How Is Weakness of the Will Possible?

An agent's will is weak if he acts, and acts intentionally, counter to his own best
judgement; in such cases we sometimes say he lacks the willpower to do what he knows, or
at any rate believes, would, everything considered, be better. It will be convenient to call
actions of this kind incontinent actions, or to say that in doing them the agent acts
incontinently. In using this terminology I depart from tradition, at least in making the class
of incontinent actions larger than usual. But it is the larger class I want to discuss, and I
believe it includes all of the actions some philosophers have called incontinent, and some of
the actions many philosophers have called incontinent.
Let me explain how my conception of incontinence is more general than some
others. It is often made a condition of an incontinent action that it be performed despite the
agent's knowledge that another course of action is better. I count such actions incontinent,
but the puzzle I shall discuss depends only on the attitude or belief of the agent, so it would
restrict the field to no purpose to insist on knowledge. Knowledge also has an unneeded,
and hence unwanted, flavour of the cognitive; my subject concerns evaluative judgements,
whether they are analysed cognitively, prescriptively, or otherwise. So even the concept of
belief is perhaps too special, and I shall speak of what the agent judges or holds.
If a man holds some course of action to be the best one, everything considered, or
the right one, or the thing he ought to do, and yet does something else, he acts incontinently.
But I would also say he acts incontinently provided he holds some available course of
action to be better on the whole than the one he takes; or that, as between some other course
of action which he believes open to him
end p.21
and the action he performs, he judges that he ought to perform the other. In other
words, comparative judgements suffice for incontinence. We may now characterize an
action that reveals weakness of the will or incontinence:
D. In doing x an agent acts incontinently if and only if: (a) the agent does x intentionally;
(b) the agent believes there is an alternative action y open to him; and (c) the agent judges
that, all things considered, it would be better to do y than to do x.1
There seem to be incontinent actions in this sense. The difficulty is that their
existence challenges another doctrine that has an air of self-evidence: that, in so far as a
person acts intentionally he acts, as Aquinas puts it, in the light of some imagined good.
This view does not, as it stands, directly contradict the claim that there are incontinent
actions. But it is hard to deny that the considerations that recommend this view recommend
also a relativized version: in so far as a person acts intentionally he acts in the light of what
he imagines (judges) to be the better.
It will be useful to spell out this claim in the form of two principles. The first
expresses the natural assumption about the relation between wanting or desiring something,
and action. 'The primitive sign of
wanting is trying to get', says Anscombe in Intention.2 Hampshire comes closer to
exactly what I need when he writes, in Freedom of the Individual,3 that 'A wants to do X' is
equivalent to 'other things being equal, he would do X, if he could'. Here I take (possibly
contrary to Hampshire's intent) 'other things being equal' to mean, or anyway to allow, the
interpretation, 'provided there
end p.22
is not something he wants more'. Given this interpretation, Hampshire's principle
could perhaps be put:
P1. If an agent wants to do x more than he wants to do y and he believes himself
free to do either x or y, then he will intentionally do x if he does either x or y intentionally.
The second principle connects judgements of what it is better to do with motivation
or wanting:
P2. If an agent judges that it would be better to do x than to do y, then he wants to
do x more than he wants to do y.
P1 and P2 together obviously entail that if an agent judges that it would be better
for him to do x than to do y, and he believes himself to be free to do either x or y, then he
will intentionally do x if he does either x or y intentionally. This conclusion, I suggest,
appears to show that it is false that:
P3. There are incontinent actions.
Someone who is convinced that P1-P3 form an inconsistent triad, but who finds
only one or two of the principles really persuasive, will have no difficulty deciding what to
say. But for someone (like myself) to whom the principles expressed by P1-P3 seem self-
evident, the problem posed by the apparent contradiction is acute enough to be called a
paradox. I cannot agree with Lemmon when he writes, in an otherwise admirable article,
'Perhaps akrasia is one of the best examples of a pseudo-problem in philosophical literature:
in view of its existence, if you find it a problem you have already made a philosophical
mistake.'4 If your assumptions lead to a contradiction, no doubt you have made a mistake,
but since you can know you have made a mistake without knowing what the mistake is,
your problem may be real.
The attempted solutions with which I am familiar to the problem created by the
initial plausibility of P1-P3 assume that P1-P3 do really contradict one another. These
attempts naturally end by giving up one or another of the principles. I am not very happy
about P1-P3 as I have stated them: perhaps it is easy to doubt whether they are true in just
their present form (particularly P1 and P2). And reflecting on the ambiguities, or plurality
of uses, of
end p.23
various critical words or phrases ('judge better', 'want', 'intentional') it is not
surprising that philosophers have tried interpreting some key phrase as meaning one thing in
one principle and meaning something else in another. But I am convinced that no amount of
tinkering with P1-P3 will eliminate the underlying problem: the problem will survive new
wording, refinement, and elimination of ambiguity. I shall mention a few of the standard
moves, and try to discredit them, but endless ways of dealing with the problem will remain.
My basic strategy is therefore not that of trying to make an airtight case for P1-P3, perhaps
by working them into less exceptionable form. What I hope rather is to show that P1-P3 do
not contradict one another, and therefore we do not have to give up any of them. At the
same time I shall offer an explanation of why we are inclined to think P1-P3 lead to a
contradiction; for if I am right, a common and important mistake explains our confusion, a
mistake about the nature of practical reason.
I
Here are some of the ways in which philosophers have sought, or might seek, to
cope with the problem of incontinence as I have stated it.
The sins of the leopard—lust, gluttony, avarice, and wrath—are the least serious
sins for which we may be eternally damned, according to Dante. Dante has these sins,
which he calls the sins of incontinence, punished in the second, third, fourth, and fifth
circles of Hell. In a famous example, Dante describes the adulterous sin of Francesca da
Rimini and Paolo Malatesta. Commentators show their cleverness by pointing out that even
in telling her story Francesca reveals her weakness of character. Thus Charles Williams
says,
'Dante so manages the description, he so heightens the excuse, that the excuse reveals itself
as precisely the sin . . . the persistent parleying with the occasion of sin, the sweet prolonged
laziness of love . . . '5 Perhaps all this is true of Francesca, but it is not essential to
incontinence, for the 'weakness' may be momentary, not a character trait: when we speak of
'weakness' we may merely express, without explaining, the fact that the agent did what he
knew to be
end p.24
wrong. ('It was one page did it.') Aristotle even seems to imply that it is impossible
to be habitually incontinent, on the grounds that habitual action involves a principle in
accord with which one acts, while the incontinent man acts against his principle. I suppose,
then, that it is at least possible to perform isolated incontinent actions, and I shall discuss
incontinence as a habit or vice only as the vice is construed as the vice of often or habitually
performing incontinent actions.6
A man might hold it to be wrong, everything considered, for him to send a valentine
to Marjorie Morningstar. Yet he might send a valentine to Marjorie Eveningstar, and do it
intentionally, not knowing that Marjorie Eveningstar was identical with Marjorie
Morningstar. We might want to say he did something he held to be wrong, but it would be
misleading to say he intentionally did something he held to be wrong; and the case I
illustrate is certainly not an example of an incontinent action. We must not, I hope it is
clear, think that actions can be simply sorted into the incontinent and others. 'Incontinent',
like 'intentional', 'voluntary', and 'deliberate', characterizes actions only as conceived in one
way rather than another. In any serious analysis of the logical form of action sentences, such
words must be construed, I think, as non-truth-functional sentential operators: 'It was
incontinent of Francesca that . . . ' and 'It was intentional of the agent that . . . ' But for
present purposes it is enough to avoid the mistake of overlooking the intentionality of these
expressions.
Incontinence is often characterized in one of the following ways: the agent intends to do y,
which he holds to be the best course, or a better course than doing x; nevertheless he does x.
Or, the agent decides to do y, which he holds to be the best course, or a better course than
doing x, and yet he does x. Or, the agent chooses y as the result of deliberation, 7 and yet
does x, which he deems inferior
end p.25
to y. Each of these forms of behaviour is interesting, and given some provisos may
be characterized as inconsistent, weak, vacillating, or irrational. Any of them might be a
case of incontinence, as I have defined it. But as they stand, they are not necessarily cases
of incontinence because none of them entails that at the time he acts the agent holds that
another course of action would, all things considered, be better. And on the other hand, an
action can be incontinent without the agent's ever having decided, chosen, or intended to do
what he judges best.
Principle 2 states a mild form of internalism. It says that a judgement of value must
be reflected in wants (or desires or motives). This is not as strong as many forms of
internalism: it does not, for example, say anything at all about the connection between the
actual value of things (or the obligatory character of actions) and desires or motives. Nor
does it, so far as I can see, involve us in any doctrine about what evaluative judgements
mean. According to Hare, 'to draw attention to the close logical relations, on the one hand
between wanting and thinking good, and on the other between wanting and doing
something about getting what one wants, is to play into the hands of the prescriptivist; for
it's to provide yet another link between thinking good and action'. 8 I confess I do not see
how these 'close logical relations', which are given in one form by P1 and P2, support any
particular theory about the meaning of evaluative sentences or terms. A possible source of
confusion is revealed when Hare says ' . . . if moral judgements were not prescriptive, there
would be no problem about moral weakness; but there is a problem; therefore they are
prescriptive' (p. 68). The confusion is between making a judgement, and the content of the
judgement. It is P2 (or its ilk) that creates the problem, and P2 connects making a
judgement with wanting and hence, via P1, with acting. But prescriptivism is a doctrine
about the content or meaning of what is judged, and P2 says nothing about this. One could
hold, for example, that to say one course of action is better than another is just to say that it
will create more pleasure and yet maintain, as Mill perhaps did, that anyone who believes a
certain course of action will create more pleasure than another (necessarily) wants it more.
So I
end p.26
should like to deny that there is a simple connection between the problem of
incontinence as I have posed it and any particular ethical theory.
Perhaps the most common way of dealing with the problem of incontinence is to reject P2.
It seems obvious enough, after all, that we may think x better, yet want y more. P2 is even
easier to question if it is stated in the form: if an agent thinks he ought (or is obligated) to do
x, then he wants to do x; for of course we often don't want to do what we think we ought.
Hare, if I understand him, accounts for some cases of incontinence in such a way; so,
according to Santas, did Plato.9
It is easy to interpret P2 in a way that makes it false, but it is harder to believe there
is not a natural reading that makes it true. For against our tendency to agree that we often
believe we ought to do something and yet don't want to, there is also the opposite tendency
to say that if someone really (sincerely) believes he ought, then his belief must show itself
in his behaviour (and hence, of course, in his inclination to act, or his desire). When we
make a point of contrasting thinking we ought with wanting, this line continues, either we
are using the phrase 'thinking we ought' to mean something like 'thinking it is what is
required by the usual standards of the community' or we are restricting wanting to what is
attractive on a purely selfish or personal basis. Such ways of defending P2, though I find
them attractive, are hard to make conclusive without begging the present question. So I am
inclined, in order to move ahead, to point out that a problem about incontinence will occur
in some form as long as there is any word or phrase we can convincingly substitute for
'wants' in both P1 and P2.
Another common line to take with incontinence is to depict the akrates as overcome
by passion or unstrung by emotion. 'I know indeed what evil I intend to do. But stronger
than all my afterthoughts is my fury', rants Medea. Hare makes this the paradigm of all
cases of weakness of the will where we cannot simply separate moral judgement and desire,
and he adds that in such cases the agent is psychologically unable to do what he thinks he
ought (Freedom and Reason, p. 77). Hare quotes Euripides' Medea when she says ' . . . an
unknown compulsion bears me, all reluctant, down', and St. Paul when he writes, 'The good
which I want to do, I fail to
end p.27
do; but what I do is the wrong which is against my will; and if what I do is against
my will, clearly it is no longer I who am the agent . . . ' ( Romans 7.) This line leads to the
view that one never acts intentionally contrary to one's best judgement, and so denies P3;
there are no incontinent actions in the sense we have defined. 10
A related, but different, view is Aristotle's, that passion, lust, or pleasure distort
judgement and so prevent an agent from forming a full-fledged judgement that his action is
wrong. Though there is plenty of room for doubt as to precisely what Aristotle's view was,
it is safe to say that he tried to solve our problem by distinguishing two senses in which a
man may be said to know (or believe) that one thing is better than another; one sense makes
P2 true, while the other sense is needed in the definition of incontinence. The flavour of this
second sense is given by Aristotle's remark that the incontinent man has knowledge 'in the
sense in which having knowledge does not mean knowing but only talking, as a drunken
man may mutter the verses of Empedocles' (Nic. Eth., 1147b).
Perhaps it is evident that there is a considerable range of actions, similar to
incontinent actions in one respect or another, where we may speak of self-deception,
insincerity, mauvaise foi, hypocrisy, unconscious desires, motives and intentions, and so
on.11 There is in fact a very great temptation, in working on this subject, to play the amateur
psychologist. We are dying to say: remember the enormous variety of ways a man can
believe or hold something, or know it, or want something, or be afraid of it, or do
something. We can act as if we knew something, and yet profoundly doubt it; we can act at
the limit of our capacity and at the same time stand off like an observer and say to
ourselves, 'What an odd thing to do.' We can desire things and tell ourselves we hate them.
These half-states and contradictory states are common, and full of interest to the
philosopher. No doubt they explain, or at least point to a way of describing without
contradiction, many cases where we find ourselves talking of weakness of the will or of
incontinence. But we ourselves show a certain weakness
end p.28
as philosophers if we do not go on to ask: does every case of incontinence involve
one of the shadow-zones where we want both to apply, and to withhold, some mental
predicate? Does it never happen that I have an unclouded, unwavering judgement that my
action is not for the best, all things considered, and yet where the action I do perform has no
hint of compulsion or of the compulsive? There is no proving such actions exist; but it
seems to me absolutely certain that they do. And if this is so, no amount of attention to the
subtle borderline bits of behaviour will resolve the central problem. 12
Austin complains that in discussing the present topic, we are prone to ' . . . collapse
succumbing to temptation into losing control of ourselves . . . ' He elaborates:
Plato, I suppose, and after him Aristotle, fastened this confusion upon us, as bad in its day and
way as the later, grotesque, confusion of moral weakness with weakness of will. I am very
partial to ice cream, and a bombe is served divided into segments corresponding one to one with
persons at High Table: I am tempted to help myself to two segments and do, thus succumbing to
temptation and even conceivably (but why necessarily?) going against my principles. But do I
lose control of myself? Do I raven, do I snatch the morsels from the dish and wolf them down,
impervious to the consternation of my colleagues? Not a bit of it. We often succumb to
temptation with calm and even with finesse.13
We succumb to temptation with calm; there are also plenty of cases where we act
against our better judgement and which cannot be described as succumbing to temptation.
In the usual accounts of incontinence there are, it begins to appear, two quite
different themes that interweave and tend to get confused. One is, that desire distracts us
from the good, or forces us to the bad; the other is that incontinent action always favours the
beastly, selfish passion over the call of duty and morality. That these two themes can be
separated was emphasized by Plato both in the Protagoras and the Philebus when he
showed that the hedonist, on nothing but his own pleasure bent, could go against his own
best
end p.29
judgement as easily as anyone else. Mill makes the same point, though presumably
from a position more sympathetic to the hedonist: 'Men often, from infirmity of character,
make their election for the nearer good, though they know it to be the less valuable; and this
no less when the choice is between two bodily pleasures than when it is between bodily and
mental.' (Utilitarianism, Chap. 11.) Unfortunately, Mill goes on to spoil the effect of his
point by adding, 'They pursue sensual indulgences to the injury of health, though perfectly
aware that health is the greater good.'
As a first positive step in dealing with the problem of incontinence, I propose to
divorce that problem entirely from the moralist's concern that our sense of the
conventionally right may be lulled, dulled, or duped by a lively pleasure. I have just relaxed
in bed after a hard day when it occurs to me that I have not brushed my teeth. Concern for
my health bids me rise and brush; sensual indulgence suggests I forget my teeth for once. I
weigh the alternatives in the light of the reasons: on the one hand, my teeth are strong, and
at my age decay is slow. It won't matter much if I don't brush them. On the other hand, if I
get up, it will spoil my calm and may result in a bad night's sleep. Everything considered I
judge I would do better to stay in bed. Yet my feeling that I ought to brush my teeth is too
strong for me: wearily I leave my bed and brush my teeth. My act is clearly intentional,
although against my better judgement, and so is incontinent.
There are numerous occasions when immediate pleasure yields to principle,
politeness, or sense of duty and yet we judge (or know) that all things considered we should
opt for pleasure. In approaching the problem of incontinence it is a good idea to dwell on
the cases where morality simply doesn't enter the picture as one of the contestants for our
favour—or if it does, it is on the wrong side. Then we shall not succumb to the temptation
to reduce incontinence to such special cases as being overcome by the beast in us, or of
failing to heed the call of duty, or of succumbing to temptation. 14
end p.30
II
Under sceptical scrutiny, P1 and P2 appear vulnerable enough, and yet tinkering
with them yields no satisfactory account of how incontinence is possible. Part of the reason
at least lies in the fact that P1 and P2 derive their force from a very persuasive view of the
nature of intentional action and practical reasoning. When a person acts with an intention,
the following seems to be a true, if rough and incomplete, description of what goes on: he
sets a positive value on some state of affairs (an end, or the performance by himself of an
action satisfying certain conditions); he believes (or knows or perceives) that an
action, of a kind open to him to perform, will promote or produce or realize the valued state
of affairs; and so he acts (that is, he acts because of his value or desire and his belief).
Generalized and refined, this description has seemed to many philosophers, from Aristotle
on, to promise to give an analysis of what it is to act with an intention; to illuminate how we
explain an action by giving the reasons the agent had in acting; and to provide the beginning
of an account of practical reasoning, i.e. reasoning about what to do, reasoning that leads to
action.
In the simplest case, we imagine that the agent has a desire, for example, to know
the time. He realizes that by looking at his watch he will satisfy his desire; so he looks at his
watch. We can answer the question why he looked at his watch; we know the intention with
which he did it. Following Aristotle, the desire may be conceived as a principle of action,
and its natural propositional expression would here be something like 'It would be good for
me to know the time' or, even more stiffly, 'Any act of mine that results in my knowing the
time is desirable.' Such a principle Aristotle compares to the major premise in a syllogism.
The propositional expression of the agent's belief would in this case be, 'Looking at my
watch will result in my knowing the time': this corresponds to the minor premise.
Subsuming
end p.31
the case under the rule, the agent performs the desirable action: he looks at his
watch.
It seems that, given this desire and this belief, the agent is in a position to infer that
looking at his watch is desirable, and in fact the making of such an inference is something it
would be natural to describe as subsuming the case under the rule. But given the desire and
this belief, the conditions are also satisfied that lead to (and hence explain) an intentional
action, so Aristotle says that once a person has the desire and believes some action will
satisfy it, straightway he acts. Since there is no distinguishing the conditions under which
an agent is in a position to infer that an action he is free to perform is desirable from the
conditions under which he acts, Aristotle apparently identifies drawing the inference and
acting: he says, 'the conclusion is an action'. But of course this account of intentional action
and practical reason contradicts the assumption that there are incontinent actions.
As long as we keep the general outline of Aristotle's theory before us, I think we
cannot fail to realize that he can offer no satisfactory analysis of incontinent action. No
doubt he can explain why, in borderline cases, we are tempted both to say an agent acted
intentionally and that he knew better. But if we postulate a strong desire from which he
acted, then on the theory, we also attribute to the agent a strong judgement that the action is
desirable; and if we emphasize that the agent's ability to reason to the wrongness of his
action was weakened or distorted, to that extent we show that he did not fully appreciate
that what he was doing was undesirable.
It should not be supposed we can escape Aristotle's difficulty simply by giving up
the doctrine that having the reasons for action always results in action. We might allow, for
example, that a man can have a desire and believe an action will satisfy it, and yet fail to
act, and add that it is only if the desire and belief cause him to act that we can speak of an
intentional action.15 On such a modified version of Aristotle's theory (if it really is a
modification) we would still have to explain why in some cases the desire and belief caused
an action, while in other cases they merely led to the judgement that a course of action was
desirable.
The incontinent man believes it would be better on the whole to do something else,
but he has a reason for what he does, for his
end p.32
action is intentional. We must therefore be able to abstract from his behaviour and
state of mind a piece of practical reasoning the conclusion of which is, or would be if the
conclusion were drawn from the premises, that the action actually performed is desirable.
Aristotle tends to obscure this point by concentrating on cases where the incontinent man
behaves 'under the influence of rule and an opinion' (Nic. Eth., 1147b; cf. 1102b).
Aquinas is far clearer on this important point than Aristotle. He says:
He that has knowledge of the universal is hindered, because of a passion, from reasoning in the
light of that universal, so as to draw the conclusion; but he reasons in the light of another
universal proposition suggested by the inclination of the passion, and draws his conclusion
accordingly . . . Hence passion fetters the reason, and hinders it from thinking and concluding
under the first proposition; so that while passion lasts, the reason argues and concludes under the
second.16
An example, given by
Aquinas, shows the plight of the the side of lust
incontinent man: the side of reason
(M 1 ) No fornication is (M 2 ) Pleasure is to be
lawful pursued
(m 1 ) This is an act of
(m 2 ) This act is pleasant
fornication
(C 2 ) This act is to be
C 1 ) This act is not lawful
pursued