but our purpose, and there is no contradiction in holding that our
purpose in acting may be the action,

I)OIj,S MORAL

P H I L OSOPH Y

R EST ON A M ISTAKE?

II

rlrrrr: liom a senseof obligation' and also when we try to apply to

rrr, lr irn action the distinction of means and end, the truth all the
tirrrr: bcing that since there is no end, there is no means either.
(,r) 'l'lrc attempt to base the senseof obligation on the recognition
ll t l rr: goodnessof something is really an attempt to find a purpose
irr rr rnoral action in the shape of something good which' as good'
rvc want. And the expectation that the goodness of something
an obligation disappearsas soon as we ceaseto look for a
rrrrrlr:rlies
l' )lrIl )ose,
'lilrc thesis, however, that, so far as we act from a senseof obligaliorr, we have no purPose must not be misunderstood. It must not
lrc titken either to mean or to imply that so far as we so act we have
wt nntiue. No doubt in ordinary speech the words 'motive' and
';rur'1lose'are usually treated as correlatives, 'motive' standing for
tirr: clcsire which induces us to act' and 'purpose' standing for the
olrjr:t:t of this desire. But this is only because, when we are looking
lirr the motive of the action, say, of some crime, we are usually prerrrpposing that the act in question is prompted by a desire,and
rr,il by the senseof obligation. At bottom, however, we mean by a
rrrotivewhat moves us to act; a senseof obligation does sometimes
ilt()vc us to act; and in our ordinary consciousnesswe should not
lresitate to allow that the action we were considering might have
lrirrl as its motive a sense of obligation. Desire and the sense of
olrlig:rtion are co-ordinate forms or species of motive.
lrr the third place, if the view put forward be right, we must
rlrirr'ply distinguish morality and virtue as independent, though
rrlrrtcd, speciesof goodness,neither being an aspect of something
ol which the other is an aspect, nor again a form or species of the
otlrr:r, nor again something deducible from the other; and we must
.rt tlrc same time allow that it is possible to do the same act either
viltrrously or morally or in both ways at once. And surely this is
llrrr:. An act, to be virtuous, mustr as Aristotle saw, be done

rl,ilit or family affection or the like, we prevent ourselves from

l,r'irrg clominated by a feeling of terror, desiring to do so from a
'rr'nS(:oll shame at being terrified. The goodness of such an act is
B

12

DO E S M O RAL P H IL OS O P H Y R ES T ON A MIS TA K E ,?

different from the goodness of an act to which we apply the term

moral in the strict and narrow sense,viz. an act done from a sense
of obligation. Its goodness lies in the intrinsic goodness of the
emotion and of the consequent desire under which we act, the goodness of this motive being different from the goodness of the moral
motive proper, viz. the senseof duty or obligation. Nevertheless,
at any rate in certain cases,an act can be done either virtuously or
morally or in both ways at once. It is possible to repay a benefit
either from dcsirc to repay it, or from the feeling that we ought to
do so, or from both motives combined. A doctor may tend his
patients either from a desire arising out of interest in his patients or
in the excrciseof skill, or from a senseof duty, or from a desire and
a senseof duty combined. Further, although we recognize that in
each casc the act possesses
an intrinsic goodness,rve regard that
action as the best in which both motives are combined; in other
words, we regard as the really best man the man in whom virtue
and morality are united.
It rnay be objected that the distinction between the two kinds of
motive is untenable, on the ground that the desireto repay a benefit,
for example, is only the manifestation of that which manifests itself
as the senseof obligationto re pay whenever wc think of something in
the action which is other than the repayment and which we should
not like, such as thc loss or pain involved. Yet the distinction can,
I think, easily be shown to be tenable. For, in the analogous caseof
revenge, the desire to return the injury and thc sensethat we ought
not to do so, leading, as they do, in opposite directions, are plainly
distinct; and thc obviousness of the distinction here seems to
remove any difficulty in admitting the existence of a parallel distinction between the desire to return a bencfit and the sensc that
we ought to return it.r
Further, the view implies that an obligation can no more be
based on or derived from a virtue than a virtue can be dcrived
from an obligation, in which latter case a virtue would consist in
carrying out an obligation. And the irnplication is surely true and
r This sharp distinction of virtue and morality as co-ordinate and independent
forms of goodness will explain a fact whicli otherwise it is difficult to account for. If we
turn from books on Moral Philosophy to any vivid account of human life and action
such as we find in Shakespeare,nothing strikes us more than the comparative remoteness of the discussionsof Moral Philosophy from the facts of actual life. Is not this
largely because,while Moral Philosophy has, quite rightly, concentrated its attention
on the fact of obligation, in the case of many of those whom we admire most and whose
lives are ofthe greatestinterest, the senseofobligation, though it may be an important,
is not a dominating factor in their lives?

II(II:I'i

MOR A L

P H ILOS OP H Y

REST

ON

A MIS TA K E ?

t3

l rnl ' ,rt.urt. ' l ' rrkc the cas e of cour age. I t is unt r ue t o ur ge t hat ,
rltrr l r r,ur;rl{()is a virtue, we ought to act courageously. It is and
llr.,t l. urrtruc, because,as we seein the end, to feel an obligation
lrr ,rrI r,rrrragcouslywould involve a contradiction. For, as I have
rrrp,,rl lrr:lirrc,we can only feel an obligationto act; we cannot feel
lrfl "l)ligirtion to actfrom a certaindesire,in this case the desire to
|,r(lu('r' one's feelings of terror arising from the sense of shame
rvlrrrlr tlr<:yarouse. Moreover, if the senseof obligation to act in a
prl tir ular way leads to an action, the action will be an action done
liorrr :r scnseof obligation, and therefore not, if the above analysis
n| r,i rl rrcbe ri ght, an act of cour age.
'l'lrr: mistake of supposing that there can be an obligation to act
rlrrr,rltt:ously seems to arise from two causes. In the first place,
tlrlrr is often an obligation to do that which involves the conquerIrrp,or r:ontrolling of our fear in the doing of it, e.g. the obligation to
w,rlk :rlong the side of a precipice to fetch a doctor for a member of
rrrrr'lirrnily. Here the acting on the obligation is externally, though
nrrly r:xternally, the same as an act of courage proper. In the second
lrl'r((: there is an obligation to acquire courage, i.e. to do such
llrirrgs as will enable us afterwards to act courageously, and this
Irr.rybe mistaken for an obligation to act courageously. The same
r rrrrsirlcrationscan, of course, be applied, mutatismutandis,to the
l tl rcl vi rtues,
'l 'lrc fact, if it be a fact, that virtue is no basisfor morality will ex;rl,rirrwhat otherwise it is difficult to account for,viz. the extreme
rrrrsr: of dissatisfaction produced by a close reading of Aristotle's
/rtlrics.Why is the Ethics so disappointing? Not, I think, because it
r r.irlly answers two radically different questions as if they were one:
(r ) 'What is the happy life?', (z) 'What is the virtuous life?' It is,
r,rtlrcr, because Aristotle does not do what we as moral philor,,;rlrerswant him to do, viz. to convince us that we really ought to
rl, rvhat in our non-reflective consciousnesswe have hitherto belilvr:d we ought to do, or if not, to tell us what, if any, are the other
tlrirrgswhich we really ought to do, and to prove to us that he is
r ilglrt.Now, if what I have just been contending is true, a systematir :rccount of the virtuous character cannot possibly satisfy this
rlcrrrand.At best it can only make clear to us the details of one of
orrr obligations, viz. the obligation to make ourselves better men;
lrrrt the achievement of this does not help us to discover what we
to do in life as a whole, and why; to think that it did would
'rrsht
lrc to think that our only business in life was self-improvement.

r+

DOE S M O RA L pH T L OS O p H y R E ST O N A M I S TA K E ?

Hence it is not surprising that Aristotle's account of the good man

strikes us as almost wholly of academic value, with little relation
to our real demand, which is formulated in Plato's words: oi yd.p
nepi roA intruy$wos 6 Adyos,d),\,i zep) roi Swwo rpdnov Xpi tiv.
I am not, of course, niticiling Aristotle for failing to satisfy this
demand, except so far as here and there he leads us to think that
he intends to satisfy it. For my main contention is that the demand
cannot be satisfied, and cannot be satisfied becauseit is illegitimate.
Thus we are brought to the question: 'Is there really such a thing
as Moral Philosophy, and, if there is, in what sense?'
We should first consider the parallel 62ss-a5 it appears to beof the Theory of Knowledge. As I urged before, at some time or
other in the history of all of us, if we are thoughtful, the frequency
of our own and of others' mistakes is bound to lead to the reflection
that possibly we and others have always been mistaken in consequence of some radical defect of our faculties. In consequence,
certain things which previously we should have said without
hesitation that we knew, as e.g. that 4x7 : zB, become subject to
doubt; we become able only to say that we thought we knew these
things. We inevitably go on to look for some general procedure by
which we can ascertain that a given condition of mind is really one
of knowledge. And this involves the search for a criterion of knowledge, i.e. for a principle by applying which we can settle that a given
state of mind is really knowledge. The search for this criterion and
the application of it, when found, is what is called the Theory of
(.nowledge. The search implies that instead of its being the fact
that the knowledge that A is B is obtained directly by consideration of the nature of ,4 and B, the knowledge that A is .8, in the full
or complete sense,can only be obtained by first knowing that I is
,8, and then knowing that we knew it by applying a criterion, such
as.Descartes's principle that what we clearly and distinctly concerve ls true.
Now it is easy to show that the doubt wheth er A is B,based on this
speculative or general ground, could, if genuine, never be set at
rest. For if, in order really to know that A is .8, we must first know
that we knew it, then really, to know that we knew it, we must first
know that we knew that we knew it. But-what is more important
-it is also easy tb show that this doubt is not a genuine dorrbt but
rests on a confusion the exposure of which removes the doubt. For
when we sal we doubt whether our previous condition was one of
knowledge, what we mean,if we mean anything at all, is that we

I)I)I,:S MOR A L

P H ILOS OP H Y

NNST

ON

A MIS TA K E ?

15

rl,,rrlrlrvlrctlrorour previous beliefwas true,alseliefwhich we should

r rlllr,,,;,rs tlrt: thinking that A is B. For in order to doubt whether
,,rrr ;,rcvirrrs condition was one of knowledge, we have to think of
lt lr rt ,rr l<rrowledge
but as only belief, and our only question can be
' \\' ,r' ,l l ri s l r< :l i eftrue?' B ut as soo n as we seet hat we ar e t hinking of
rrr llrcvious condition as only one of belief, we see that what we
frlr lr)w rlorrbting is not what we first said we were doubting, viz.
rvlrrrlr.r :r prcvious condition of knowledge was really knowledge.
llr r,,,', t() rcmove the doubt, it is only necessaryto appreciate
llrr rr',rl lr:rture of our consciousnessin apprehending, e.g. that
'.r11,
and thereby see that it was no mere condition of bei
|
lt, r rrrti lrrrt a condition of knowing, and then to notice that in our
rlrr<'rrldoubt what we are really doubting is not whether this
11s1,,,r
|,r,r rrusr(:sswas really knowledge, but whether a consciousness
rfl rlrr ftlrcr kind, viz. a belief that 7 X 4 : zB, was true. We thereby
lr rlr,rt tlrough a doubt based on speculative grounds is possible,
ll rr rr,| ;r rloubt concerning what we believed the doubt concerned,
is impossible.
'rrl tlr.rt :r tloubt concerning this latter
I rr',,r csrrltsfollow. In the first place, if, as is usually the case,we
rrr,,rrrlr), tlrr: "fheory of Knowledge' the knowledge which supplies
l l rr' ,rrr:l vr:r' to the questi on' Is what we have hit her t o t hought
l 11,,ryl .rl r1r'
r' cal l yknow l edge?' , t her e is and can be no such t hing,
,rrrrltl rr' :;rr1> posi ti on
that there can is sim ply due t o a conf usion.
I lr r l r ;rrr bc no answer to an illegitimate question, excePt that the
rl rr,' ,tr,rri s i l l cgi ti mate. N everth elesst he quest ion is one which we
r rrlltlnu. t() I)ut until we realize the inevitable immediacy of knowlr r11,, ,,\rrrlit is positive knowledge that knowledge is immediate
!rrrI rr rtlrcr r:un be, nor needsto be, improved or vindicated by the
Ilr r lr.r Lrrowlcdgethat it was knowledge. This positive knowledge
ir |, ,rt r,'st the inevitable doubt, and, so far as by the 'Theory of
h rr,,rvlcrlrl:'is meant this knowledge, then even though this knowlr,l1r lrr tlrt: knowledge that there is no Theory of Knowledge in
tlrr l,,r rn('r s(:nsc,to that extent the Theory of Knowledge exists.
lrr tlrt sr:<:onclplace, suppose we come genuinely to doubt
,rl rr fl rr r, (' .11.,1 X 4 : eB ow i ng t o a genuine doubt whet her we
{ \ l f I r' l l rt i rr bcl i evi ng yesterday t hat 7 x4 : zB, a doubt which
'
i rr rr l.rr I ouly arise if we have lost our hold of, i.e. no longer
r r rrrrrrrlr.r, llr<:rezrlnature of our consciousnessof yesterday, and
,', tlrrrrl, il as consisting in believing. Plainly, the only remedy is
'l
r,, rl ,, tl rr' :;urn:rgai n. Or, to put t he m at t er gener ally, if we do
, ' ,,1r, l , rl ,rrrl l tw hether i t i s true t hat A is. B, as we once t hought ,

16

DO E S M O R AL PH IL O S O P H Y R E ST ON A MIS TA K E ?
the remedy lies not in any process of reflection but in such a reconsideration of the nature of A and B as leads to the knowledge that

Ais B.
With theseconsiderationsin mind, considerthe parallel which,
as it seemsto me, is presented-though with certain differencesby Moral Philosophy.The sensethat we ought to do certain things
arisesin our unreflectiveconsciousness,
being an activity of moral
thinking occasionedby the varioussituationsin which we find ourselves.At this stageour attitude to theseobligationsis one of unquestioning confidence. But inevitably the appreciation of the
degreeto which the execution of theseobligations is contrary to
our interestraisesthe doubt whether after all theseobligationsare
really obligatory, i.e. whether our sensethat we ought not to do
certain things is not illusion. We then want to have it prouedto us
that we ought to do so, i.e. to be convinced of this by a process
which, as an argument, is different in kind from our original and
unreflectiveappreciationof it. This demand is, as I have argued,
illegitimate.
Hence, in the first place, if, as is almost universally the case,by
Moral Philosophyis meant the knowledgewhich would satisfythis
demand, there is no such knowledge,and all attempts to attain it
are doomed to failure becausethey rest on a mistake,the mistake
of supposingthe possibility of proving what can only be apprehended directly by an act of moral thinking. Neverthelessthe
demand, though illegitimate, is inevitable until we have carried
the processof reflection far enough to realize the self-evidenceof
dur obligations,i.e. the immediacy of our apprehensionof them.
This realization of their self-evidence
is positiveknowledge,and so
far, and so far only, as the term Moral Philosophy is confined to
this knowledgeand to the knowledgeof the parallel immediacy of
the apprehensionofthe goodnessofthe variousvirtues and ofgood
dispositionsgenerally, is there such a thing as Moral Philosophy.
But since this knowledgemay allay doubts which often affect the
whole conduct ofJife, it is, though not extensive,important and
even vitally important.
In the second place, suppose we come genuinely to doubt
whether we ought. for example, to pay our debts, owing to a
genuine doubt whether our previous conviction that we ought to
do so is true, a doubt which can, in fact, only arise if we fail to
rememberthe real nature of what we now call our past conviction.
The only remedy lies in actually getting into a situation which

l ) ( ) 1 , ) SM O R A L

P HIL OSOPH Y

R EST ON A M ISTAKE?

r7

or r;rsiorrsthe obligation, or-if our imagination be strong enough

irr irnirgining ourselvesin that situation, and then letting our

tubligutionto originate I in that situation.