Grice to the Mill -- on freedom

by JLS
for the GC

From Rickaby, "Free will in four English philosophers" (1906): Hobbes, Locke, Mill, and Grice.


JOHN STUART MILL

Logic, Boo{ VI, Chap. II. Of Liberty and Necessity.
Examination of Sir William Hamilton s Philo
sophy, Chap. XXVI. On the Freedom of the
Will

"CORRECTLY conceived, the doctrine called
\^4 Philosophical Necessity is simply this: that,
given the motives which are present to an individual s
mind, and given, likewise, the character and disposi
tion of the individual, the manner in which he will
act may be unerringly inferred; that if we knew the
person thoroughly, and knew all the inducements
which are acting upon him, we could foretell his con
duct with as much certainty as we can predict any
physical event. This proposition 1 take to be a mere
interpretation of universal experience, a statement in
words of what every one is internally convinced of. No
one who believed that he knew thoroughly the cir
cumstances of any case, and the characters of the dif
ferent persons concerned, would hesitate to foretell
how all of them would act. Whatever degree of doubt
he may in fact feel arises from the uncertainty whether
he really knows the circumstances, or the character of
some one or other of the persons, with the degree of
accuracy required; but by no means from thinking
that if he did know these things, there could be any
uncertainty what the conduct will be."
Mill strives to rest his doctrine, which is one with
that of Hume, upon experience. But I observe that
the experience which he invokes is not any know
ledge of fact, but a belief about an unobserved con
tingency: it is not an experience of what is, but an
expectation of what would be in a certain issue which
never occurs. No one ever does know any person
thoroughly, nor the relative values of all the motives
affecting any person s conduct out upon a new field
of choice where he has never been tried before, where
he cannot proceed by force of habit, where he will
have to make up his mind afresh, the very situation
in which, if anywhere, free will must come into play.
Even a successful prediction in such a case would
prove nothing. The success might be due, three-
quarters to shrewdness and the remaining quarter to
luck, as when one has backed the winner of the Derby.
"Three-quarters to shrewdness," I say, for I admit
that a free volition may be predicted with probability.
1 deny only that it can be predicted with certainty even
under the fullest knowledge of antecedent conditions
of choice. Not with certainty, because the volition is
not essentially contained in those conditions. Against
this position Mill alleges "a mere interpretation of
universal experience," his interpretation, to wit, but
certainly not his experience. As I have shown against
Hume, the libertarian interpretation, properly guarded
and explained, suits all experienced facts of predic
tion as well as "the doctrine called Philosophical
Necessity." Nothing, then, is thereby proved on
cither side.
"The religious metaphysicians who have asserted
Lhc freedom of the will have always maintained it to
be consistent with divine foreknowledge of our actions;
and if with divine, then with any other foreknowledge."

As religious metaphysicians we speak of" the GOD of
the Theist and of the Christian ; a GOD who is numeri
cally One, who is Personal; the Author, Sustainer and
Finisher of all things, the Life of Law and Order, the
Moral Governor; One who is Supreme and Sole; like
Himself, unlike all things besides Himself, which are
but His creatures; distinct from, independent of them
all; One who is self-existing, absolutely infinite, who
has ever been and ever will be, to whom nothing is
past or future; who is all perfection, and the fullness and
archetype of every possible excellence, the Truth Itself,
Wisdom, Love, Justice, Holiness; One who is All-
powerful, All-knowing, Omnipresent, Incomprehen
sible."* I am not concerned with the correctness of
this representation: my sole purpose is to show that
they who believe it to be correct are not committed to
the inference that if the freedom of the will is con
sistent with the divine foreknowledge of our actions,
it must be consistent likewise with any other fore
knowledge. The foreknowledge ascribed to "the GOD
of the Theist and of the Christian" not standing on
a level with any other foreknowledge, Mill s argu
ment a part becomes inadmissible.

God is"One,who is self-existing, absolutely infinite,
who has ever been and ever will be, to whom nothing is
* Grammar of\fssfftf, p. 98.



1 68 FREE WILL

past or future." He is the perfed realisation of all that
can be, filling all bounds of being, filling all space and
time, yesterday and to-day and for ever the same,
stationary in the plentitude of being. Like His being,
His knowledge is measured by eternity; it all exists
together, it embraces all time. Whatever things come
to be in time, are to God eternally present. His vision
ranges from eternity over all things as they are under
His unvarying all-pervading gaze.

To us the past and the future, when we know them,
are present in their images or in their signs.* But to
God they are present in themselves, for they are in
Him as in their first principle. We are placed at the

* " If the future and the past are, I would know where they are.
And if I cannot yet compass that, still I know that, wherever they
are, they are not there future or past but present. For if there also
they are future, they are not yet there; if there they are past, they
are no longer there. Wherever, therefore, they are, and whatever they
are, they are not save in the present. When the past is related truly,
it is not the past things themselves that are produced from memory,
but words formed from the images of them, like footprints which
in passing by they have impressed on the mind through the senses.
My boyhood for instance, which is no more, is in the time past,
which is no more; but when I con over and tell my impression of
it, I am looking at an objecl in the present time, because the im
pression is still in my memory. . . When the future is said to be
seen, it is not the things themselves which as yet are not, or which
are future, it is their causes or signs perchance, that are seen, which
signs already are. . . It is now plain and clear that neither the future
nor the past is. Nor is it properly said, There are three tenses, the
present, the past, and the future; but perhaps it might properly be
said: There are three tenses, the present of things past, the present
of things present, and the present of things future. For these are
three certain realities in the mind, and elsewhere I sec them not;
the present of things past, which is memory; the present of things
present, which is intuition; and the present of things future, which
is expectation." St Augustine s Confessions, xi, 18, 20.



JOHN STUART MILL 169

circumference of the circle of which He is the centre.
The instant in which we are is one now out of many:
from the divine now all nows radiate, and it is equiva
lent to them all. Thus to God there is no foreknow
ledge or afterknowledge, but simply knowledge of the
present. This knowledge, as applied to adual creation,
receives in theology the name of the " science of vision."
By it God sees, He sees in the a6l itself, He does not
calculate from antecedents, all that He Himself is
freely about to do, or rather is doing, in the way of
creating, working miracles and the like, as also all the
effects that will proceed from natural causes, whether
from the necessary determination of their natures, or
through the use made of them by free agents. God,
looking at a creature, sees its history all at once before
Him, albeit that, to the creature, the facls are evolved
successively. The generation that shall be alive thirty-
five years hence will behold what the ruler of France
at that time does.* They will not calculate his actions
from the motives, they will watch them being done.
Thirty-five years hence is present in the now of God.
He is a spectator of what is to go on then.

This is marvellous doctrine. If it were not marvel
lous, it would hardly be likely to be true of Him whose
name is called Wonderful. But it is not on the mar-
vellousness, nor even on the truth of the doclrine that
I here wish to insist, but on the bare facl that this
is the doctrine of those "religious metaphysicians"
who assert the freedom of the will and maintain it to
be consistent with the divine foreknowledgeofourcon-
* Written about the year 1872.



i?o FREE WILL

duct. Such eternal foreknowledge is a thing without
parallel in the human mind. It gives, therefore, no
ground for the inference set up by Mill.

Ill

" It is not the doctrine that our volitions and actions
are invariable consequents of our antecedent states of
mind, that is either contradicted by our consciousness
or felt to be degrading. But the doctrine of causation,
when considered as obtaining between our volitions
and their antecedents, is almost universally conceived
as involving more than this. Many do not believe, and
very few practically feel, that there is nothing in causa
tion but in variable, certain and unconditional sequence.
There are few to whom mere constancy of succession
appears a sufficiently stringent bond of union for so
peculiar relation as that of cause and effect. Even if the
reason repudiates, the imagination retains the feeling
of some more intimate connexion, of some peculiar tie
or mysterious constraint exercised by the antecedent
over the consequent. Now this it is which, considered
as applying to the human will, conflicts with our con
sciousness and revolts our feelings. We are certain that,
in the case of our volitions, there is not this myster ious
constraint. We know that we are not compelled, as by a
magical spell, to obey any particular motive. W r e feel
that if we wish to prove that we have the power of
resisting the motive, we could do so (that wish being,
it needs scarcely be observed, a new antecedent); and
it would be humiliating to our pride and paralysing
to our desire of excellence if we thought otherwise. But
neither is any such mysterious compulsion now sup
posed, by the best philosophical authorities, to be exer
cised by any cause over its effect. Those who think
that causes draw their effects after them by a mystical



JOHN STUART MILL 171

tie, are right in believing that the relation between
volitions and their antecedents is of another nature.
But they should go further and admit that this is also
true of all other effects and of their antecedents. If
such a tie is considered to be involved in the word
necessity, the doctrine is not true of human actions;
but neither is it then true of inanimate objects. It
would be more correct to say that matter is not bound
by necessity than that mind is so."

This language is, of course, no more than an echo
of Hume (nn. 6, 7). But because it is striking and
clear, and is caught up with approval by men of our
time, and even by boys, it had better be listened to
attentively and judged for what it is worth. Mill
teaches that volition does not differ from mechanical
action, so far as the invariable and unconditional se
quence of consequents upon antecedents is concerned.
Let the simultaneous facts, A, B, C, D, and others,
be followed by the fact Z. Then, says Mill, whatever
be the character of the facts, whether mental or phy
sical, it is certain that wherever A, B, C and the rest
go before, without addition or diminution, there Z
will come after. Experience may show that Z occurs
whether A precedes or not. Therefore A may be left
out of the account, as likewise may other antecedents,
as B, C, D, for the same reason. But it will be found
that some antecedents, such as F and G, cannot be
omitted without the result Z failing to appear. These
antecedents must be retained. Again, the insertion of
some new antecedents, as P, Q, R, may be found to
prevent the appearance of Z, even though F and G,



172 FREE WILL

and all others whose presence is indispensable, are
duly there. The omission of these obstructive ante
cedents must be bargained for. Let the indispensable
antecedents, F, G, etc., be summed under the general
expression E, and the impeding antecedents, P, Q, R,
etc., under the general expression E. Then the expres
sion E - E will stand for what Mill calls the cause of Z,
Z being any fact either of mind or of matter.

Let us take an illustration from each department.
And first of matter. A smith takes a piece of iron,
heats it red-hot in the forge, and beats it flat on the
anvil. The iron becoming flat is a fact or phenomenon
of matter. The antecedents to it are the smith s hav
ing got up that morning, having had his breakfast,
having work to do, having put the iron in the fire,
having hammered it, there are these and other ante
cedents too numerous to mention. The result is that
the iron gets flattened out. Any similar iron would
get flattened out in similar circumstances. Even a varia
tion of circumstances, up to a certain point, is compa
tible with the attainment of the result. That smith,
we will suppose, said his morning prayers. But iron
will yield to beating, whether the hands that strike it
have been previously clasped in prayer or not. GOD
rains upon the just and the unjust. On the other hand,
if the smith is stricken blind, his blow is likely to
fail, and the metal will go unflattened. Thus some
conditions are requisite to the effect, and some are
superfluous. Further, there are conditions of which
the absence is positively required. The hot bar must
not be cooled in water, else the beating will make no



JOHN STUART MILL 173

impression. When all the indispensable conditions are
there, and the preventive conditions are all absent, the
result, the flattening of the iron, will be brought to pass,
infallibly^ M ill would say : I should add, and necessarily.
Let us pass to a phenomenon of will. A man has
gained an importantsuccess,somethingthat he imagines
will fix his name in history: he has vindicated his
country s honour in the field, or amended her consti
tution at home, or he has come forward in the ranks
of her poets, her artists or her men of science: and
as he thinks of his achievement, his heart is lifted up
within him, as was the heart of Lucifer of old, taking
the glory to himself away from GOD. If the person
deliberately consents to this movement of vainglory,
he commits a sin: so all moralists who recognise the
rights of the Creator agree in teaching. If we are to
believe Mill, the guilty consent there follows upon
the temptation with a sequence as indefectible as the
flattening of a hot iron consequent upon percussion.
When a smith hammers a bar that has been properly
heated, and when there is no interference, natural or
supernatural, with the operation, it is incredible to
Mill, as it is to every reasonable man, that the shape
of the bar should remain unchanged. Suppose now
two persons are placed together in the situation of
trial which I have described. Their antecedent dis
positions, their present motives, arising as well from
nature as from grace, are essentially alike in number
and in kind. In that case it is simply incredible to
Mill that one man should sin and the other remain
innocent. Crimes, he thinks, are ruled by the same



i 7 4 FREE WILL

laws as landslips. One cliff will not stand in the exacl:
situation in which a similar cliff has fallen: neither
will Abel ever do right, if placed, with Cain s charac
ter, in an occasion similar to that in which Cain has
done wrong.

Our author indeed says: "We know that we are
not compelled, as by a magical spell, to obey any par
ticular motive. We feel that if we wished to prove
that we have the power of resisting the motive we
could do so." That wish, he adds with emphasis, would
be "a new antecedent." Just as well may it be said
that a cliff is not compelled, by any magical spell or
natural necessity, to give way under any particular
mining operation. If, as you cut away the rock, you
judiciously replace it with iron pillars, they will bear
up the superincumbent mass as it stood before. Those
pillars are new antecedents. Without them, or some
support like them, the rock, being undermined, surely
will fall. With them, if they are sufficient, it as surely
will stand. So, on Mill s showing, a man in tempta
tion surely will sin, unless it occurs to him that it
would be a fine thing to show his power of resistance.
Without that, or some deterring thought of that sort,
his offence is calculable, with mathematical precision,
from the occasion given him. But supply him with
motive sufficient, or if you like to speak theologically,
with grace sufficient, to keep him out of sin, and
there is no more danger of his yielding to temptation
than there is of his sinking through a stone pavement.

We have here a system of necessarianism, rigid as that
of any Calvinist divine. The recognition of anything



JOHN STUART MILL 175

that possibly might be other than what actually is cannot
stand with the doctrine of philosophical necessity, taught
nowadays by "the best philosophical authorities," as
Mill complacently styles himself and friends. The se
quence of antecedent and consequent in this system
is so close, so invariable, so uniform, as to leave no
room anywhere for edging in a might be, A bridge
has given way with a train upon it. It might not be
that the train should not fall into the river. It might
not be that the bridge, constructed as it was, should
not give way under that pressure. It might not be that
the railway officials, with their individual characters and
incentives to action, should have had the forethought
and energy to prevent the train from going upon the
bridge. It might not be that the engineer of the bridge
should have constructed it in any other way. It might
not be that anything which has happened should have
happened otherwise. Everywhere, event follows event
with rigid calculable precision, till we come to the pri
meval arrangement, the original collocation of mate
rials in the universe. That, one is tempted to say,
might have been arranged quite differently. But here
those self-styled "best philosophical authorities" de
clare human knowledge to stop short. Nothing, they
tell us, can be known as to how the first position of
things came about. Then it cannot be known that
things might have been arranged in the beginning in
any other fashion than as they actually were arranged.
Consequently, so far as we know, all that happens is
inevitable; what happens not, is impossible; andnothing
might have been, or might be, except what has been,



1 76 FREE WILL

is, or shall be. This is what the doctrine of philosophi
cal necessity comes to, Hobbism, pure and simple.

A "mystical tie," indeed, would that relation be, of
which there were no terms! The doctrine that "there
is nothing in causation but invariable, certain and un
conditional sequence," abolishes the terms of the re
lation of cause and effect, and cuts the relation afloat
to go by itself; this event before, and that event after,
no permanent being anywhere. If there is nothing of
permanent being in ourselves, and nothing permanent
in nature, on what ground do we assert the perma
nence of any law of nature? Why must the future re
semble the past, if nothing of the past stands over into
the future? No wonder if "many men do not believe,
and very few practically feel," that there is nothing
in the universe but a ghostly procession of phantoms
going before and phantoms coming behind. No wonder
if many men persist in looking for substantial realities,
and for ties, "mystical" or otherwise, so long as
they are real, that is to say, "real relations" between
cause and effect. We divide these substantial realities
into persons and things, persons habitually conscious of
self, things totally unconscious: dumb animals, who need
not here be considered, come in between. A thing es
sentially acts upon whatever comes within the range
of its action, as the earth on the moon, the sun on the
planets, every particle of matter upon every other par
ticle to which its power extends. The effect of such
action is some determination to motion. This action
of things is called transient, because the term of action
lies without the agent. Therefore are things called



JOHN STUART MILL 177

inert ^ because they do not aft within or upon them
selves, as it were setting themselves in motion. A
pefsotijOn the other hand, the only person I here speak
of is thinking man, as such, is impressed and acted
upon by objects without entering into his ken, and to
this impression there is a responsive action from
within. This action is immanent, for it remains within
the agent. This is the act of perceiving and liking, or
disliking, and in its first stage this action is necessary,
being determined, as determinists truly say, by envi
ronment and character. It is only in a further stage,
when the ego consciously awakes to judge of this spon
taneous and necessary like or dislike, that the exercise
of free will begins.

Libertarians have this abiding dissatisfaction with
Hume and Mill and the modern determinist school,
that, as men blinded by physics to everything above
the physical and material order, they ignore a vital
difference between beings conscious of the ego and beings
totally unconscious^ between persons in fact and things.
Still, dissatisfied as we are, we are not surprised: we
remember that we are dealing with men who have shut
out from their philosophical purview all such con
cepts as that of substantial, permanent Being and Per
sonality (oufft o, u7ro<TTo<rit;), yea even of Body and Soul,
and exercise their speculation solely upon transient

States of consciousness (ytrtaiQ, a taOijaii;, iravra pit).

Such exclusiveness leaves no place for free will, nor for
much else that is valuable in human nature: nay,
c nature itself loses all persistency and is carried away
in the stream of the definitely and determinately <be-

12



1 78 FREE WILL

coming. On all which dissolving views see Plato,
The<etetus i 796-1 Sjb.

IV

" I am inclined to think that . . . error . . . would
be prevented by forbearing to employ, for the expres
sion of the simple fact of causation, so extremely in
appropriate a term as necessity. That word, in its other
acceptations, involves much more than mere unifor
mity of sequence; it implies irresistibleness. Applied
to the will, it only means that the given cause will be
followed by the effect, subject to all possibilities of
counteraction by other causes; but in common use it
stands for the operation of those causes exclusively,
which are supposed too powerful to be counteracted
at all. When we say that all human actions take place
of necessity, we only mean that they will certainly
happen if nothing prevents: when we say that dying
of want, to those who cannot get food, is a necessity,
we mean that it will certainly happen whatever may
be done to prevent it. The application of the same
term to the agencies on which human actions depend,
as is used to express those agencies of nature which
are really uncontrollable, cannot fail, when habitual, to
create a feeling of uncontrollableness in the former
also. This, however, is a mere illusion. There are
physical sequences which we call necessary, as death
for want of food or air; there are others which are not
said to be necessary, as death from poison, which an
antidote, or the use of the stomach-pump, will some
times avert. It is apt to be forgotten by people s feel
ings, even if remembered by their understandings,
that human actions are in this last predicament; they
are never (except in some cases of mania) ruled by any
one motive with such absolute sway that there is no
room for the influence of any other. The causes,



JOHN STUART MILL 179

therefore, on which action depends are never uncon
trollable; and any given effect is only necessary, pro
vided that the causes tending to produce it are not
controlled. That whatever happens, could not have
happened otherwise unless something had taken place
which was capable of preventing it, no one surely
needs hesitate to admit."

This is a distinct advance upon Hume, who thought
(n. 7) that necessity added nothing to mere uniformity
of sequence. Mill recognises that it adds an element
of what we may call uncounteraftableness. No doubt Mdl
is right. Only Mill s position is none the better for this
correction of his predecessor; for in Mill s philosophy,
as in Hume s, whatever is actually uncounterafled\s prac
tically and in the concrete uncounteraftable; and there
fore to happen and to happen of necessity are one.

At Minster Lovel on the Windrush, some fifteen
miles west of Oxford, may be seen what remains ot
the house of Sir William Lovel, the trusted minister
of Richard III. In the next reign Sir William took part
in a rising against Henry VII, was defeated near Stoke,
and never heard of again. Some said that he was
drowned in the Trent, but others that he found his
way back to his Oxfordshire home, and ensconced him
self in one of those hiding-places which in those tumul
tuous days were an indispensable adjunct to every great
mansion. One old housekeeper knew his secret, and
she suddenly died. A skeleton, supposed to be Sir
William s, was found in the hiding-hole in 1708. We
may imagine the unhappy plight of the refugee lord
and master of that house. He hears the village clock



i8o FREE WILL

striking his usual dinner-hour: it does not call him to
eat. Twenty-four hours pass away to the solitary
prisoner, and forty-eight, and how many more? Men
keep aloof from him, and he cannot go to them: nor
do the angels come and minister to him. Nothing is
left for the man in that situation but death: inevitably,
irresistibly, necessarily, he must die. He may weep or
sing, sit or stand or lie down, but he must die. The
sun may shine or the rain fall, there may be feasting
or mourning in the house, his acquaintance may love
him or love him not: happen what will, if he remain
in that situation, he must die. And so he did die, and
there was no help for him.

Sir William died necessarily, as the case stood. We
can readily conceive how it might have stood other
wise, how he might have been discovered in time and
had food brought him by some faithful domestic. Let
us pass to a case of volition, and clothe the volition in
those circumstances which best make for freedom, if
volition ever is free. I speak of what I know and where
I have experience, in contradiction to blind prejudice
and lack of experience, when I say that the most per
fectly free volition possible is the choice of a state of
life, made according to St Ignatius s "method of elec
tion" in the Spiritual Exercises. But the particular in
stance chosen matters not. If any one will not take
mine, let him pick another for himself, let us say
Wellington s resolution to give battle to Marmont at
Salamanca. I follow up the instance which I have taken.
It is all-important that the "exercitant s" election be
his own. The director of the exercises is warned on



JOHN STUART MILL 181

no account to express a preference: nay, so far as may
be, he is not to feel a preference of one state over an
other for the exercitant s choice; he is not to advise
this choice or that, much less to dictate. He is to allow
the Creator to work alone upon the soul that He has
created. Days of careful thought are bestowed on the
choice. Mere emotion is discounted; prayerful reason
ing must decide. The decision is made, and in this case
we will suppose it to be, not to become a priest or
religious, but to go into the army. Wrong or right, it
is a thoroughly free election, the exercitant s own
choice. Now I say, considered in the concrete and
under the circumstances in which it is actually made,
that choice is every whit as uncounteractable, in Mill s
philosophy, and quite as necessary, as the refugee s
death in his hiding-hole. Only by violating your hypo
thesis, and bringing in discovery where discovery was
none, can you save Sir William s life. Only by altering
the exercitant s character, making him antecedently
more of a churchman by disposition than he actually
was, or by striking him with an alarm that in fact he
felt not, or kindling in him an enthusiasm that in his
breast did not burn, could you, on Mill s showing,
bring that exercitant to choose to be a priest. As
things stood, Mill would say, any choice of the
priesthood in him was quite out of the question and
impossible. Nay, taking a wider view of both positions,
we must avow that it was not in the nature of things,
as they lay from the beginning, for Sir William to be
discovered and saved; nor was it part of the existent
order of nature (and there can be only one order of



182 FREE WILL

nature) for that exercitant to have approached his elec
tion in other dispositions or under any other play of
motive. Mill would have allowed, I think, the neces
sity of Sir William s death. Most men would allow it, I
should allow it myself. No one who holds by Mill can
draw any distinction subtle and potent enough to dis
allow the similar necessity of that exercitant s choice.

There is a children s story of a certain Dutch ship,
which encountered a great storm at sea, whereupon the
sailors chose one of their number to tie all the rest
fast to the mast and spars. And so that one did. Then
he fastened himself up, in such a way that, when the
storm was over, he might loose first himself and then
his comrades. But the ship happening to give a great
lurch, he was turned head over heels, and hung un
able to release himself. Thus the whole crew were put
to drift at the mercy of the weather. Mill depidls man
kind in the plight of these unfortunate Dutchmen.
Any man might act otherwise than he does, if he could
get fresh motives, which would be forthcoming if any
one else could give them; but every man is tied up in
invariable and unconditional sequences like his fellow-
man: thus the world drifts underthe breath of necessity.

This sad consequence results from a too unqualified
admission of the principle that "whatever happens,
could not have happened otherwise unless something
had taken place which was capable of preventing it."
Man, in certain cases, could have elicited the mere
inward, deliberate acl: of his will otherwise than as he
actually has elicited it, and that apart from anything
else taking place, other than what has actually taken



JOHN STUART MILL 183

place, antecedently to his willing. Man is the one un
bound sailor in the ship of the physical universe.

V

"Though the doctrine of necessity, as stated by
most who hold it, is very remote from fatalism, it is
probable that most necessarians are fatalists, more or
less, in their feelings. A fatalist believes, or half be
lieves, for nobody is a consistent fatalist, not only
that whatever is about to happen will be the infallible
result of the causes which produce it, which is the
true necessarian doctrine, but, moreover, that there is
no use in struggling against it; that it will happen how
ever we may strive to prevent it. Now, a necessarian,
believing that our actions follow from our characters,
and that our characters follow from our organization,
our education and our circumstances, is apt to be, with
more or less of consciousness on his part, a fatalist as
to his own actions, and to believe that his nature is
such, or that his education and circumstances have so
moulded his character, that nothing can now prevent
him from feeling and acting in a particular way, or at
least that no effort of his own can hinder it. In the
words of the sect which in our own day has most per-
severingly inculcated and most perversely misunder
stood this great doctrine, his character is formed for
him, and not by him; therefore his wishing that it had
been formed differently is of no use; he has no power
to alter it. But this is a grand error. He has, to a cer
tain extent, a power to alter his own character. Its being,
in the ultimate resort, formed for him, is not incon
sistent with its being in part formed by him as one of
the intermediate agents. . . We are exactly as capable of
making our own character, if we will, as others are
of making it for us. Yes, answers the Owenite, but



i8 4 FREE WILL

these words, "if we will," surrender the whole point,
since the will to alter own character is given us, not
by any efforts of ours, but by circumstances which we
cannot help; it comes to us either from external causes,
or not at all. Most true: if the Owenite stops here, he
is in a position from which nothing can expel him.
Our character is formed by us as well as for us;
but the wish which induces us to attempt to form it
is formed for us, and how? Not, in general, by our
organization, nor wholly by our education, but by our
experience; experience of the painful consequences of
the character we previously had, or by some strong feel
ing of admiration or aspiration, accidentally aroused."

The Owenite whom Mill combats is his own veri
table shadow; or Mill is the shadow of the Owenite.
The attitudes of the two precisely correspond. The
Owenite alleges that man s character is formed for
him and not by him. Mill answers that we are capable
of making our own characters, if we will. The Owen
ite contends that this "if we will" surrenders the
whole point, and Mill ingenuously replies, "Most
true." The Owenite lays it down that man has no
power to alter his character by his wishing. Mill
thinks that he has, to some extent. The Owenite points
out that the will to alter our character is given us, not
by any efforts of ours, but by circumstances which we
cannot help; and Mill hastens to assure him that so
long as he stops there he is in a position from which
nothing can expel him. Is there any difference be
tween the disputants? Whatever there is comes only
to this, that the one would have our characters to be
formed for us and not by us; the other both for us



JOHN STUART MILL 185

and by us. But this difference disappears upon the
explanation which Mill affords, that our characters are
formed for us "in the ultimate resort," but by us "as
intermediate agents." Bearing in mind what Mill adds,
that this intermediate agency of ours is determined
by "external causes," this explanation is everything
that the Owenite could desire. Man starts with an
organisation which is none of his contriving: he re
ceives an education, that is, a supply of motives from
without, tending to direct him in a certain way: he
gets experience of painful consequences which he did
not mean to encounter: he also has strong feelings,
accidentally aroused. These and the like adventitious
determinants are the making of the man s character.
Character determined from without, and motive com
ing in from without, rule the man s every choice jointly.
It must be so in the absence of free will.

VI

"To think that we have no power of altering our
character, and to think that we shall not use our
power unless we desire to use it, are very different
things and have a very different effect on the mind.
A person who does not wish to alter his character can
not be the person who is supposed to feel discouraged
or paralysed by thinking himself unable to do it. The
depressing effect of the fatalist doctrine can only be
felt where there is a wish to do what that doctrine re
presents as impossible. It is of no consequence what
we think forms our character, when we have no desire
of our own to form it; but it is of great consequence
that we should not be prevented from forming such
a desire by thinking the attainment impracticable, and



1 86 FREE WILL

that if we have the desire, we should know that the
work is not so irrevocably done as to be incapable of
being altered."

To think that we shall not use our power to alter
our character unless we desire it, and further that we
shall not desire it except in accordance with some in
variable sequence analogous to the sequence of a feel
ing of heat from hot weather, seems to be the very
same thing and to have the very same effect upon the
mind as thinking that we have no power of altering
our character. We shall alter it, perhaps, when the de
sire of amendment supervenes: well, we wdl await the
desire, and when it comes, float out upon it to repen
tance and amendment.

It is of consequence what we think forms our cha
racter, even when we have no present desire of a re
formation. For it is important, as Mill well says, "that
we should not be prevented from forming such a de
sire." But we should be prevented, if we thought that
the desire, when it came, would take hold of our minds,
as the tide of a log lying upon the beach, without our
seconding it, and without our being on the alert to
transmute by our conscious sanction the spontaneous
craving, the velleity for better things, into a solid and
effective purpose of amendment.

VII

"And indeed, if we examine closely, we shall find
that this feeling of our being able to modify our own
character if we wish, is itself the feeling of moral free
dom which we are conscious of. A person feels morally
free who feels that his habits or his temptations are not



JOHN STUART MILL 187

his masters, but he theirs: who even in yielding to
them knows that he could resist; that were he desirous
of altogether throwing them off, there would not be
required for the purpose a stronger desire than he knows
himself to be capable of feeling. . . The free will doc
trine, by keeping in view precisely that portion of the
truth which the word Necessity puts out of sight,
namely, the power of the mind to co-operate in the
formation of its own character, has given to its adherents
a practical feeling much nearer to the truth than has
generally, I believe, existed in the minds of necessarians.
The latter may have had a stronger sense of the impor
tance of what human beings can do to shape the cha
racters of one another; but the free will doctrine has,
I believe, fostered in its supporters a much stronger
spirit of self-culture."

Here Mill has imitated the tactics of his admired
master, Locke. I have remarked how Locke (11.9) shifts
his ground, and without express adherence to free will
nearly becomes a libertarian. And it has been observed
of Mill, by one of the ablest of his opponents,* that he
answers objections by yielding to them, and yet will
not resign the pretensions of his school. The present
passage, if it means anything, means a withdrawal of
the application of the doctrine of invariable and uncon
ditional sequence to the operations of the will. But
Mill has not written a book of Retractations.

Let us " examine closely," as he suggests, and ac
cording to his description, " the feeling of moral free
dom which we are conscious of." When we feel morally

* John Grotc, in his Examination of the Utilitarian Ththsophy. It is
an Oxford saying: "The best things in Mill are his admissions."



1 88 FREE WILL

free, we are conscious, according to Mill, of three facts:
(i) that we are able to modify our own character, if we
wish: (2) that we are masters of our habits and temp
tations, not they of us: (3) that we could resist habit
or temptation even when we yield to it. The first of
these facts has already been discussed. The wishing, on
which our ability to shape our character is conditioned,
must rest consciously with us: else how can we be con
scious of possessing that ability? As for the second fact,
our conscious mastery over our habits and temptations,
the said " habits " and " temptations " are the same as
the "dispositions" and "motives" respectively, which
Mill formerly declared to be the causes whence our
actions flow in uniform sequence. But if the sequence
is uniform, we are not masters of our actions, and there
fore not of our habits and temptations: they rule us,
not we them. Indeed this we is a new term, not intro
duced before. Before there were antecedent circum
stances followed by consequent acts; now there comes
on the scene a person, a conscious agent, who claims
the acts for his own and disputes the mastery of them
with the circumstances. A transition appears to have
been made from physics to psychology. The third fact
which Mill learns from consciousness is that we can resist
temptation even when we yield to it. The very thing
that libertarians say, and the one thing that they care
to keep to! The great champion of the uniformity of
nature acknowledges free will, he avows that he is con
scious of it. Let all that is here written against him be
cast into the fire, and let his literary executors cancel
his chapter on Liberty and Necessity, all except the



JOHN STUART MILL 189

present passage; and the little world that reads our
books will be delighted with the unusual spectacle of
a philosopher come to terms with his adversary. For
if in yielding to a temptation we know that we can
resist, we know that our yielding is not a sure con
sequence of the circumstances of trial in which we
stand. Therefore the chain of uniformity does not bind
volition. Catching at a quibble to hold him from being
drawn into this concession, Mill might insist that his
word is could) not can resist; and he might explain
himself to mean that we could resist, if circumstances
were different, but cannot as they are. But is conscious
ness of what we might help in another case, but can
not help in the present, a consciousness of not being
here and now overpowered? If this is freedom, no man
ever was a slave, for never was man placed in circum
stances in which he could not have broken his bonds had
not the said bonds been there and then too strong for
him. In Mill s work On Liberty there is a chapter "Of
Individuality as one of the conditions of well-being."
In that enthusiastic and paradoxical vindication of indi
viduality, one may mark the loathing with which the
author turned from rigid necessarianism, a loathing
which has got the better of his respecT: for consistency,
and wrung from him a confession of free will in the
midst of a treatise that argues universal uniformity.

VIII

"What experience makes known is the facl: of an
invariable sequence between every event and some
special combination of antecedent conditions, in such
sort that wherever and whenever that union of ante-



1 90 FREE WILT.

cedents exists, the event does not fail to occur. Any
must in the case, any necessity, other than the uncon
ditional universality of the fact, we know nothing of.
. . . The so-called Necessitarians . . . affirm, as a truth
of experience, that volitions do, in point of fact, follow
determinate moral antecedents with the same unifor
mity, and, when we have sufficient knowledge of the
circumstances, with the same certainty as physical
effects follow their physical causes. . . . This is what
Necessitarians affirm, and they court every possible
mode in which its truth can be verified. They test it
by each person s observation of his own volitions. They
test it by each person s observation of the voluntary
actions of those with whom he comes into contact, and
by the power which every one has of foreseeing actions
with a degree of exactness proportioned to his previous
experience and knowledge of the agents, and with a
certainty often quite equal to that with which we pre
dict the commonest physical events. They test it fur
ther by the statistical results of the observation of
human beings acting in numbers sufficient to elimi
nate the influences which operate only on a few, and
which on a large scale neutralise one another, leaving
the total result about the same as if the volitions of the
whole mass had been affected by such only of the deter
mining causes as were common to them all. In cases
of this description the results are as uniform, and may
be as accurately foretold, as in any physical enquiries
in which the effect depends upon a multiplicity of
causes. The cases in which volitions seem too uncer
tain to be confidently predicted are those in which our
knowledge of the influences antecedently in operation
is so incomplete, that with equally imperfect data there
would be the same uncertainty in the predictions of the
astronomer and chemist. On these grounds it is con-



JOHN STUART MILL 191

tc rule*.! that our choice between the conflicting incon-
ceivables should be the same in the case of volitions
as of all other phenomena; we must reject equally in
both cases the hypothesis of spontaneousness, and con
sider them all as caused. A volition is a moral effect,
which follows the corresponding moral causes as cer
tainly and invariably as physical effects follow their
physical causes. Whether it must be so, I acknowledge
myself to be entirely ignorant, be the phenomenon
moral or physical; and I condemn, accordingly, the
word Necessity as applied to either case. All I know is
that it always does"

This, and the following extracts, are from Mill s
Examination oj Sir William Hamilton s Philosophy. Mill
here repeats what he has written in his Logic, that is
to say, he repeats Locke and Hume. I am, therefore,
compelled to repeat myself. I count it no disadvantage,
on a difficult topic, to be led into some repetition.

We have experience of necessity, and of what Mill
terms the "a priori must," quite as much as of inva
riable sequence and the " a posteriori does." Experience
is either immediate or mediate: it takes the form either
of intuition or inference. W T e cannot be always, nor
go everywhere; we cannot, therefore, gain immediate
experience of the working of a law in all times and
places. Creatures of a day, we cannot crowd an inva
riable sequence into our field of view. If we know any
such sequence, we know it only by inference, by as
suming that an observed uniformity obtains beyond
the sphere of observation. But how justify this infer
ence, how warrant the passage from "It does so far as
I have seen" to "It always does"? The a posteriori



192 FREE WILL

"does" does very well; but the a posteriori "will do,"
I fear, will not do.

The a priori "must" comes to the rescue. If we
know what must be, we have ground to predial what
will be. How, then, do we know what must be? And,
first, how do we conceive it? By considering our in
tuitions of what is. Surely we do right to examine
ideas which we have already got. We do no violence
to experience by counting the treasure which experi
ence has bestowed on us. Intuition of self reveals "I
am," "I do," and thence "I can." The reverse of that
is "I cannot"; whence, by reduplication of the nega
tive, "I cannot not," which is "I must." Transferring
the idea from self to not-self, we conceive "thou
must," "it must." But when do we know that a thing
must be ? So far as we are concerned, that must be
which we find ourselves unable to prevent. To pre
vent a thing by our personal exertions we require to
know of it: ignorance in us means incapacity of inter
ference. Given, therefore, an agent without under
standing, we know that it cannot help whatever it does
or suffers; that it must do all that it can do, and suffer
all it can suffer under the circumstances in which it is
placed; that it is, in fad, a necessary agent. Whatever
it can do it does, and must do, if there be a term to
work upon within its sphere of action. What a brute
agent once does, it must ever do, ilne sail pas faire au-
trement. This necessity cleaves to the substantial abi
ding nature of a brute agent. To know that nature,
then, in the present, contains a knowledge of its aclion
in the future. A phenomenon of matter will be, because



JOHN STUART MILL 193

it must be. That positis ponemlis it must be we see in
the cognition of the noumenon, the material substance
which is at once the necessary agent, efficient cause of
formerly observed phenomena, and the guarantee of
like effects to follow under like conditions to come.

Rational natures often act unconsciously: they wake
up to consciousness of mental states not of their own
choosing. But then they can refrain from enhancing
such a state within themselves: at the same time they
can enhance it. The fuller their reflection, the greater
their liberty in this particular. Looking into them
selves, they become masters of their affections. Thus
they are free to will or, as the Elizabethan writers said,
to afeff, or not to will and affect the objects that occur.
But for this freedom, the verb can in our mouths
would be foolish. "I can go to bed" means, in the first
place, "I can make up my mind to do so." If my re
solve were determined for me by the accidents of my
position, it would be as idle for me, sitting in my chair,
to say, "I can go to bed," as for the stones of West
minster Abbey, could they find a voice, to cry to the
Dean and Chapter assembled below, "We can fall and
crush you." Allow the possible as distinct from the
actual, and one must allow the impossible, and thence
the necessary and the contingent. To deny necessity
is to deny possibility and impossibility. But it is as
egregious wilfulness in a psychologist to set aside any
element of human consciousness as for a chemist to
expel from his laboratory chlorine and its compounds,
forsooth because he has a theory with which those
bodies do not square. As such a theory of chemistry

13



194 FREE WILL

would be "done all on one side," so, too, is the phi
losophy one-sided that ignores necessity. And Mill
does ignore necessity: he will not hear of it in physi
cal science: he banishes it from the science of mind.
Still the phrase, "It must be," has a meaning: every
man understands it: a philosopher should take account
of it.

Mill alleges three proofs, not of the necessary, but
of the invariable sequence of volitions from certain
antecedents. He appeals, first, to self-consciousness.
Here is room not so much for controversy as for re
flection. Reflection upon self is indispensable to the
psychologist. It is, nevertheless, a somewhat untrust
worthy source of knowledge. Entering into ourselves,
we see what we go to see, and few of us go to see
ourselves as we are. However, there is one reason for
thinking that the insincerity in this question rests not
with those psychologists who affirm the consciousness
of free will. For what does that imply? "My will is
free," in the mouth of a man on earth, implies, among
other things, this: "I am capable of sinning." That,
again, if we consider who the speaker is, further im
plies, to a greater or less extent, "I have sinned." Be
lievers in free will believe in sin. One section, also, of
disbelievers in free will have professed to believe in
sin: I mean the Old Calvinists. The greater number,
however, of disbelievers in free will, including all who
deny the doctrine on other than theological grounds,
have no belief in sin. They believe in noxious actions,
restrainable by motives, but not in sinful actions.
"Thou art the man" is not their word to the evil-



JOHN STUART MILL 195

doer. They would shield him with Adam s excuses,
his wife, the serpent, his temper and his circumstances.
They put crime in the same rank with disease: they
would have a criminal operated on for his cure; not
punished for his guilt. "Sweet shall be thy rest," says
the author of the Imitation of Christ, "sweet shall be thy
rest, if thy heart do not rebuke thee." The rest which
phenomenalists enjoy ought to be delicious indeed:
their heart cannot rebuke them, if they are, as they
represent themselves, unconscious as babes of the very
possibility of sinning. There is but one way heartily
to enjoy this world; that is, to put sin out of the list
of possibilities to be thought of. This comfortable way
phenomenalists have found. But the author quoted
above, speaking of ungodly men, who say they are at
peace, gives this warning, "Believe them not, for the
wrath of GOD shall rise of a sudden, and their deeds
shall be brought to nothing, and their speculations
shall perish." To deny sin is hardly the way to escape
the wrath to come, if it be to come. And men know
it is to come, and they know why, because in spite of
themselves they know that they have sinned. The
sense of sinfulness is written too deep in man s heart,
it has operated too widely amongst mankind, to be a
misconception, a psychological solecism. But it is no
more unless the will be free. Sin impossible? Would
it were so! But I fear that, were it not possible, men
could never have imagined such a horror. I conclude
that, of the philosophers who find free will in their
consciousness and sin upon their conscience, and of
those others who declare that they are unconscious



196 FREE WILL

alike of being free and of having sinned, the latter are
the more likely to be deceiving themselves, and to
have not the truth in them.

Mill s second proof is borrowed from the fact that
men can foretell each other s behaviour better or
worse as they know more or less about one another.
This fact proves nothing for him, if it stands as well
with liberty as with uniformity in volition. And so it
does. Free will is not indifference to motives: it is ab
sence of any absolute constraint from the particular
motive that is uppermost in the mind at any given
moment. But there may be more or less an approach
to constraint. A person is left more free under some
motives than under others. A knowledge of his mo
tives is a probable clue to his action. Still more is the
probability of the estimate increased, if, along with
motives, we know also his character, which we may
know by knowing how he has behaved on similar
occasions before. Every time a man does a thing, he
diminishes his liberty of not doing it next time;
he makes the act in some degree natural to him, and
necessary in so far as it is natural. A habit is not
broken without a special motive. The better a man s
habits and motives are known, the more calculable his
action becomes, calculable, I mean, with an ever in
creasing probability. Nor do I care to deny that some
of man s actions may be calculated with absolute
certainty. Such actions, if such there be, are neces
sary; but frequently they are what is called "free in
their cause," being acts proceeding from a habit which
was engendered originally of free acts. While Mill



JOHN STUART MILL 197

holds that all acts are absolutely calculable in them
selves, and are relatively incalculable to us because of
our ignorance of their antecedents, libertarians will
have it that some acts are absolutely beyond calcula
tion, as not following rigidly from antecedents. Neither
view is inconsistent with the facts of our experience.
Necessarianism is not provable a posteriori.

Mill s third proof from "statistical results" shows
no more than this, that many men are sure to do what
all are inclined to do. Probability for each is certainty
for some, out of a large number, but not for any defi
nite individuals. Free will, however, is an attribute of
men taken individually, not collectively. And antece
dent probability of action is compatible with a degree
of freedom.

I reject, equally with Mill, "the hypothesis of spon-
taneousness" about volitions, and "I consider them
all as caused." That is to say, I do not believe an act
of the will to come out of nothing, a causeless pheno
menon. I hold that the person who wills causes his
own volition, under certain motives as conditions. To
Mill the person is nobody; that is why he would call
a free act "spontaneous," meaning that it has no cause.
I do not, however, agree that volitions are "caused"
in Mill s sense of the term, or that an "explanation"
can be found for them, as for physical events.

"A volition," says Mill, "is a moral effect, which
follows the corresponding moral causes as certainly and
invariably as physical effects follow their causes." The
absurdity of this proposition is manifest, when we con
vert it into the following equivalent form: "A volition



198 FREE WILL

is a conscious act, which is done by a conscious agent
or person, as necessarily as an unconscious act is done
by an unconscious agent or thing."

IX

"To be conscious of free will must mean, to be
conscious before I have decided that I am able to de
cide either way. Exception may be taken in limine to
the use of the word consciousness in such an applica
tion. Consciousness tells me what I do or feel. But
what I am able to do is not a subject of consciousness.
Consciousness is not prophetic; we are conscious of
what is, not of what will or can be. We never know
that we are able to do a thing, except from having done
it, or something equal and similar to it. We should
not know that we were capable of action at all, if we
had never acted. Having acted, we know, as far as that
experience reaches, how we are able to act; and this
knowledge, when it has become familiar, is often con
founded with and called by the name of consciousness."

/ do, I can, and I am, are three facets of the same
truth, all three known together in present conscious
ness. / do implies / can. I do and / can imply / am,
for there is no activity nor power in non-existence.
Again, / am signifies / can and also / do; there is no
substantial being without power, and there is no
power where there is no act, though outward action is
not coextensive with power. / do in the present, / can
in the present, and / am in the present. Mill acknow
ledges the present truth of / do and / am, but not of
/ can. He thinks that when I declare / can, I announce
some future fact; but "consciousness," he says, "is
not prophetic." Conscious I am of being, Mill allows,



JOHN STUART MILL 199

and conscious of doing, but not conscious of power.
My belief in any power of my own he holds to be an
inference from what I have done to what I shall do
again in like circumstances. But surely, "I can do a
thing" does not mean "I shall do it." When I act, I
am conscious alike of action and of power, both in
the present. The action passes, but the consciousness
of power remains. There is nothing "prophetic" about
it. It is true that we learn our powers by exercising
them. And we learn that we have a free will by exer
cising it. It is a consciousness that comes of experi
ence. There is no innate idea of free will. The will
is not free in childhood. To say that a child has come
to the use of reason means that his will is now be
ginning to assume command of his conduct. We learn
to will as we learn to lift. There are weights that we
cannot so much as stir. And about many circumstances
and conditions of life our will is utterly powerless. We
learn to know hard necessity, things that we cannot
help, in contrast with what we can help. Necessity
strikes us most when it is about feelings of our own,
of a pleasurable or painful kind. Many such feelings,
e.g., those of temperature in our own bodies, are partly
under our control and partly beyond our control.
Such experience especially helps on the cognition of
free will. But free will comes out most of all in the
matter of impulses. Moral education begins in the
checking of impulses, notably those passionate out
bursts of crying characteristic of infantsgenerally. Aided
by much persuasive impulse from without, the child
comes to cry a little less. There is nothing of free will



200



FREE WILL



here, because there is not as yet any reflex conscious
ness, nothing more than that formative process which
we observe in the higher animals under the training
of man. The assertion of self against impulse is very
gradual. When that assertion takes definite shape, free
will has begun. One day an impulse is curbed in this
way; another day it is allowed free scope. But in
giving it scope, the young agent remembers, " I helped
crying, or getting angry, or frightened, yesterday."
The inference thence is not beyond the range of a
child of six or seven, " I might have helped getting
angry to-day." There we have an initial consciousness
of free will. There is nothing mysterious in the pro
cess, nothing inconsistent with the nature of a con
scious a<5t. It is a reading of one s own present state
in the light of a remembered similar past. There is
no reference at all to the future; nothing of the ele
ment that Mill calls "prophetic."

Mill s mistake, common to him with Locke, is that
of confusing the will to act in a certain way with the
power of executing such volition. Nothing certainly
is more frequent than for people to fancy themselves
conscious of abilities, which further experience proves
that they do not possess. Conscious of his swimming
powers, so he thinks, the unfortunate youth jumps
into the quarry pond and is drowned. We have
in such cases to distinguish between man and his cir
cumstances. Man is conscious of what depends on
himself; he is not conscious of what depends on cir
cumstances. He makes an effort and hopes it will suc
ceed. The effort is perhaps the main element of success,



JOHN STUART MILL 201

but it is not success. To be conscious of ability is to
be conscious of that which in us lies, not of what lies
without us. Therefore, * I am conscious I can swim,
is a twofold judgement of consciousness and of infer
ence. It comes to this: I am conscious I can try, and
I argue from past experience that my attempt will be
successful. The consciousness here is infallible, but
further experience in unwonted circumstances may
overthrow the inference. Hence we may learn to dis
tinguish what truth there is in Mill s saying, that the
assertion / can is " prophetic." So far as it means, * I
can use my endeavours, the assertion / can is a fact
of present consciousness: so far as it means, those
endeavours will be adequate to the occasion, it is an
inference from the past to the future.

X

"But this conviction, whether termed conscious
ness or only belief, that our will is free what is it ?
Of what are we convinced ? I am told that whether I
decide to do or to abstain, I feel that I could have
decided the other way. I ask my consciousness what I
do feel, and I find, indeed, that I feel (or am con
vinced) that I could have chosen the other course if
1 had preferred it; but not that I could have chosen
one course while I preferred another. When I say pre
ferred, I, of course, include with the thing itself, all
that accompanies it. . . Take any alternative: say to
murder or not to murder. I am told that if I elect to
murder, I am conscious that I could have elected to
abstain: but am I conscious that I could have abstained
if my aversion to the crime, and my dread of its con
sequences, had been weaker than the temptation? If



202 FREE WILL

I elect to abstain, in what sense am I conscious that
I could have elected to commit the crime ? Only if I
had desired to commit it with a desire stronger than
my horror of murder; not with one less strong. When
we think of ourselves hypothetically as having acted
otherwise than we did, we always suppose a difference
in the antecedents : we picture ourselves as having
known something that we did not know, or not known
somethingthatwedidknow; which isa differencein the
external motives; or as having desired something or dis
liked something more or less than we did; which is a
difference in the internal motives. I, therefore, dispute
altogether that we are conscious of being able to act in
opposition to the strongest present desire or aversion."

Let us for a moment suppose that the doctrine,
here laid down by Mill, is true. Let us take his exam
ple of a man who has before him an alternative to mur
der or not to murder. Then, if that man s aversion to
the crime and his dread of its consequences are wea
ker than the temptation, he cannot abstain from the
murder: he needs must commit it in that case. If, on
the other hand, his desire to commit the crime is wea
ker than his horror of it, he cannot commit the mur
der, but must needs abstain from it. Whence I argue
thus. Either the temptation is stronger than the hor
ror of the crime, or the horror of the crime is stronger
than the temptation, the case of the two being equal
is a blank. Which ever way it is, the man s election is
necessitated; and as of this, so of every other election
that a man may be called upon to make, all are neces
sitated; therefore, the true theory of volition, as Mill
expounds it, is absolute necessarianism.



JOHN STUART MILL 203

Having followed Mill to a goal which he himself
somewhat deprecates, let us retrace our steps to the
point where we differ from him. It is a slight point, so
slight that he has overlooked its being a possible occa
sion of difference. He " disputes altogether that we are
conscious of being able to act in opposition to the
strongest present desire or aversion." I dispute it also;
indeed, in strict parlance, though, of course, not in the
popular sense, I deny it. We cannot act in opposition
to the strongest present desire, while that desire is at
the present strongest. But frequently we are able to
refrain from acting in accordance with the strongest
present desire (or aversion). Suppose I have a desire
to pull a house down because it is inconvenient, and
also a desire to leave it standing because it is endeared
to me by old associations. I cannot feel two such in
compatible desires both at exactly the same instant, but
I feel now one and now the other. Each in turn is the
stronger at the instant at which I feel it, though one
may be stronger than the other on the whole, as com
ing oftener and being more intense when it does come.
If the proposition, which Mill and I alike dispute,
simply means that I can finally act against the desire
which on the whole is stronger, I cannot stand with
Mill, for the proposition in that sense is true. But if
the meaning is that I can do the very reverse of what
at the moment of my action I supremely long to do,
I protest with Mill against the proposition. To me, as
to Mill, it appears incredible that a man should choose
one course, and at the same time prefer, altogether
prefer, the reverse. Such a choice would turn the laws



204 FREE WILL

of volition topsy-turvy. Let us go back to the example
of the house. At this instant, we will say, the desire
of pulling down the building is uppermost in my soul.
By the very fact that I have that desire now, I do not
desire at the same moment to let the building stand.
I have a spontaneous complacency in the idea of de
struction; and that, while it lasts, prevents me from
being complacent in the idea of conservation.* If I
consummate any volition now by a reflex approval of
a spontaneous complacency, the approval must fall on
the complacency which I have now. I cannot at pre
sent make up my mind to keep the house standing: for
the one "bill," so to speak, at present awaiting my
royal assent is a bill to pull it down. A man cannot
will in opposition to, I do not say his animal or physi
cal, but his psychical and volitional impulse while that
impulse actually reigns; nor being spontaneously com
placent in one purpose can he become then and there
reflexly complacent in another. Thus far I go along
with Mill. He proceeds tacitly to assume that a man
must positively acl and reflexly will in accordance with
his strongest present desire, and there I fall off" from
him. I say the man can wait. Once more to the house.
Desiring to pull the old place down I cannot resolve
to keep it standing, but I can stay and view my desire.
And while I view it, the desire fades away, and I re
main thinking, but not willing, what I shall do with
the old place. The desire to keep it now rises and be
comes predominant. I cannot will to pull the building

*The coiner s press must stamp just that bit of metal which at
that moment lies under it, if at that moment it stamp? anything at all.



JOHN STUART MILL 205

down while I feel an actual desire to keep it; but at the
same time I need not will to keep it. So I go on till at
last I will in accordance with some present desire.

When the volition is completed, I look back upon
my ad. I say it was freely done, by which I mean,
not that I could at the instant have acted otherwise,
but that I could at the instant have refrained from
acting in the way I did. In the moment when the act
of my will was done, though I could not have acted
otherwise, I need not have acted at all. I might have
been quiescent: I might simply not have approved
the complacency at that time being. Without any dif
ference in the antecedents, without any learning of
anything that I did not know, or becoming ignorant of
aught that I did know, or desiring or disliking more
or less than I spontaneously did desire or dislike, I
might have held aloof from that complacency which
I sanctioned and made into a full volition. But for me
then to have embraced a different complacency, and to
have performed a different act of the will, does sup
pose a difference in the antecedents, just such a dif
ference as Mill says "we always suppose when we think
of ourselves hypothetically as having acted otherwise
than we did." Mill s dictum is right, formally for the
psychological instant of decision, but not for the whole
of the deliberative process which is popularly called
the "action."*

See Locke, n. 9.



206 FREE WILL

XI

"It is not the belief that we shall be made account
able, which can be deemed to require or presuppose the
free-will hypothesis; it is the belief that we ought so to
be; that we are justly accountable; that guilt deserves
punishment. It is here that the main issue is joined
between the two opinions. In discussing it, there is no
need to postulate any theory respecting the nature or
criterion of moral distinctions. It matters not for this
purpose whether the right and wrong of actions de
pends on the consequences they tend to produce or
on an inherent quality of the actions themselves. It
is indifferent whether we are utilitarians or anti-
utilitarians; whether our ethics rest on intuition or
on experience. It is sufficient if we believe that there
is a difference between right and wrong, and a natural
reason for preferring the former. . . The real question
is one of justice the legitimacy of retribution or
punishment. On the theory of necessity, we are told,
man cannot help acting as he does, and it cannot be
just that he should be punished for what he cannot
help. Not if the expectation of punishment enables
him to help it, and is the only means by which he can
be enabled to help it? ... There are two ends which
on the necessitarian theory are sufficient to justify
punishment: the benefit of the offender himself and
the protection of others. The first justifies it, because
to benefit a person cannot be to do him an injury. To
punish him for his own good, provided the inflictor has
any proper title to constitute himself ajudge, is no more
unjust than to administer medicine. . . In its other
aspect punishment is a precaution taken by society in
self-defence. . . If it is possible to have just rights, it
cannot be unjust to defend them. Free will or no free
will, it is just to punish so far as is necessary for this



JOHN STUART MILL 207

purpose, exactly as it is just to put a wild beast to
death, without unnecessary suffering, for the same ob
ject. Now, the primitive consciousness we are said to
have, that we are accountable for our actions, and that,
if we violate the rule of right, we shall deserve punish
ment, I contend is nothing else than our knowledge
that punishment will be just; that by such conduct we
shall place ourselves in the position in which our
fellow-creatures, or the Deity, or both, will naturally
and may justly inflict punishment upon us. By using
the viordjust/y I am not assuming in the explanation
the thing I profess to explain. As before observed,
I am entitled to postulate the reality and the know
ledge and feeling of moral distinctions."

Mill is righting against an objection which may be
put into syllogism thus.

We cannot know that we ought to be punished for
our misdeeds, without knowing also that our wills
are free.

But we do know that we ought to be punished for
our misdeeds.

Therefore we know also that our wills are free.

Mill denies the major of this syllogism, and under
takes to prove the contradictory, that we can know that
we ought to be punished for our misdeeds, without
knowing also that our wills are free; in other words,
that the notion of just punishment does not involve
the notion of free will. The way to prove this thesis
would be to explain the meaning of the phrase, "we
ought to be punished," and to show, if possible, that the
phrase does not contain any reference to free will. But
Mill starts with the surprising announcement that the
reason of the right and wrong of actions, which explains



2o8 FREE WILL

why we ought to be punished when we do wrong,
matters not for the purpose of his proof. Surely, it is
on that very reason that the proof depends. How can
anyone discuss why punishment is just, without his
argument involving some view as to the essential
nature of justice? But that again involves some theory
of morals. Indeed, one of the greatest treatises on the
theory of morals ever written, the Republic of Plato,
starts from this very inquiry, What is justice? Mill s
antagonists here contend that the denial of free will
puts quite a new face on justice and just punishment;
and that the ordinary notion of justice and just punish
ment is founded on the assumption of free will; that
consequently, to ordinary minds, to punish a man for
a deed which there and then he could not help is
unjust. Mill s reply, fair enough in its way, is that the
ordinary notion of justice is altogether a mistake. He
proceeds to inculcate instead his own notion of justice
and just punishment, which is the blankest utilitari
anism. Mill s compeer, Bain, correctly writes: "Assu
ming that the imposition of punishment is the distinc
tive property of acts held to be morally wrong, we are
next to inquire on what grounds such acts are forbidden
and hindered by all the force that society or individuals
possess. What are the reasons or considerations re
quiring each one to abstain from the performance of
certain actions, and to concur in a common prohibition
of them, enforced by stringent penalties? The answer
to this is the theory of morals."* This is saying, and it
is well said, that some theory of morals is implied in the
belief that certain actions ought to be punished. How

* Emotions and H lll, second edition, p. 254.



JOHN STUART MILL 209

then can Mill pretend that "it is indifferent whether
we are utilitarians or anti-utilitarians?" His whole
argument is constructed on a basis of utilitarianism.
When a philosopher writes, " There is no need to
postulate," let the reader beware, and till he sees it
to be otherwise, let him expect that his author is going
to take for granted the point which he says there is no
need to postulate. It is not dishonesty on the philoso
pher s part that prompts him to this stretch of the
" privy paw." The stealth is ascribable to a mixture of
zeal and mistrust. Observing some pet doctrine in
want of a particular support, and doubting of our
ability to secure it in face of the opposition of our
adversary, we yield to the nervous eagerness of desire
which makes men say the reverse of what they should
say, and we bid the other party distinctly to take notice
how we scorn that support on which all the while our
doctrine rests. Thus Mill needs the utilitarian morality
to bear out his assertion, that crime ought to be pun
ished, free will or no free will. But his opponents are not
utilitarians, and to convert them is not worth his while
to try. Therefore he denies his need, at the same time
taking what he needs for granted. He lays down utili
tarian definitions concerning crime and punishment
and justice. He lays beside them the necessarian prin
ciple, that a man cannot help the crime that he com
mits. He applies the said definitions to the said prin
ciple, and the result appears accordingly, that it is just
to punish a man for the crime that he cannot help. In
other words, it is expedient for the greatest happiness
of the greatest number, that a man who has been com-



210 FREE WILL

pelled to mar that happiness for want of a motive to
maintain it, should suffer such an amount of pain as
shall furnish a motive to compel him on the next temp
tation to respect the common interests of humanity.
A just procedure, on condition that crime be made
out an accident, punishment a surgical operation,
justice expediency, and man a motive-worked automa
ton. At that rate I readily understand how, " free will
or no free will, it is just to punish so far as is neces
sary for this purpose, exactly as it is just to put a wild
beast to death."

Utilitarianism is a ruthlessly logical system, but it
is not a system of morals. Elsewhere I have styled it
" an abyss of chaos and confusion," in which " moral
philosophy finds her grave."* Abiding by that verdict,
I say that to punish a man for what he cannot help is
an insult to the dignity and a violation of the rights of
man. To punish is not simply to pain: it is to pain and
to blame together. Though it be sometimes just, for
a man s own benefit and for the protection of others,
to make him suffer pain for what he cannot help, it can
never be just to blame him for what he cannot help. The
castigations which we inflict on children and brute
animals are only styled punishments in an improper
sense of the term, inasmuch as they are not accom
panied with moral reproach. It is from an exclusive
study of this improper sense that utilitarians have
evolved their theory of punishment, a theory which
supposes that a wicked man, a " naughty boy," and a

* Ethics and Natural Law, Stonyhurst Series, pp. 177-189.



JOHN STUART MILL 211

restive horse, are all on a level as objects of punishment.
A moment s consideration destroys this supposition.
Man, boy, and horse receive stripes alike; but the man
is blamed severely, the boy perhaps slightly, the horse
not at all. The blame is an essential portion of the man s
punishment; it is that which gives it the sting. The
animosity shown against men of blood, marking them
off from mere beasts of prey, Mill would set down to
the desire to see an example made of noxious men, lest
they should breed imitators. That is indeed a reason why
the murderer should suffer. But it is not a reason for
holding him in abomination. Abomination is not pro
spective, precautionary, prudential, as Mill would
have the entire treatment of wrong-doers to be. To
punish is not to dispense suffering as a chemist dispen
ses drugs: punishment is sufferingattended with blame.
Blame supposes that the delinquent could and ought
to have done otherwise.

As there seems to be something incompatible with
utilitarian ideas in the reprobation heaped on crime
by the common people, it will be desirable in the day
when criminals shall be confined among the beasts, for
some disciple of Mill to stand beside the cages to
rectify the vulgar errors of the visitors. The tenor of
his lecture might be as follows:

" That bear there you observe, ladies and gentle
men, yesterday morning hugged his keeper: the man
was carried out dead from that fatal embrace. The
animal in the next compartment is a man, who has
murdered his wife. As the other bears are not likely



212 FREE WILL

to be influenced by that one bear s example, it has not
been thought necessary to punish him: but the new
keeper has received instructions not to take liberties
with his charge. The man, however, is to be made a
warning to his fellows. It were in evil precedent for
our species if a husband could kill his wife with im
punity. Therefore is he deprived of his liberty, and
any afternoon at one o clock you may see him publicly
flogged. Not that his guilt is greater than the bear s:
but prudence requires that he should undergo a more
exemplary punishment. Be pleased, therefore, not to
censure, blame, loathe or abominate the man any more
than you loathe the brute. You call the murderer a
brute, and it is well you should: only remember that
nature made him what he is, no less inevitably than
she made the bear. The keeper s death took place in
accordance with an invariable law. Had he indeed not
enraged the animal, the law would not have come into
operation. But he did enrage the animal, and the animal
killed him necessarily. You do not blame Bruin for
that. The man is as blameless as the bear. He had
from nature, to start with, a certain organisation and
certain susceptibilities of character. He grew up in
the midst of circumstances, which followed other cir
cumstances in unvarying sequence, like the heat and
cold, and sun and rain, under which a water-lily springs
on the bosom of a lonely lake. A plant s growth is
determined by two fadors, germinal capacity and en
vironment: so is a man s character made for him by
nature and by circumstances. This poor fellow could
not help killing his wife. There are motives which



JOHN STUART MILL 213

would have saved him from the crime, but they were
out of his reach. He had them not, and could not get
them. If his wife had been wearing an iron helmet,
the blow would not have proved fatal. But she had
no helmet to wear, and so was fain to die, as her hus
band was fain to kill. One is no more to be blamed
than the other for what neither could help. It was the
uniformity of nature, the same which tempers the
heat of the sun and measures the orbit of the moon,
that brought the husband to strike with the cleaver
and the wife to die of the wound. Let us hope, ladies
and gentlemen, that nature is not steadily preparing,
in the order of her sure sequences, a similar fate for
you and me."

This, it will be said, is turning philosophy into buf
foonery. It may be buffoonery, but, to the writer at
least, it is not mirth. It saddens my heart to read
utterances like the following: "You discern nothing
while your eye is fixed on Archelaus himself. . . But
when you turn to the persons whom he has killed,
banished, or ruined . . . there is no lack of argument
to justify that sentiment which prompts a reflecting
spectator to brand him as a disgraceful man. . . It will
indeed be at once seen that the taint or distemper with
which Archelaus is supposed to inoculate himself when
he commits signal crime ... is a pure fancy or poetical
metaphor on the side of Plato himself."* To say that
sin is no stain is to say that it is not sin. A criminal

*Grote s Plato, n, 109, ill. There is nothing "disgraceful" in
being a usurper and a tyrant, if one cannot help it. Men should not
be reproached for their natural defects nor for the circumstances of
their education.



2i 4 FREE WILL

in this view is toned down to a hurtful agent: "a
Borgia and a Catiline" appear no worse than " storms
and earthquakes." Wicked men certainly are influenced
by motives, hurricanes are not. The inference which
I should draw, as an utilitarian, measuring moral evil
by material damages, is that hurricanes are more wicked
than men, as being more incorrigibly noxious. A deed
of ours which we cannot help may hurt our fellow-
men, but we are quite aware that that is no sin. We
should not resent being put to inconvenience to pre
vent the recurrence of the mishap, being shut up by
ourselves, for instance, when we have unwittingly com
municated an infection. But we should resent being
punished for it, that is, being reproached as well as
inconvenienced. It is the earliest excuse of a child,
I could not help it. The stupid, rude answer, But I ll
make you help it, has a ring of tyranny and injustice
in the ear of a little one.

One last word on the theory of punishment. The
theory, as we have seen, needs to be modified to fit in
with the hypothesis of determinism. But, I observe,
not only does the theory, as a theme of academic
discussion, need modification: an important change
must likewise be made in our criminal jurisprudence,
and in the practice of our courts of justice, those
arbiters of life and death. I refer to the case of the
criminal lunatic, afflidled with homicidal mania. Like
other madmen, these unhappy persons are by no means
inaccessible to motives, especially of the more violent
sort. It is quite possible to inspire them with fear and
so deter them from offending; and this possibility is



JOHN STUART MILL 215

greater, and the deterrent more effectual, ere they have
yet shed blood. These early stages of the malady should
be contemplated by the preventive eye of the judge.
If a new legal maxim were introduced, and enforced
by example before the eyes of all men, that insanity
shall no longer enter into the verdict, and that crimi
nal lunatics henceforth shall be hung for homicide as
inexorably as other murderers are hung, then persons
of unsettled reason, and others whose criminal habits
are gradually unsettling their reason, would have a
strong motive provided them to keep them from
shedding blood, and this provision would be a notable
addition to the safety of the community. Why should
not the law provide this additional security ? If any
criminal at all should be hung for murder on deter-
minist principles, we should hang the criminal lunatic
for murder. Of all dangerous persons he is the most
dangerous: his type is the most clearly marked: his
character is the most set, and his execution would be
the most exemplary.

The sole reason for sparing the life of this dangerous
person is drawn from the ancient belief of that Chris
tianity in which European States were conceived and
nurtured, the belief that, not being a free agent, the
lunatic is not responsible for his deed; that with his
character and under his circumstances he could not
help it; that, therefore, he has not sinned before GOD,
and, consequently, should not be visited with extreme
punishment by man.* Civilly noxious, and, therefore,

* " Never by human judgement ought a man to be punished with
the pain of the lash (pana flagelli), so as to be put to death, or



216 FREE WILL

to be kept in confinement, he is still morally inno
cent, and retains the right of a man to live, a right
which no man forfeits except voluntarily and freely,
by choosing to behave like a wild beast.* This poor
lunatic still claims the benefit of the medieval axiom:
"The life of the just makes for the preservation and
promotion of the good of the community; and, there
fore, it is nowise lawful to slay the innocent. "f

But all these considerations of the old libertarianism
are swept away by modern determinism, as ruthlessly
as they were abolished by Hobbes. To any determinist,
or necessarian, a punishable murderer is not one who,
being what he was by nature and character, and tempted
as he found himself, could possibly have acted other
wise than as he did act to the slaying of his fellow-
man: he is simply a highly noxious element of society,
whose extirpation will be a good riddance, and will act
as a motive to deter similar characters from imitating
his conduct. If the determinist judge will hang any
man, let him hang this criminal lunatic.

XII

"If any one thinks that there is justice in the
infliction of purposeless suffering, that there is a natu
ral affinity between the two ideas of guilt and punish-

maimed, or beaten with [grievous] stripes, without his own fault.
But with the pain of loss (paena damn ?) one is punished even in
human judgement without fault, but not without cause." Aqui
nas, Sum. Theol. 23 zae, q. cviii, art. 4 ad 2.

* I beseech the reader to whom these ideas are strange to study
them in St Thomas, Sum. Theol. 2a zx, q. Ixiv, articles I, 2, 3, 6;
they may be read in English in dquinas Sthicus, n, pp. 39-42, 46;
cf. Ethics and Batumi Law, pp. 203, 349, 350.

t St Thomas, I.e.



JOHN STUART MILL 217

merit, which makes it intrinsically fitting that wherever
there has been guilt, pain should be inflicted by
way of retribution, I acknowledge that I can find no
argument to justify punishment inflicted on this prin
ciple. As a legitimate satisfaction to feelings of indig
nation and resentment which are on the whole salutary
and worthy of cultivation, I can in certain cases admit
it; but here it is still a means to an end. The merely
retributive view of punishment derives no justification
from the doctrine I support. But it derives quite as
little from the free-will doctrine. Suppose it true that
the will of a malefactor, when he committed an offence,
was free, or, in other words, that he acted badly, not
because he was of a bad disposition, but for no reason
in particular, it is not easy to deduce from this the con
clusion that it is just to punish him. That his acts were
beyond the command of motives might be a good
reason for keeping out of his way, or placing him
under bodily restraint; but no reason for inflicting pain
upon him, when that pain, by supposition, could not
operate as a deterring motive. While the doctrine I
advocate does not support the idea that punishment in
mere retaliation is justifiable, it at the same time fully
accounts for the general and natural sentiment of its
being so. From our earliest childhood the ideas of
doing wrong and of punishment are presented to our
mind together, and the intense character of the im
pressions causes the association between them to attain
the highest degree of closeness and intimacy. Is it
strange, or unlike the usual processes of the human
mind, that in these circumstances we should retain the
feeling, and forget the reason on which it is grounded?
But why do I speak of forgetting? In most cases the
reason has never in our early education been presented
to the mind. The only ideas presented have been those



2i8 FREE WILL

of wrong and punishment, and an inseparable associa
tion has been created between these directly, without
the help of any intervening idea. This is quite enough
to make the spontaneous feelings of mankind regard
punishment and a wrong-doer as naturally fitted to
each other as a conjunction appropriate in itself, in
dependently of any consequences."

There is no strategic advantage in marching in force
upon a position entirely removed from the seat of war.
" He acted badly for no reason in particular " is such
an irrelevant position in the present controversy: no
necessarian need attack what no libertarian holds. Of
course the malefactor acts for some reason and some
motive: of course certain reasons and motives appeal
strongly to his peculiar disposition. But every true
human act, though necessarily done on some motive,
is not done according to one motive rather than
according to another except under the conscious
superintendence and final arbitrement of the presiding
Ego. Whichever way the arbitrement goes, it goes
upon "some reason in particular": the crux of the
question is, Why upon this reason rather than upon
that? Libertarians assert that no "why" of physical
determination, like the " whys " of physical science,
is assignable; and that, therefore, "moral" science is
not, what necessarians make it, a "physical" science.
For my own part I should keep out of the way of a
man to whom it was the same thing to have a motive,
supervening upon a particular disposition, and to act
accordingly. Such is the behaviour of somnambulists, of
patients in delirium and of lunatics generally, but not



JOHN STUART MILL 219

of men in their right senses. A man in his right senses,
that is to say, a free agent, is one whose ads are not
beyond the command of motives, nor yet wholly
within the command of motives. Give a person a mo
tive, and you incline him to act; you do not compel
him. Motives, therefore, are useful means to employ,
though they do not quite act like weights in a scale.
It is advisable to lead your horse to the water, even
though you cannot make him drink. He certainly will
not drink unless he comes to water. And thus much
of deterrent punishment on the theory of free will.

Now for free will and retributive punishment. An
evil deed freely done calls for retribution. Not only
should repetition be guarded against in the future,
but the past wrong should be revenged. There should
be " sorrow dogging sin." This is the keynote of tra
gedy, to which the human heart has responded sym
pathetically in all ages.

kpaaavra Tra&tv,
rpiytpwv /tuOoc TriSe fytovti.

What is it that a man does when he commits a crime ?
He does damage, and he does wrong: his act is evil
physically and morally. The damage is to be repaired
and hindered from recurring, like any other damage.
If it is a wound inflicted, we send the patient to the
hospital, and lay restraints upon the hasty temper and
violent hand of the striker. Thus the physical evil is
corrected. But the offending person has not only done
damage, a stone may do that: he has, moreover, willed

* ^Eschylus, Qkofphori, 314.



220 FREE WILL

to do damage, freely and wantonly. He himself per
sonally was the main author of the mischief, not his
motives. Without motives he could not have done
evil: but the motives that he had were void of effect
without his sanction. Not only, therefore, shall we la
bour to readjust his motives; we shall also blame him
for having yielded to his motives as they stood, and
blaming him we shall avenge the majesty of the moral
law upon him, making him suffer for the wrong that
he has freely done.

Man starts life with much good about him, the gift
of nature and of GOD. He has as a duty of serving the
Giver by keeping the law of nature; and that law he
will discern in various measures according to circum
stances of age, place and race. He can and he ought
to help himself by the aid of his liberty to keep the
law, as he understands it, and as the observance of it
lies within his power. If he wilfully breaks the law,
not as it binds me, and as it would be atrocious fcr
me to break it, but as it binds him, and with a dis
obedience atrocious even in him, then he has entered
upon a quarrel with his Maker, in which he, the man,
is the aggressor. God essentially loves Himself and
hates whatever is opposed to Him. He cannot be op
posed but by a free agent. Ill will alone can set up
against the Almighty; and an ill will is the single ob
ject of His hate. GOD, we are told, is love. The earth and
the fullness thereof is the monument which GOD S love
has built. But if His love is so efficacious, His hatred
is not feeble. St Ignatius in his Spiritual Exercises, in the
first Exercise on sin, puts before us the case of a man,



JOHN STUART MILL 221

no matter who, "who for one mortal sin has gone to
hell ": he bids us ponder " how in sinning and acting
against Infinite Goodness such a one has been justly
damned forever." Eternal punishment is the consum
mation of retributive justice, a consummation to over
awe, but not to astonish us. We are not to be sur
prised at wilful, flagrant opposition to the Supreme
Goodness having its issue in endless evil. Holiness
and happiness are in GOD. It is not unnatural that
he who cuts himself off from holiness, should be cut
off f-om happiness ; and that impiety, if not accom
panied with misery, should at any rate end in misery.
Whoever renounces the law cannot expect to retain
the joy of the LORD. Whoever will not share GOD S
holiness, shall not share His happiness. We say com
monly that man sins and GOD punishes. We might put
it otherwise that man, so far as in him lies, casts off
GOD, and then finds himself forlorn. Punishment is
not so much the remedy as the result of sin. When
GOD leaves things to take their course, the sinner is
chastised: mercy is more of a divine interference than
justice. When the angels sinned, we are told, " their
place was no more found in heaven." * Heaven had
become for them a foreign country, a climate in which
they could not thrive, and they fell down as dead
leaves drop from the trees in autumn.

The separation of the wicked from GOD, and their
consequent destruction, are thus pictured by Plato:

GOD, as the old tradition declares, holding in His hand
the beginning, middle and end of all that is, moves according
* Apocalypse xii, 8.



222 FREE WILL

to His nature in a straight line towards the accomplishment
of His end. Justice always follows Him, and is the punisher
of those who fall short of the divine law. To that law he
who would be happy holds fast and follows it in all humi
lity and order ; but he who is lifted up with pride, or
money, or honour, or beauty, who has a soul hot with
folly and youth and insolence, and thinks that he has no
need of a guide or ruler, but is able himself to be the guide
of others, he, I say, is left deserted of GOD, KaraAfnrrrcu tpripof
Oeov ; and being thus deserted he takes to him therso who
are like himself, and dances about in wild confusion, and many
think that he is a great man, but in a short time he pays a
penalty which justice cannot but approve, and is utterly de
stroyed.*

Mill and his school enter vigorous protests against
any introduction of theology into philosophy. But
natural theology is a part of philosophy, the end and
crown of the science. To treat of crime and of pun
ishment, without reference to the Supreme Ruler and
Judge of all the earth, is impossible to a serious
theist.

Having spoken of chastisements immediately divine,
I pass to those inflicted by civil society. Is it right for
civil society to punish one of its members merely with
a view to satisfaction for a past offence, without hope
either of the reformation of the culprit or the protec
tion of society? It is not right, as I will show with all
possible brevity. To punish, one must have authority
over the delinquent. Against an equal there exists the
right of self-defence but not of punishment. The
awarding of punishments is a function of distributive
ustice, the actual exercise of which justice belongs to

* Lotus, 716, Jowett s translation.



JOHN STUART MILL 223

rulers, not to subjects.* The civil ruler must not
punish beyond the measure of his authority. That
measure is determined by the end of civil govern
ment, which end is the temporal happiness of the civil
community, that the citizens may live together in peace
and justice, with a sufficiency of wordly goods, and
with so much of moral probity as is requisite for the
outward good order and happiness of the State. f The
civil magistrate cannot punish, motu proprio, except
for this end. To the extent to which this end may
reasonably be expected to be furthered, to that extent,
and not beyond, may pains and penalties be imposed
by the civil power.

But, though the State should not punish any man
further than there is a prospect of good to mankind,
yet the punishment inflicted under this limitation is
retributive as well as corrective. A murderer should
not be hung, except where the hanging is likely to
hinder bloodshed; but when he his hung in that likeli
hood, men may well rejoice that he has got his deserts.
We may rejoice to see sin expiated by suffering,
though we should not inflict suffering on another
person without his consent, solely for the expiation of
his sins. We blame sin wherever we discover it. But,
in order to award pain as well as blame, that is, to
award punishment to a sinner, the civil magistrate
should have some prospect of amending the offender
or protecting society against him and his example. A
wicked man deserves punishment, but his fellow-men

* St Thomas Aquinas, Summa, 2a, 2X, q. Ixi, art. I ad 3.
t Suarez, De Lcgibus, \. m, c. I i, n. 7.



224 FREE WILL

are not always the persons to punish him. Where they
have the right to inflict punishment for their own
purposes, they become, at the same time, the ministers
of the vengeance of the LORD, and should consider
themselves as such.

Mill draws an argument from "the punishment of
crimes committed in obedience to a perverted con
science." He alludes by name to Ravaillac and Bal-
thasar Gerard. Men like these, he thinks, are justly
immolated to political expediency, without any regard
to the "state of mind of the offender, further than
as this may affect the efficacy of punishment as a means
to its end." I observe that laws are made to deal with
fads as they ordinarily occur. If a criminal s conscience
has been perverted, it is commonly his own fault. The
law, therefore, will not admit the plea of perversion
of conscience. It supposes guilt in the man who, being
of sound mind, does a criminal aft.

These, then, are the positions which I advance
against Mill, (i) It is unjust to punish a man, blame
and distress him, for a deed which he could not help.
(2) Even for deeds that he could help, the civil power
should never punish a man further than the good of
society requires; but within that limit he should be
punished as well as in retribution for the past as in
precaution against the future. (3) GOD does justly
punish in certain cases by way of mere retaliation for
the wrong done to His Divine Majesty.

One word in conclusion on the educational bearing
of this discussion, as our pedagogists now insist on
psychology for teachers. Let free will be an article of



JOHN STUART MILL 225

the teacher s psychological creed, if not on the higher
ground of truth, then on the lower ground of utili
tarian and pragmatic expediency. Where free will is
denied, punish as we may, the training of the young
in virtue will prove no easy task. When a child is
punished, unless he confess at heart that he deserves
it for his own waywardness and wilfulness, the pun
ishment will not appear to him in any moral light
but as a mere odious infliction. If there is no self-
reproach, no iteration of the rod will ever lead a delin
quent to think that wrong-doing is wrong and ought
to be punished. His thought will be that, unfortu
nately, it is hard to do forbidden things and escape
scot-free, a widely different conclusion. But self-
reproach brings in the consciousness of free will: we
do not reproach ourselves for what we think we could
not help. Evil, inevitable under the circumstances, is
a matter of pure compassion.

XIII

" Suppose that there were two peculiar breeds of
human beings, one of them so constituted from
the beginning that, however educated or treated, no
thing could prevent them from always feeling and
acting so as to be a blessing to all whom they ap
proached; another, of such original perversity of na
ture that neither education nor punishment could
inspire them with a feeling of duty, or prevent them
from being active in evil-doing. Neither of these races
of human beings would have free will; yet the former
would be honoured as demigods, while the latter would
be regarded and treated as noxious beasts: not punished

15



226 FREE WILL

perhaps, since punishment would have no effect upon
them, and it might be thought wrong to indulge the
mere instinct of vengeance: but kept carefully at a
distance, and killed like other dangerous creatures
when there was no other convenient way of being
rid of them. We thus see that even under the utmost
possible exaggeration of the doctrine of necessity the
distinction between moral good and evil in conduct
would not only subsist, but would stand out in a more
marked manner than now, when the good and the
wicked, however unlike, are still regarded as of one
common nature."

If we consider a man s act apart from the manner
in which it is elicited, in other words, if we abstract
from free will and determinism, neither affirming nor
denying either, no doubt a distinction between good
and evil actions might still be kept up. Nay, even under
the denial of free will, the action which we now know
as good remains good, and the action which we know
as evil remains evil. But under such denial we should
regard the doers of such actions in quite a different light
from that in which we view good men and bad men
now. On this point I have written elsewhere:

And first of what remains of our present moral system,
when it comes to be worked on determinist principles. The
Ten Commandments remain unchanged. The list of vices and
virtues remains unchanged. The ethical motives for virtue and
against vice remain unchanged. The State continues to frame
laws, commanding and forbidding the same things as before.
The same conduct is praised and rewarded, or blamed and
punished, as before, albeit not quite with the same intention.
The portraits of the good man and of the bad man respectively
have lost none of their external lineaments. The one is still



JOHN STUART MILL 227

self-controlled and self-denying, brave, loving, magnanimous
and just. The other remains a sensualist, cruel and cowardly,
frivolous, idle, heartless and untrustworthy. Nero is still bad,
and Paul good. The exigencies of human nature and of hu
man society have not lost their value. The good and happi
ness of the individual, and the prosperity of the society to
which he belongs, require of him the same conduct as before.
Goodness has not become less profitable, nor wickedness less
detrimental and deplorable, now that both are recognised
necessities. Wickedness is what it was in every respect save
one; and the same deeds are wicked that were wicked. Good
ness has lost only one of its attributes. Formerly the good man
did what it befitted a man to do, having at the same time in
the very act and circumstances of his well-doing the power to
swerve from goodness: still he does the same things, but fur
ther it is to be noted that, with his charadler and circumstances,
he cannot help doing them. And conversely of the wicked man,
who is rightly enough pronounced by the determinist a dan
gerous, disgusting and offensive animal. Ugly conduct fits in
with the determinist hypothesis as well as ugly architecture.
We praise a flower, or a gun, or the "points" of a horse.
There would be no difficulty in praising in that way a man
in whose conduit we recognised no free will. Still he might
be to us a grand fellow, a very useful creature. We might
further encourage him with prospective praise, as an induce
ment to serve us still better, much in the same way that a
driver pats his horse and utters kind cries to it on a hard
road. Such praise, however, and the corresponding blame,
cannot be called moral approbation and reprobation.*

There are two cases conceivable in which no edu
cation nor treatment could prevent a human being
from going about doing good in this world. The
first would be the case of a being too unsusceptible of
education, too stupidly insensible of the treatment he
received, to be diverted from gratifying a blind incli-
* Political and Moral Essays, pp. 253-255.



228 FREE WILL

nation that he had to make himself agreeable and pro
fitable to others, a being that would exercise among
men a genial and healthy influence, as a tench is said
to do amongst fishes, without understanding. Such a
being, though useful, would not be morally good, nor
would his utility be of the highest order; indeed he
would be scarcely human. Secondly, we may conceive
a man, possessed of such a lively and ever actual in
sight into the paramount excellence of doing good,
that he would no more think of failing to do good in
seasonable circumstances, notwithstanding any per
verse training or harsh treatment that he might have
undergone, than we should think of cutting off our
heads to appease our hunger. This man s will would
not be free to turn away from doing good. At the
same time he would be a moral agent, as distinguished
from a physical one, for he would act with an appre
ciation of what he did. His would be a case of an in
tellectual necessity, similar to the necessity under
which GOD and the angels and saints are of being holy,
from seeing the clear vision of the beauty of holiness.
Brute necessity, on the other hand, is the state of an
agent that must act, without knowing what it does.
Brute necessity is incompatible with either moral ex
cellence or turpitude: intellectual necessity is incom
patible with moral evil, but quite compatible with
moral good.

A human being, lying under a brute necessity of
evil doing, may be discussed in three shapes. In the
first place, he may do evil from stupidity, not mean
ing what he does: but then it is no moral evil. Mill



JOHN STUART MILL 229

seems to have wished to exclude this case by his phrase
"adive in evil-doing." Secondly, we may speak of
"an original perversity of nature," that intends known
evil with a resistless necessity from the first. Such a
being would not be of sound mind: would be what is
called a " criminal lunatic " ; and his aclions, horrible ot
themselves, would not be morally evil in him. Lastly,
we may conceive human beings with their wills set in
wickedness, whence no motive can convert them, not,
however, created in this state, but having come to it
by their own abuse of their free will. Such men would
be in the state and condition of devils. Of the devils
St Thomas writes, "The evil angels sin mortally in all
things whatsoever they do of their own will,"* and
he assigns as a reason this property of angelic nature,
that when once an angel takes a decisive resolution,
his will becomes eternally fixed in the same: "The
free will of man is flexible one way and another both
before and after election; whereas the free will of an
angel is flexible before election, but not after."f Such
a life, however, we may venture to think, is not studded
and diversified with a multitude of distinct sins, but
is one long sin, the beginning of which was a free acl,
albeit the continuance is a necessity. We see some
approximation to this state in a confirmed habit of vice
even in a man on earth.

* Sumrtta, la zx, q. Ixxxix, art. 4. t la, q. Ixiv, art. 2.



230 FREE WILL

XIV

" Real fatalism is of two kinds. Pure, or Asiatic
fatalism, the fatalism of the QEdipus, holds that our
actions do not depend upon our desires. Whatever our
wishes may be, a superior power, or an abstract destiny,
will overrule them, and compel us to ad, not as we de
sire, but in the manner predestined. Our love of good
and hatredof evil are of no efficacy, and thoughinthem-
selves they may be virtuous, as far as conduct is con
cerned, it is unavailing to cultivate them. The other
kind, modified fatalism I will call it, holds that our
actions are determined by our will, and our will by our
desires, and our desires by the joint influence of the
motives presented to us and of our individual cha
racter; but that, our character having been made for
us and not by us, we are not responsible for it, nor for
the actions it leads to, and should in vain attempt to
alter them. The true doctrine of the Causation of hu
man actions maintains, in opposition to both, that not
only our conduct, but our character, is in part amen
able to our will; that we can, by employing the proper
means, improve our character, and that if our character
is such that while it remains what it is, it necessitates
us to do wrong, it will be just to apply motives which
will necessitate us to strive for its improvement, and
so emancipate ourselves from the other necessity: in
other words, we are under a moral obligation to seek
the improvement of our moral character. . . When we
voluntarily exert ourselves, as it is our duty to do, for
the improvement of our character, or when we act in
a manner which, either consciously on our part or
unconsciously, deteriorates it, these, like all other
voluntary acts, presuppose that there was already
something in our character, or in that combined with



JOHN STUART MILL 231

our circumstances, which led us to do so, and accounts
for our doing so."

The interest of this interesting passage, with which
the previous passage, n. 5, the reply to the Owenite,
should be compared, lies in the affirmation that "not
only our conduct, but our character, is in part ame
nable to our will." How so? That is the question the
answer to which should light up Mill s whole position,
and reveal the gulf, if any gulf there be, between him
and that " Modified Fatalism " which he reprobates.
This then is his reply or replies:

R. i. " We can, by employing the proper means,
improve our character."

The reply merits all praise from the libertarian
point of view. It is quite true. But how is it consis
tent with " the true doctrine of the Causation of hu
man actions," as laid down by Mill? To get that doubt
solved, we are obliged to ask for an explanation of the
terms of the reply received. What are the " proper
means" by employing which we can improve our
character? Mill replies as follows:

R. 2. "The proper means for improving our cha
racter are our own voluntary exertions."

This reply is gathered from the last sentence in the
extract quoted: "When we voluntarily exert our
selves, as it is our duty to do, for the improvement of
our character," etc. Again, an excellent reply, and, as
it stands, conceived quite in a libertarian spirit. But
this libertarian spirit Mill hastens to exorcise and cast
out. For the sentence goes on: " These [voluntary ex
ertions], like all other voluntary acts, presuppose that



232 FREE WILL

there was already something in our character, or in that
combined with our circumstances, which led us to do
so, and accounts for our doing so." Alas, alas, here we
are back in the squirrel s cage, the vicious circle, from
which Mill seems impotent to escape. By a singular
method, which we may call Roundabout Fatalism, he
derives our volitions from our character and circum
stances, our character from our volitions and circum
stances (one most important circumstance being no
doubt that of heredity), and those volitions again from
our character and circumstances. Then, except our
character and circumstances cause and determine us so
to do, we shall make no voluntary efforts for the im
provement of our character. In what does this account
differ from the "Modified Fatalism" which represents
" our character having been made for us and not
by us"?

The only way to strike a difference, and it is a direct
and very true way, is by saying that although character
and circumstances must concur to induce us to make
voluntary exertions to improve our character, for we
can do nothing without motives, and motives suffer a
sort of refraction in the character upon which they im
pinge; nevertheless, it rests with us finally, having the
motive for voluntary exertion, to ad upon it or to let
it drop void and ineffectual, and this is an alternative
ultimately ruled, not by motive and character, but by
our own personal self. But this a statement of free will;
the direct contrary of that doctrine of the "causation
of human actions," i.e., their physical determination,
which Mill maintains.



JOHN STUART MILL 233

"We are under a moral obligation," writes Mill,
"to seek the improvement of our moral character."
This may be accomplished by "voluntary exertions";
but, as we have seen, those voluntary exertions are
determined by that very character which needs them
for its improvement. Should the character not be re
sponsive to the need, Mill provides another remedy.
"If our character is such that while it remains what
it is, it necessitates us to do wrong, it will be just to
apply motives which will necessitate us to strive for
its improvement"; which means that we may be justly
punished to set us on the way of reform. Here is a
duty which we cannot do until we are punished for
not doing it; a moral obligation, the fulfilment of which
rests with the strong arm of the law that grips us.
Awaiting that, we lie like over-turned motor cars,
helpless for all good on the wayside of life. It is diffi
cult to see any moral obligation in such a position of
necessity.

Disagreeing with Mill in many things, I have never
ceased to cherish for him a certain admiration. Since
I first opened his pages, nearly forty years ago, I have
everadmired his clear, incisive thought, his logical acu
men, and his candour, shining out, as it often does, at the
expense of his consistency. He is too ingenuous, too
adverse to fatalism, too great a lover of individualism
and liberty, to be a thorough determinist. In all this
he forms a strong contrast to Hobbes, who drives his
necessarianism, as he drives every other point of his
grim philosophy, steadily and remorselessly to the
final conclusion.



234 FREE WILL

The difference between determinism and fatalism is
not so much in theory as in practice. The fatalist acts
upon his theory, and either sits idle in the absence of
strong emotion, or surrenders himself to the impulse
in which he recognises his destiny. The determinist,
in England at least, shuts his determinism up with his
books; and, in active life, uses his free will vigorously.
Whatever academicians may say, an illogical escape into
the realms of truth is preferable to detention in the
logical bonds of error once entered upon. Thus esca
ping, on the whole we prosper in England, notwith
standing much bad philosophy.