6 Morality as freedom

Elevating though man's privilege is, of being capable of such an idea as

freedom of choice - [those who are accustomed only to physiological expla-
nations] are stirred up by the proud claims of speculative reason, which feels
its power so strongly in other fields. They are stirred up just as if they were
allies, leagued in defense of the omnipotence of theoretical reason and
roused by a general call to arms to resist the idea of freedom of choice and
thus at present, and perhaps for a long time to come (though ultimately in
vain), to attack the moral concept of freedom and, if possible, render it
suspect.
Immanuel Kant (MPV 378)
Kantian ethical philosophy has often been criticized for its depen-
dence on an untenable conception of the freedom of the will. Kant
is supposed to have asserted that we are morally responsible for all
of our actions because we have free will, and that we have free will
because we exist in a noumenal world in which we are uninflu-
enced by the temptations of desire and inclination. If we existed
only in the noumenal world, we would invariably act as the cate-
gorical imperative requires, but because we are also phenomenal
beings we sometimes go wrong. The view so understood gives rise
to several problems. First, the claim that purely noumenal persons
would act as the categorical imperative requires may be questioned.
It is not obvious why persons uninfluenced by causality should act
morally rather than any other way. Secondly, if it can be established
that insofar as we are noumena we obey the moral law, then the ac-
count of moral imputability becomes unintelligible. If we are only
responsible because we are noumena and if insofar as we are nou-
mena we only do what is right, then we cannot be responsible for
our evil actions. Or, if we are responsible, it is so radically that no

159

https://doi.org/10.1017/CBO9781139174503.007 Published online by Cambridge University Press

 160 RANT'S MORAL PHILOSOPHY

room is left for excuses. For how can we take into account the
terrible temptations to which the wrongdoer was subjected, when
the choosing noumenon was uninfluenced by those temptations?
Finally, the view seems to require an unappealing ontological com-
mitment to the existence of "two worlds/' and to give rise to a
variety of puzzles about how what occurs in the one can influence
the other.
In this paper my aim is to address these problems. In the first
part of the paper, I show why Kant thinks that the moral law is the
law of a free will, and why he thinks we must regard ourselves as
free. I then argue that the supposed problems about responsibility
and ontology arise from a common source: a failure to appreciate
the radical nature of Kant's separation of theoretical and practical
reason, and of their respective domains of explanation and delibera-
tion. When these domains are separated in the way that Kant's
philosophy requires, the problems about responsibility disappear,
and we see that Kant's theory of freedom does not commit him to
an ontological dualism.1 In the second part of the paper I show
what it does commit him to: a certain conception of the moral
virtues.

I LAW AS FREEDOM

i. Freedom enters Kant's moral philosophy as the solution to a prob-

lem. The categorical imperative is not analytic, and disregarding its
claims is therefore not inconsistent. Yet it is supposed to present us
with a rational necessity. In order to show that morality is not a
"mere phantom of the mind" (G 445), Kant seeks to provide a deduc-
tion of (or a credential for)2 the moral law: he must link being ra-
tional to acting on the moral law. The third idea through which
rationality and morality are linked is the positive conception of
freedom. By showing, first, that a free person as such follows the
moral law, and, second, that a rational person has grounds for regard-
ing herself as free, Kant tries to show that insofar as we are rational,
we will obey the moral law.
It was making the second of these two connections that trou-
bled Kant - the connection between rationality and freedom. The
arguments intended to demonstrate this connection in the Ground-

https://doi.org/10.1017/CBO9781139174503.007 Published online by Cambridge University Press

 Morality as freedom 161
work of the Metaphysics of Morals and in the Critique of Practical
Reason are obscure and appear to be different from one another. In
Groundwork III, Kant calls his argument a "deduction" of the
moral law (G 454), and connects freedom and reason through the
capacity of reason for pure spontaneous activity which is exhibited
in its production of ideas. This spontaneous activity shows we are
members of the intelligible world and therefore free (G 452). In the
Critique of Practical Reason, we are instead offered what Kant
calls a "credential" for morality (C2 48) and told that "the objec-
tive reality of the moral law can be proved through no deduction"
(C2 47). The credential is provided by the fact that freedom can be
deduced from morality. Kant does not comment on the difference
between these two arguments, and his readers do not agree about
whether they come to the same thing, or are different arguments
serving different purposes, or are incompatible arguments resulting
from a change of mind.3
But Kant was not in doubt about his success in making the first
connection, between morality and freedom. Kant was confident that
"if freedom of the will is presupposed, morality together with its
principle follows from it by the mere analysis of its concept" (G 447).
In Groundwork III, the argument for this point takes about a page; in
the second Critique, it is a mere paragraph, posed as
Problem II
Granted that a will is free, find the law which alone is competent to deter-
mine it necessarily.
Since the material of the practical law, i.e., an object of the maxim, can-
not be given except empirically, and since a free will must be independent of
all empirical conditions (i.e., those belonging to the world of sense) and yet
be determinable, a free will must find its ground of determination in the
law, but independently of the material of the law. But besides the latter there
is nothing in a law except the legislative form. Therefore, the legislative
form, in so far as it is contained in the maxim, is the only thing which can
constitute a determining ground of the [free] will. (C2 29)
Not everyone has found this connection so perspicuous. In his well-
known appendix to The Methods of Ethics,* Sidgwick complains
that Kant's whole moral philosophy is vitiated by a confusion be-
tween two senses of "freedom." "Moral or neutral" freedom is the
freedom we exercise when we choose between good and evil. "Good

https://doi.org/10.1017/CBO9781139174503.007 Published online by Cambridge University Press

 162 RANT'S MORAL PHILOSOPHY

or rational" freedom is the freedom we exercise when we act mor-

ally, and so are not "enslaved" by our passions and desires. Sidgwick
accuses Kant of being unaware of the distinction. This accusation is
unfair, for the distinction Sidgwick makes is closely related to
Kant's own distinction between negative and positive freedom. As
we shall see, Kant rejects moral or neutral freedom as a conception
of freedom; but it is a consequence of negative freedom, or the ab-
sence of all determination.
We may put Kant's reply to Sidgwick in these terms. Following
John Rawls, we may distinguish the concept of X, formally or func-
tionally defined, from a conception of X, materially and substan-
tively defined.5 The Kantian concept of free will would be "a will
which makes choices independently of all alien influences/7 that is,
a will which is negatively free. A positive conception of freedom
would be a material account of what such a will would in fact
choose. Kant's reply to Sidgwick will then be that there is a single
concept of freedom, of which the moral law is the unique positive
conception. My aim in the next section is to explain Kant's claim
that the moral law is the unique positive conception of freedom.

2 Kant argues that when you make a choice you must act "under
the idea of freedom" (G 448). He explains that "we cannot conceive
of a reason which consciously responds to a bidding from the outside
with respect to its judgments" (G 448). You may of course choose to
act on a desire, but insofar as you take the act to be yours, you think
you have made it your maxim to act on this desire. If you feel that
the desire impelled you into the act, you do not regard the act as a
product of your will, but as involuntary. The point is not that you
must believe that you are free, but that you must choose as if you
were free. It is important to see that this is quite consistent with
believing yourself to be fully determined. To make it vivid, imagine
that you are participating in a scientific experiment, and you know
that today your every move is programmed by an electronic device
implanted in your brain. The device is not going to bypass your
thought processes, however, and make you move mechanically, but
rather to work through them: it will determine what you think.
Perhaps you get up and decide to spend the morning working. You no
sooner make the decision than it occurs to you that it must have

https://doi.org/10.1017/CBO9781139174503.007 Published online by Cambridge University Press

 Morality as freedom 163
been programmed. We may imagine that in a spirit of rebellion you
then decide to skip work and go shopping. And then it occurs to you
that that must have been programmed. The important point here is
that efforts to second guess the device cannot help you decide what
to do. They can only prevent you from making any decision. In order
to do anything, you must simply ignore the fact that you are pro-
grammed, and decide what to do - just as if you were free. You will
believe that your decision is a sham, but it makes no difference.6
Kant's point, then, is not about a theoretical assumption necessary
to decision, but about a fundamental feature of the standpoint from
which decisions are made.? It follows from this feature that we must
regard our decisions as springing ultimately from principles that we
have chosen, and justifiable by those principles. We must regard
ourselves as having free will.
Kant defines a free will as a rational causality that is effective
without being determined by an alien cause. Anything outside of the
will counts as an alien cause, including the desires and inclinations
of the person. The free will must be entirely self-determining. Yet,
because it is a causality, it must act on some law or other. " Since the
concept of a causality entails that of laws . . . it follows that freedom
is by no means lawless" (G 446). The free will therefore must have
its own law. Alternatively, we may say that since the will is practical
reason, it cannot be conceived as acting and choosing for no reason.
Since reasons are derived from principles, the free will must have its
own principle. Kant thinks that the categorical imperative is the free
will's law or principle. But it may seem unclear why this more than
anything else should be the free will's principle. If it is free to make
its own law, why can't it make any law whatever?
To see why, imagine an attempt to discover the freely adopted
principle on which some action is based. I ask to know why you are
doing some ordinary thing, and you give me your proximate reason,
your immediate end. I then ask why you want that, and most likely
you mention some larger end or project. I can press on, demanding
your reason at every step, until we reach the moment when you are
out of answers. You have shown that your action is calculated to
assist you in achieving what you think is desirable on the whole,
what you have determined that you want most.
The reasons that you have given can be cast in the form of maxims

https://doi.org/10.1017/CBO9781139174503.007 Published online by Cambridge University Press

 164 KANT'S MORAL PHILOSOPHY

derived from imperatives. From a string of hypothetical imperatives,

technical and pragmatic (G 416-17), you have derived a maxim to
which we can give the abbreviated formulation:
I will do this action, in order to get what I desire.
According to Kant, this maxim only determines your will if you
have adopted another maxim that makes it your end to get what you
desire. This maxim is:
I will make it my end to have the things that I desire.
Now suppose that I want to know why you have adopted this
maxim. Why should you try to satisfy your desires?
There are two answers which we can dismiss immediately. First,
suppose you appeal to a psychological law of nature that runs some-
thing like "a human being necessarily pursues the things he or she
desires." 8 To appeal to this causal law as an answer would be to deny
your freedom and to deny that you are acting under the idea of
freedom. The answer does not have the structure of reason-giving: it
is a way of saying "I can't help it." Second, suppose you claim that
you have adopted this maxim randomly. There is nothing further to
say. You think you could have adopted some other maxim, since you
regard your will as free, but as it happened you picked this one. As
we know, Kant rejects this, as being inconsistent with the very idea
of a will, which does what it does according to a law, or for a reason.
It seems as if the will must choose its principle for a reason and so
always on the basis of some more ultimate principle.
We are here confronted with a deep problem of a familiar kind. If
you can give a reason, you have derived it from some more funda-
mental maxim, and I can ask why you have adopted that one. If you
cannot, it looks as if your principle was randomly selected. Obvi-
ously, to put an end to a regress like this we need a principle about
which it is impossible, unnecessary, or incoherent to ask why a free
person would have chosen it. Kant's argument must show that the
categorical imperative has this status.
Although Kant does not think that a free will exists in time, we
may imagine that there is a ''moment" when the free will is called
upon to choose its most fundamental principle. In order to be a will,
it must have a principle from which it will derive its reasons. The
principle it chooses will determine what it counts as a reason. But

https://doi.org/10.1017/CBO9781139174503.007 Published online by Cambridge University Press

 Morality as freedom 165
precisely because at this "moment" the will has not yet determined
what it will count as a reason, it seems as if there could be no reason
for it to choose one principle rather than another. Kant calls this
feature of the will its "spontaneity."9
As the argument stands now, it looks as if the will could adopt any
maxim we can construct. If you have a free will you could adopt a
maxim of pursuing only those things to which you have an aversion,
or perhaps all and only the things your next-door neighbor enjoys. For
us human beings, however, these are not serious options, for reasons
that come out most clearly in Religion Within the Limits of Reason
Alone. Kant uses the term "incentive" (Thebfeder) to describe the
relation of the free person to the candidate reasons among which she
chooses. An incentive is something that makes an action interesting
to you, that makes it a live option. Desires and inclinations are incen-
tives; so is respect for the moral law. An inclination by itself is merely
an incentive, and does not become a reason for action until the person
has adopted it freely into her maxim (R 23-24; 44). Although incen-
tives do not yet provide reasons for the spontaneous will, they do
determine what the options are - which things, so to speak, are candi-
dates for reasons. And having an aversion to something is not, for us
human beings, an incentive for pursuing it, and so will not become a
reason. In the Religion, Kant claims that it is impossible for a human
being not to be moved at all by incentives; our freedom, rather, is
exercised in choosing the order of precedence among the different
kinds of incentives to which we are subject (R 30; 36). So the real
choice will be between a maxim of self-love, which subordinates the
incentives of morality to those of inclination, and the moral maxim,
which subordinates incentives of inclination to moral ones. The
maxim of self-love says something like:
I will do what I desire, and what is morally
required if it doesn't interfere with my self-love.
and the moral maxim says something like:
I will do what is morally required, and what
I desire if it doesn't interfere with my duty.
More specifically stated, of course, the moral maxim is the maxim
derived from the categorical imperative:
I will act only on a maxim that I can will as a universal law.

https://doi.org/10.1017/CBO9781139174503.007 Published online by Cambridge University Press

 166 RANT'S MORAL PHILOSOPHY

It looks at first as if the problem here is to show that there is some

reason for the spontaneous will to choose the moral maxim rather
than the maxim of self-love. Yet this seems impossible, since the
spontaneous will by hypothesis has not yet determined what it
counts as a reason. But on reflection we shall see that this problem
can be circumvented. We need only consider the standpoint of the
spontaneous will, and the content of the categorical imperative.
At the standpoint of spontaneity, the will must, in order so to speak
to commence operations, choose a principle or a law for itself. Noth-
ing provides any content for that law. All that it has to be is a law.
Suppose that it chooses the categorical imperative, as represented
by the Formula of Universal Law. This formula merely tells us to
choose a law. Its only constraint on our choice is that it have the
form of a law. Nothing provides any content for that law. All that it
has to be is a law.
By making the Formula of Universal Law its principle, the free
will retains the position of spontaneity. Or, to put it a better way, the
argument shows that the free will need do nothing to make the
Formula of Universal Law its principle: it is already its principle.
The categorical imperative is thus shown to be the law of spontane-
ity. In a sense, the Formula of Universal Law simply describes the
function or task of an autonomous will. The moral law does not
impose a constraint on the will; it merely says what it has to do in
order to be an autonomous will at all. It has to choose a law.
On the other hand, suppose the will chooses the maxim of self-
love. In that case, it departs from its position of spontaneity and puts
itself in the service of inclination. A constraint on its choice is
acquired. The important thing to see is that there is no incentive for
the spontaneous will to do this. Since we are just talking about the
will itself right now, and not the whole person, the incentives of
inclination cannot provide a temptation to adopt the maxim of self-
love. Incentives of inclination cannot move the will to abandon its
position of spontaneity, since they cannot move the will at all until
it has already abandoned that position by resolving to be moved by
them.
This argument, which I will call the Argument from Spontaneity,
shows that there are not really two choices, morality and self-love,
on an equal footing. The will that makes the categorical imperative
its law merely reaffirms its independence of everything except law

https://doi.org/10.1017/CBO9781139174503.007 Published online by Cambridge University Press

 Morality as freedom 167
in general. Its dependence on law in general is not a constraint, for
that is just a consequence of the fact that it is a will. Making the
categorical imperative its principle does not require the spontaneous
will to take an action-it is already its principle. Adopting the
maxim of self-love is surrendering the position of spontaneity, and
does require an action (R 31-32). And it is an action for which there
could be no reason. Thus, not only are the two options not on a
footing, but the choice of the maxim of self-love over that of moral-
ity is unintelligible. Morality is the natural condition of a free will.
The free will that puts inclination above morality sacrifices its free-
dom for nothing.

3. A crucial point in the Argument from Spontaneity is that the

spontaneous will is not tempted by incentives of inclination. Now,
we human beings are not so situated with respect to the incentives
of inclination, because we are imperfectly rational beings. Or rather,
this is what makes us imperfectly rational beings. Our inclinations
may be alien to our purely rational wills, but they are not alien to us,
and they do tempt us. Letting our wills serve our happiness therefore
does not seem pointless to us. So although the Argument from Spon-
taneity explains why a purely rational will would have the moral
law as its first principle, it does not show us exactly why we should
do so. In Kant's language, it does not explain "the interest attaching
to the ideas of morality" (G 448).
Without an account of moral interest, Kant complains, there will
be a circle in our explanation of moral obligation (G 449-50). Now,
what exactly this circle is is rather difficult to see. Kant has already
claimed that, as creatures who must act under the idea of freedom,
we are bound by the laws of freedom (G 448). But he thinks this does
not yet explain how "the worth we ascribe" to moral actions (G 449)
can so completely outweigh the worth of our condition - that is, our
happiness or unhappiness. We are willing to grant the importance of
the autonomy we express in moral conduct only because we already
think that morality is supremely important. But it is still unclear
why we think so. What is needed is an incentive for us to identify
with the free and rational side of our nature. To provide this, Kant
introduces the distinction between the intelligible and sensible
worlds, or noumena and phenomena.10 This distinction introduces
two new elements into the argument.

https://doi.org/10.1017/CBO9781139174503.007 Published online by Cambridge University Press

 168 RANT'S MORAL PHILOSOPHY

The first element is the emphasis on complete causal determina-

tion in the phenomenal world. Up until now, I have spoken of the
will that adopts the maxim of self-love as adopting an unnecessary
constraint. But the addition of the two-worlds picture makes the
consequence of adopting the maxim of self-love look even worse.
The will that adopts self-love as its maxim is determined by inclina-
tions, and inclinations, in the world of phenomena, are completely
determined by natural forces, by the nexus of causal laws. So such a
will becomes a mere conduit for natural forces. The person who acts
from self-love is in a sense not actively willing at all, but simply
allowing herself to be controlled by the passive part of her nature,
which in turn is controlled by all of nature. From the perspective of
the noumenal world, ends we adopt under the influence of inclina-
tion rather than morality do not even seem to be our own.
The other element is introduced with the claim that "the intelligi-
ble world contains the ground of the sensible world and hence of its
laws" (G 453).11 Although we can know nothing of the noumenal
world, it is what we conceive as lying behind the phenomenal world
and giving that world its character. To conceive yourself as a member
of the noumenal world is therefore to conceive yourself as among the
grounds of the world as we know it.12 And if you hold this position in
so far as you have a will, then that means that the actions of your will
make a real difference to the way the phenomenal world is.
Combining these two new elements we can generate a very stark
contrast between choosing the maxim of morality and choosing
that of self-love. We can think of the noumenal world as containing
our own wills and whatever else forms part of "the ground of the
sensible world and its laws/' In particular, the noumenal world
contains the ground, whatever it might be, of the laws of nature
(for these are not objects of our wills).^ We can influence the phe-
nomenal world, and these other forces do so as well. Of course,
nothing can be known about the nature of this influence or its
mechanisms, or of how these various agencies together generate
the world of appearances. But we can still say this: if by choosing
the maxim of self-love you allow the laws of nature to determine
your actions, then you are in effect surrendering your place among
"the grounds of the sensible world and its laws/7 The existence of
your will in the noumenal world makes no difference to the char-
acter of the phenomenal world. For your will is determined by the

https://doi.org/10.1017/CBO9781139174503.007 Published online by Cambridge University Press

 Morality as freedom 169
laws of nature, and those in turn can be accounted for by other
forces in the noumenal world. Although you are free, you could
just as well not have been. Your freedom makes no difference. But
if you will in accordance with the moral law, you do make a differ-
ence. You actually contribute - we might say to the rational, as
opposed to the merely natural, ordering of the sensible world. The
choice of the moral maxim over the maxim of self-love may then
be seen as a choice of genuine activity over passivity; a choice to
use your active powers to make a difference in the world.
Recall that all of this is supposed to solve the problem of moral
interest. Kant thinks of the idea of our intelligible existence as be-
ing, roughly speaking, the motivating thought of morality, and so
what makes morality possible. In the Religion, Kant tells us that one
who honors the moral law cannot avoid thinking about what sort of
world he would create under the guidance of practical reason, and
that the answer is determined by the moral idea of the Highest Good
(R 5). In the second Critique, Kant says in one place that our intelligi-
ble existence gives us a "higher vocation" (C2 98). This vocation is
to help to make the world a rational place, by contributing to the
production of the Highest Good.1*
This argument also explains why Kant thinks that unless the High-
est Good is possible the moral law is "fantastic, directed to empty
imaginary ends, and consequently inherently false" (C2 114). The
difficulty arises in this way. We have explained moral interest in
terms of a stark contrast between being a mere conduit for natural
forces on the one hand and making a real difference in the world
through one's intentions on the other. But in between these two
possibilities we discover a third - that our intentions and actions
will make a real difference in the world, but that we will have no
control over what sort of difference they make - because the conse-
quences of our actions will not be what we intend. This can happen
because we are not the only elements of the noumenal world and the
various forces it contains combine, in ways we cannot comprehend,
to generate the world of appearances. The forces of nature and the
actions of other persons mediate between our intentions and the
actual results of our actions, often distorting or perverting those
results. This possibility then makes the appeal of freedom seem like
a fraud. If the motivating thought of morality is that freedom means
that we can make a difference in the world, but we then find that we

https://doi.org/10.1017/CBO9781139174503.007 Published online by Cambridge University Press

 I7O KANT'S MORAL PHILOSOPHY

have no control over the form this difference ultimately takes, then
the motivating thought is genuinely threatened. Postulating God as
the author of the laws of nature is a way of guaranteeing that other
noumenal forces will cooperate with our good intentions, and leaves
our moral interest in place. In the Groundwork, Kant says:
the idea of a pure intelligible world as a whole of all intelligences to which
we ourselves belong as rational beings . . . is always a useful and permissible
idea for the purpose of a rational faith. This is so even though all knowledge
terminates at its boundary, for through the glorious idea of a universal realm
of end-in-themselves (rational beings) a lively interest in the moral law can
be awakened in us. (G 462)

The Two-Worlds Argument is worked out better in the Critique of

Practical Reason than in Groundwork III. In Groundwork III, Kant
wants to argue that the idea of our existence in the intelligible world
suggests our freedom to us: our capacity for pure spontaneous activ-
ity, which reveals itself in reason's production of ideas, makes us
members of the intelligible world. As such we may regard ourselves
as free. In the second Critique, Kant develops the reverse argument
that freedom leads us to the conception of our existence in the
intelligible world. It is morality, in turn, that teaches us that we are
free. So morality itself "points" us to the intelligible world (C2 44).
The argument of the Critique of Practical Reason is superior be-
cause freedom requires not just that we exist in the intelligible
world, but that we exist there insofar as we have wills - that we can
be motivated from there, so to speak. The Groundwork argument
places our theoretical capacity to formulate pure ideas in the intelli-
gible world, but that by itself does not imply that we can be moved
by them.1* And the latter is what the argument must show. The
second Critique argument starts firmly from the fact that we can be
motivated by pure ideas. That we can be so motivated is what Kant
calls the Fact of Reason.
But the function of the idea of our intelligible existence as an
incentive is essentially the same in both books. The famous address
to Duty in the Critique of Practical Reason, like Groundwork HI,
demands to know the source of the special worth we assign to moral-
ity (C2 86-87). And the answer is again that respect for the moral
law is produced by the thought of our intelligible nature. Kant says
that the incentive of pure practical reason is

https://doi.org/10.1017/CBO9781139174503.007 Published online by Cambridge University Press

 Morality as freedom 171
nothing else than the pure moral law itself, so far as it lets us perceive the
sublimity of our own supersensuous existence. (C2 89)16
The Argument from Spontaneity shows why a free and spontaneous
will, uninfluenced by anything, makes the moral law its principle.
The Argument from the Two Worlds shows us why we imperfectly
rational beings, influenced by sensibility as well as morality, should
do so as well. If we are free we are members of the intelligible world,
the ground of the sensible world and its laws. This gives us a "higher
vocation" than the satisfaction of our own desires. We can help to
bring about the Highest Good in the world. The thought of that
higher vocation is the motive of morality.

4. But the result of the Argument from Spontaneity may seem too
strong. If the will is free, moral evil is unintelligible, for if this
argument is correct, moral evil is the pure will's wholly unmoti-
vated abandonment of its freedom. However, this is exactly Kant's
view: evil is unintelligible. Neither a good will nor an evil will
admits of explanation, for both must be regarded as grounded in the
person's own free and spontaneous choice. If these choices could be
explained, they would be derived from something else, and then
they would not be the spontaneous choices that they purport to be
(R 21). Yet it is evil that is unintelligible, for it is in the evil choice
that the will falls away from its freedom. Kant says:
Evil could have sprung only from the morally-evil (not from mere limita-
tions in our nature); and yet the original predisposition is a predisposition to
good; there is then for us no conceivable ground from which the moral evil
in us could originally have come. (R 43)
Moral evil is a Fall, in the Biblical sense, and it is exactly as hard to
understand as the Fall in the Bible (R 19,- 4iff.).
In fact, Kant goes so far as to deny that what Sidgwick calls moral
or neutral freedom, the freedom to choose between good and evil, is
really a conception of freedom at all:
freedom can never be located in the fact that the rational subject is able to
make a choice in opposition to his (legislative) reason, even though experi-
ence proves often enough that this does happen (though we cannot compre-
hend how this is possible). . . . Only freedom in relation to the internal
legislation of reason is properly a capacity; the possibility of deviating from
it is an incapacity. (MM 226-27)

https://doi.org/10.1017/CBO9781139174503.007 Published online by Cambridge University Press

 172 RANT'S MORAL PHILOSOPHY

Many readers, among them Sidgwick,1? have complained that so

strong an identification of freedom and morality should force Kant
to give up his account of moral imputability. If the moral law is the
unique positive conception of freedom, then it seems as if only mor-
ally good actions are really free. Kant does say that if we were solely
members of the intelligible world we would always act in accor-
dance with the moral law. How then are we to account for the
imputability of bad actions and characters? Your noumenal self
would not have chosen them. Your phenomenal self, being wholly
determined, cannot be held responsible.
But these complaints ignore the status of the positive conception of
freedom, and its corollary, intelligible existence, in the Kantian sys-
tem. The positive conception of freedom, understood as noumenal
causality, is a postulate of practical reason, in the sense developed in
the Dialectic of the Critique of Practical Reason. Kant explains the
basis of such postulates this way:

The postulates of pure practical reason all proceed from the principle of moral-
ity, which is not a postulate but a law by which reason directly determines the
will. This will. . . requires these necessary conditions for obedience to its
precept. These postulates are not theoretical dogmas but presuppositions of
necessarily practical import; thus, while they do not extend speculative
knowledge, they give objective reality to the ideas of speculative reason in
general (by means of their relation to the practical sphere), and they justify it
in holding to concepts even the possibility of which it could not otherwise
venture to affirm.
These postulates are those of immortality, of freedom affirmatively re-
garded (as the causality of a being so far as he belongs to the intelligible
world), and of the existence of God. (C2 132)

A postulate of practical reason is an object of rational belief, but the

reasons for the belief are practical and moral. The person needs the
belief as a condition for obedience to the moral law, and it is this,
combined with the categorical nature of that law, that justifies the
belief. Although the beliefs are theoretical in form - the will is free,
there is a God - their basis and their function are practical. As Kant
says in the passage quoted above, and as he constantly emphasizes in
the second Critique, the postulates play no theoretical or explana-
tory role whatsoever. They provide us with concepts that define the
intelligible world, but we have no intuitions to which we may apply

https://doi.org/10.1017/CBO9781139174503.007 Published online by Cambridge University Press

 Morality as freedom 173
those concepts, and consequently no theoretical knowledge of their
objects (e.g. C2 54-56; 133; 136).
The fact that the postulates of practical reason play no theoretical
role has two important implications. One is that we cannot con-
clude from the Argument from Spontaneity that evil is impossible,
or that a person who does something evil has not done it freely. A
free but evil will is shown to be unintelligible from the standpoint of
pure practical reason, but not to be theoretically impossible. It can-
not be explained, but no act of freedom can be explained. And we are
whole persons, not just pure spontaneous wills. Unlike the pure will
in the Argument from Spontaneity, we are imperfectly rational, be-
cause we are subject to temptation from inclinations. There is no
problem about explaining how we go wrong.
A central feature of Kant;s philosophy as a whole is brought out
here. The deliberating agent, employing reason practically, views the
world as it were from a noumenal standpoint, as an expression of the
wills of God and other rational agents. This is the philosophical
consequence of the fact that we act under the idea of freedom, and of
the way in which freedom leads to the other practical postulates: the
ethical world replaces the world of speculative metaphysics. Kant
tells us that "a moral principle is nothing but a dimly conceived
metaphysics, which is inherent in every man's rational constitu-
tion" (MPV 376). The theorizing spectator, on the other hand, views
the world as phenomena, mechanistic and fully determined. The
interests of morality demand a different conceptual organization of
the world than those of theoretical explanation (MM 217; 221; 225).
Both interests are rational and legitimate. And it is important that
neither standpoint is privileged over the other - each has its own
territory. Or, if either is privileged, it is the practical, because, accord-
ing to Kant, "every interest is ultimately practical" (C2 121).l8 From
the explanatory standpoint of theoretical reason, nothing is easier to
understand than that a human being might evade duty when it is in
conflict with her heart's desire. From the normative standpoint of
practical reason her sacrifice of her freedom for some mere object of
inclination is completely unintelligible. These two standpoints give
us two very different views of the world. To suppose that the Argu-
ment from Spontaneity shows anything at all about what can hap-
pen is to mix the theoretical and explanatory standpoint with the
practical and normative one in an illegitimate way.

https://doi.org/10.1017/CBO9781139174503.007 Published online by Cambridge University Press

 174 KANT ; S MORAL PHILOSOPHY

The second implication follows from the first. The standpoint

from which you adopt the belief in freedom is that of the deliberat-
ing agent. You are licensed to believe in the practical postulates
because they are necessary conditions of obeying the moral law.
Thus it is primarily your own freedom that you are licensed to
believe in, and, as a consequence, it is primarily yourself that you
hold imputable. The result is that the business of praising and blam-
ing others occupies a somewhat unstable position in Kantian ethics.
It is true that you are supposed to regard others as free, and to treat
them accordingly. But the necessity of doing so comes from the
moral law, which commands the attribution of freedom to persons,
and not from theoretical reasoning about how their wills actually
function. J9 The moral sentiments of approval and disapproval, praise
and blame, are, when directed to others, governed by the duties
associated with the virtues of love and respect. And these duties, as
Kant understands them, may actually demand attitudes of us that
exclude or curtail theoretical reasoning about the motives of others.
To the extent that we respect others and regard them as free, we
must admit that we do not know the ultimate ground of their mo-
tives. And not knowing it, we are obligated wherever possible to
take a generous attitude. Even when dealing with an actual wrong-
doer, Kant says we must
not deny the wrongdoer all moral worth, because on that hypothesis he
could never be improved either - and this latter is incompatible with the
idea of man, who as such (as a moral being) can never lose all predisposition
to good. (MPV 463-64)20

And Kant urges us to "cast the veil of philanthropy over the faults of
others, not merely by softening but also by silencing our judgments"
(MPV466).21
The positive conception of freedom, then, is not to be given a
theoretical employment. The idea of positive freedom is not sup-
posed to show that moral evil is so irrational that it is impossible.
Indeed, Kant does not propose that we should explain actions theo-
retically by referring them to the free choice of maxims in an intelli-
gible world. The role of the idea of freedom and the intelligible world
is, rather, a practical one. It provides a conception of ourselves which
motivates us to obey the moral law.
In Kant's philosophy, freedom of the will cannot be theoretically

https://doi.org/10.1017/CBO9781139174503.007 Published online by Cambridge University Press

 Morality as freedom 175
established. To establish it would be to achieve knowledge of the
noumenal world, and this is something that we cannot have. The
freedom of the will is asserted, but as a practical postulate, and so
only from a practical point of view. But surely, one is tempted to say,
it cannot simply fail to matter to the moral agent who is to be
motivated by this conception whether she is in point of actual fact
free or mechanistically determined.
In one sense Kant's response to this worry is contained in the idea
of the Fact of Reason. The Fact of Reason is our consciousness of the
moral law as a determining ground of the will (C2 31). Kant says:
"We can come to know pure practical principles in the same way we
know pure theoretical principles, by attending to the necessity with
which reason prescribes them to us and to the elimination from
them of all empirical conditions, which reason directs" (C2 30). The
moral law is thus presented to us by reason "as soon as we construct
maxims for the will" (C2 29) and it reveals our freedom to us. It does
this by showing us that we are able to act against even our strongest
inclinations, because there are cases in which we ought to. Kant says
that a person considering such a case:
judges . . . that he can do something because he knows that he ought, and he
recognizes that he is free - a fact that, without the moral law, would have
remained unknown to him. (C2 30)
Putting this together with the argument from the Groundwork about
acting under the idea of freedom, we arrive at an account of the possi-
bility of morality with a rather complicated structure, (i) We must act
under the idea of (at least negative) freedom; (ii) we must therefore act
on maxims we regard ourselves as having chosen; (iii) by the Argu-
ment from Spontaneity (or, as Kant puts it here, by eliminating all
empirical conditions, as reason directs) we are led to the moral law
(the positive conception of freedom); (iv) our ability to act on the
moral law teaches us that we are (negatively) free; (v) if so, we are
members of the intelligible world, and have a higher vocation than
the satisfaction of our desires; and (vi) this provides us with the incen-
tive to be positively free - that is, moral.
But all of this still remains at the level of the practical postulate.
For the sense in which our ability to act on the moral law teaches us
that we are free (step iv) and so are members of an intelligible world
(step v) is that we must believe these things in order to obey the

https://doi.org/10.1017/CBO9781139174503.007 Published online by Cambridge University Press

 176 RANT'S MORAL PHILOSOPHY

categorical imperative. And articles of belief we hold because they

are necessary conditions of obedience to the moral law are practical
postulates, with no theoretical employment.
And, in a sense, Kant's answer to the question whether it matters
if we are in fact (theoretically) free is that it does not matter. Kant's
deduction of freedom from the moral law in the Critique of Practical
Reason concludes:
Thus reason, which with its ideas always became transcendent when pro-
ceeding in a speculative manner, can be given for the first time an objective,
although still only practical, reality; its transcendent use is changed into an
immanent use, whereby reason becomes, in the field of experience, an effi-
cient cause through ideas. (C2 48; my emphasis)

Reason becomes an efficient cause by telling us how a free person

would act and by providing the conception of our higher vocation
that motivates us to act that way. For if the moral law does indeed
provide the positive conception of freedom, then we know how a
person with a completely free will would act. Motivated by the idea
of the higher vocation freedom gives us, we can act that way our-
selves. But if we are able to act exactly as we would if we were free,
under the influence of the idea of freedom, then we are free. Nothing
is missing: the will in the Argument from Spontaneity, when mak-
ing its original choice of a principle, could not do more. It chooses to
act on the moral law for the sake of maintaining its freedom; and we
can do the same. By acting morally, we can make ourselves free.

II VIRTUE AS FREEDOM

5. At this point a natural objection arises. The proposed solution to

the free will problem depends on our being able to act according to
the moral law for the sake of our freedom. I have claimed that what
interests us in our freedom is the higher vocation of contributing to
the Highest Good. But if this interest determines our moral actions,
how can we be free? To answer this question, we must turn to Kant's
theory of virtue, or "internal freedom."
It is Kant's view that all human action is purposive. A human
being always acts for the sake of an end. Kant speaks of this as being
the result of our finite and sensible nature. In a footnote in Religion I
Kant says:

https://doi.org/10.1017/CBO9781139174503.007 Published online by Cambridge University Press

 Morality as freedom 177
All men could have sufficient incentive if (as they should) they adhered
solely to the dictation of pure reason in the law. What need have they to
know the outcome of their moral actions and abstentions . . . ? Yet it is one
of the inescapable limitations of man and of his faculty of practical reason (a
limitation, perhaps, of all other worldly beings as well) to have regard, in
every action, to the consequence thereof - which consequence, though last
in practice . . . is yet first in representation and intention. ... In this end, if
directly presented to him by reason alone, man seeks something that he can
love} therefore the law, which merely arouses his respect, even though it
does not acknowledge this object of love as a necessity, does yet extend itself
on its behalf by including the moral goal of reason among its determining
grounds. (R 6-yn)
The objective necessity in the law ought to motivate us directly, but
a human being always acts for the sake of an end. This is why, in the
Groundwork, it is after explaining the Formula of Universal Law
that Kant embarks on the project of showing the possibility of rea-
son determining conduct a priori, and launches into a discussion of
ends (G 427). The Formula of Universal Law explains the objective
necessity of moral conduct, but it does not explain the subjective
necessity: that is, it does not explain how pure reason secures "ac-
cess to the human mind" (C2 151). Pure practical reason itself must
gain access to us through ends. Thus it is necessary to introduce the
Formula of Humanity, which directs that we make humanity, and
other aims which may be derived from it, our ends. The Religion
footnote continues:
This extension is possible because of the moral law's being taken in relation
to the natural characteristic of man, that for all his actions he must conceive
of an end over and above the law (a characteristic which makes man an
object of experience). (R jn)21
Kant also says that it is because of our susceptibility to temptation
that ethics extends to ends.
For since sensible inclinations may misdirect us to ends (the matter of
choice) which may be contrary to duty, legislative reason cannot guard
against their influence other than, in turn, by means of an opposing moral
end, which therefore must be given a priori independently of inclination.
(MPV381)
This sounds like a different account of the need for ends, but I
believe it is not. The same element in our nature-the passive,

https://doi.org/10.1017/CBO9781139174503.007 Published online by Cambridge University Press

 178 RANT'S MORAL PHILOSOPHY

sensible, representational element that makes us require an end, is

also what makes us susceptible to temptation.^
What this implies is that for human beings, freedom must take the
form of virtue: the adoption and pursuit of moral ends. Kant explains
why, in The Metaphysical Principles of Virtue, by setting up a prob-
lem (MPV 388-89). Every action has an end, and choice is always
determined by an end (G 427; MPV 381, 384-85; R4). So a maxim of
action, or of the means to an end, is adopted freely only when you
have adopted the maxim of holding that end. But the moral law only
says that the maxim we adopt must have a certain form, not that we
must have certain maxims. How can it be necessary to have certain
maxims? The answer is that if there are ends that are duties, there
will be maxims that it is a duty to have: maxims of actions that
promote those ends. Since we must believe that we are morally
obligated (that is, that there are maxims we ought to have), we must
believe that there are such obligatory ends. For example, Kant says
that the (external) duties of justice can be done from a moral motive
and so done freely by one who makes the rights of humanity one's
end (MPV 390). The possibility of internal freedom is secured by the
"Supreme Principle of the Doctrine of Virtue" which runs "Act
according to a maxim whose ends are such that there can be a univer-
sal law that everyone have these ends" (MPV 395). This principle is
deduced from pure practical reason by the following argument:
For practical reason to be indifferent to ends, i.e., to take no interest in
them, would be a contradiction; for then it would not determine the max-
ims of actions (and the actions always contain an end) and, consequently,
would not be practical reason. Pure reason, however, cannot a priori com-
mand any ends unless it declares these ends to be at the same time duties;
such duties are then called duties of virtue. (MPV 395)

In the introduction to The Metaphysical Principles of Virtue, Kant

says that the obligatory ends are one's own perfection and the happi-
ness of others. But in fact, a number of different ends appear in this
text and elsewhere in the ethical writings. One's own perfection
includes moral perfection and so subsumes the whole of morality, as
well as natural perfection, which involves the development of our
physical and intellectual capacities. The duties of respect make the
rational autonomy of others an end. Securing the rights of humanity
is an end (MPV 390). In the political writings, the development of

https://doi.org/10.1017/CBO9781139174503.007 Published online by Cambridge University Press

 Morality as freedom 179
republican forms of government is made a necessary end for sover-
eigns (MPJ 340), and peace is an end for everyone (MPJ 354-55). The
Highest Good, the whole object of practical reason, is a necessary
end, as we have already seen (C2 108-14; R 3-6). It is because there
are various ends, Kant says, that there are various virtues, even
though virtue is essentially one thing (MPV 395; 406). All of these
ends are determined by the moral law and so are necessary ends
(ends of reason); and all of them can be derived from the uncondi-
tional value of humanity. When we act for the sake of these ends, we
act from the moral law, for it determines them. It is because the law
determines ends that creatures like ourselves, who always act for
ends, can be free.2*

6. But this may not seem to resolve the problem. Clearly it is not
enough that we act for moral ends; we must also do so because they
are moral ends. We must adopt the ends themselves freely, as ends
determined by the moral law. But if we must be free in order to adopt
moral ends, then adopting moral ends cannot be what makes us free.
The answer to this objection lies in the special nature of internal
freedom. To explain the answer, we must take a detour through
another problem about the adoption of moral ends. Kant argues that
the duties of virtue are all of broad obligation; they do not require
definite acts which may simply be discharged (MPV 39o).25 The duty
to advance a moral end is one of broad obligation because it is an
imperfect duty; the law does not say exactly what or how much we
should do to advance the end. But what about the duty to adopt a
moral end? Kant thinks that the adoption of an end is necessarily a
free act, for he says:
Another may indeed force me to do something which is not my end (but is
only the means to some other's end); but he cannot force me to make it my
own end, for I can have no end except of my own making. (MPV 381-82)

Making something your end is a kind of internal action, and it is

these internal actions that are commanded by the Supreme Principle
of the Doctrine of Virtue. The duty to adopt these ends (and so also
the duty not to act against them) is a perfect duty. The law does say
exactly what we must do. So why does Kant count such duties as
creating only broad obligations which may not be discharged?
One of the things we expect of a person who has an end is that she

https://doi.org/10.1017/CBO9781139174503.007 Published online by Cambridge University Press

 180 RANT'S MORAL PHILOSOPHY

will notice facts that are associated with that end in a certain way,
and things that bear on the promotion of the end will occur to her.
This is a general point about ends, and does not apply only to moral
ends. To see this, imagine that I claim that I am Charlotte's friend,
and that I have her happiness as my end. But imagine also that it
seldom or never occurs to me to do anything in particular to make
Charlotte happy. When I see something in a shop window that suits
her taste exactly, I do not think, "now that is something Charlotte
would really like/ 7 and go in to purchase it. When I look at the
calendar on what happens to be her birthday, it does not occur to me
that it is her birthday, and I should telephone. When I hear of some
catastrophe happening in her neighborhood, I do not wonder about
the possible bearing of this event on her safety and comfort. These
things just do not come to me. Under these circumstances, surely
Charlotte would be entitled to complain that there is no real sense
in which I have her happiness as my end.26 It would not be pertinent
for me to reply that I have no direct control over what occurs to me.
To find certain features of the world salient is part of our notion of
what it is to have an end. To have an end is to see the world in a
certain way. But what determines salience most directly lies in our
sensory and representational capacities - and so in the passive part
of our nature. To adopt an end is to perform an internal action. But it
is also to undergo certain changes, changes in your representational
capacities. It is to come to perceive the world in the way that having
the end requires.
When the end is one that is suggested by natural inclination, we
are already inclined to perceive the world in the relevant way. In-
deed, that you are inclined to perceive the world that way is the form
that the incentive takes. Our sensible nature here helps us out. But
when the end is one prompted by reason this may not be the case.
Here, you are imposing a change on your sensible nature, and your
sensible nature may, and probably will, be recalcitrant. Although
adopting an end is a volitional act, it is one that you can only do
gradually and perhaps incompletely.
This is why the duty to adopt an end is of broad obligation. You
cannot, just by making a resolution, acquire a virtue or recover from
a vice. Or better still, we will say that you can, because you are free,
but then we must say that only what happens in the future estab-
lishes whether you have really made the resolution or not. I do not

https://doi.org/10.1017/CBO9781139174503.007 Published online by Cambridge University Press

 Morality as freedom 181
mean that only the future will produce the evidence: I mean that
only what you do in the future will enable us to correctly attribute a
resolution to you. There is a kind of backwards determination in the
construction of one's character. Whether you have made it your
maxim to be more just, helpful, respectful, or honest depends on
what you do in the future - on whether you make progress towards
being the sort of person you have (presumably) resolved to be. Be-
cause the materials we have to work with in these cases are recalci-
trant, it is in the progress, not in the success, that Kant places virtue
(MPV 409). But the work must show up in progress. Suppose, for
instance, that I am selfish, but resolve to be more attentive to the
needs of others. As a selfish person I will also be self-absorbed, and
fail to notice when others are in trouble and to be perceptive about
what they need. At first, others may have to draw my attention to
the cases where I can help. But if I continue indefinitely to fail to
notice when others are in need and I can help, then I just did not
resolve. On the other hand, if I do progress, I will count as having
resolved, even if I am not consistently unselfish all of the time.
This is Kant's explicit view in Religion Within the Limits of Rea-
son Alone. According to Kant, we must think of our free actions and
choices as being unconditioned by time. If they were conditioned by
time they would be subject to causality and so not free (R 40). Still,
time is a condition of our thinking, and this means that for us,
temporally unconditioned choice must be represented as choice that
either is before or in a certain way follows from the events of our
lives. For purposes of holding ourselves responsible, we think of the
free adoption of our most fundamental maxim as if it were before
our phenomenal choices: the evil in us is present from birth, Kant
says, as if it were innate (R 21-22; 41; see also C2 100). But if our
maxims were innate, we could not change for the better, for our
most fundamental reasons would be self-interested ones. So, for pur-
poses of regarding ourselves as free to change, we see the free choice
of our character as something to which the whole conduct of our life
adds up. Kant explains:
duty demands nothing of us which we cannot do. There is no reconciliation
possible here except by saying that man is under the necessity of, and is
therefore capable of, a revolution in his cast of mind, but only of a gradual
reform in his sensuous nature (which places obstacles in the way of the
former). That is, if a man reverses, by a single unchangeable decision, that

https://doi.org/10.1017/CBO9781139174503.007 Published online by Cambridge University Press

 182 RANT'S MORAL PHILOSOPHY

highest ground of his maxims whereby he was an evil man (and thus puts on
the new man), he is, so far as his principle and cast of mind are concerned, a
subject susceptible of goodness, but only in continuous labor and growth is
he a good man. That is, he can hope ... to find himself upon the good
(though strait) path of continual progress from bad to better. For Him who
penetrates to the intelligible ground of the heart (the ground of all maxims
of the will [Willkiir]) and for whom this unending progress is a unity, i.e., for
God, that amounts to his being actually a good man (pleasing to Him); and,
thus viewed, this change can be regarded as a revolution. (R 47-48)27
The appearance of freedom in the phenomenal world, then, is
virtue - a constant struggle to love and respect the humanity in
oneself and others, and to defeat the claims inclination tries to make
against that humanity. So far from committing him to a mysterious
dualism, Kant's theory of the atemporal nature of freedom permits
him to harmonize freedom with a temporal account of the acquisi-
tion of virtue. One achieves virtue through a gradual habituation,
and, as in Aristotle's ethics, the sign of success is gladness in its
practice. In the Religion, Kant says:
This resolve, then, encouraged by good progress, must needs beget a joyous
frame of mind, without which man is never certain of having really attained
a love for the good, i.e., of having incorporated it into his maxim. (R 2in)
To the extent to which moral ends have really become our ends, we
will take pleasure in the pursuit of them. Indeed we have all of the
emotions appropriate to having an end. In the Metaphysical Princi-
ples of Virtue, Kant speaks of gratitude (MPV 454-56) and even
sympathetic feeling (MPV 456-58) as being required. He is quick to
qualify these remarks, for we have no direct control over our feel-
ings. Yet it is his view that one who does adopt an end will normally
come to have the feelings that are natural to a person who has this
end. If the end were suggested by sensibility, we would already have
had the feelings, but though the end is adopted on moral grounds we
should still come to have them eventually.28 When explaining the
relation between inclination and morality in the duty of benefi-
cence, for example, Kant says:
Beneficence is a duty. Whoever often exercises this and sees his beneficent
purpose succeed comes at last really to love him whom he has benefited.
When therefore it is said, "Thou shalt love thy neighbor as thyself/' this
does not mean you should directly (at first) love and through this love

https://doi.org/10.1017/CBO9781139174503.007 Published online by Cambridge University Press

 Morality as freedom 183
(subsequently) benefit him; but rather, "Do good to your neighbor/' and this
beneficence will produce in you the love of mankind (as a readiness of
inclination toward beneficence in general). (MPV 402)

Kant does not mean that we will come to act solely from the inclina-
tion, but rather that the inclination will be in harmony with reason
and so will no longer be an impediment. As long as we do not act
from inclination, but because the ends are dictated by the law, this is
no detriment to our moral character. On the contrary, it shows that
we have advanced toward the complete control over our sensuous
nature that is implied by freedom.
So we do not exactly need to adopt moral ends freely in order to be
free. If we come, over time, to act purely for the sake of moral ends,
it will come to be true that we are, timelessly, free.

7. Kant's theory of the freedom of the will involves neither extrava-

gant ontological claims nor the unyielding theory of responsibility
which seems to follow from those claims.2? These problems arise
only from a misunderstanding of a fundamental feature of the
Kantian philosophy - the radical split between the theoretical and
practical points of view. The idea of intelligible causality is a practi-
cal conception, and our belief in it is an article of practical faith. It is
not supposed to be theoretically employed, and it cannot be used to
explain anything that happens. It is true that the positive conception
of freedom makes practical freedom possible - but not because it
explains how it is possible. It makes practical freedom possible be-
cause we can act on it.
Kant sees positive freedom as pointing to a higher vocation, the
thought of which moves us to moral conduct, and explains how we
can take an interest in such conduct. This interest leads us to adopt
moral ends, and so to struggle against the temptations that beset us.
If we reach the point where we are indeed moved wholly by ends
determined by the law, we are in fact free - practically free. Nothing
in this development requires any ontological claims, or requires that
we be radically different sorts of creatures than the mundane ra-
tional animals we suppose ourselves to be. All that Kant needs is the
conclusion that the moral law does indeed represent the positive
conception of freedom. The idea of freedom motivates us to culti-
vate the virtues, and, in turn, virtue makes us free.

https://doi.org/10.1017/CBO9781139174503.007 Published online by Cambridge University Press

 184 RANT'S MORAL PHILOSOPHY

NOTES

I would like to thank Manley Thompson, Andrews Reath, Stephen

Engstrom, and Onora O'Neill for valuable comments on earlier drafts of
this paper,
i For another treatment of some of these same difficulties, but centered
more on Kant's views in the Critique of Pure Reason, see Henry E.
Allison, "Empirical and Intelligible Character in the Critique of Pure
Reason."
i The alternative language is used because of the difference in Kant's own
two accounts of what he is doing. I discuss this below.
3 For some important discussions of this question see the following
works: H.J. Paton, The Categorical Imperative: A Study in Kant's Moral
Philosophy (1947), Book IV; W.D. Ross, Kant's Ethical Theory-, Karl
Ameriks, "Kant's Deduction of Freedom and Morality"; Dieter Henrich,
"Die Deduktion des Sittengesetzes: iiber die Griinde der Dunkleheit des
letzen Abschnittes von Kants Grundlegung zur Metaphysik der Sitten."
My own view on the matter is explained in Section 3.
4 The appendix, "The Kantian Conception of Free Will [Reprinted with
some omissions, from Mind, 1888, Vol. 13, no 51]" was first attached to
the sixth edition in 1901.
5 John Rawls, A Theory of Justice, p. 5. Rawls is in turn drawing upon
H.L.H. Hart, The Concept of Law, pp. 155-59. Rawls uses the distinc-
tion in separating the concept of justice, "a characteristic set of princi-
ples for assigning basic rights and duties and for determining. . . the
proper distribution of the benefits and burdens of social cooperation,"
from conceptions of justice, that is, various substantive accounts of
what those principles are.
6 You may take the belief into account in other ways, like other beliefs.
For instance, you may decide to warn your friends that you may do
something uncharacteristic today, and that if so they should not be
upset, since you are, as we say, "not yourself."
7 This is brought out well by Thomas Hill, Jr., in "Kant's Argument for
the Rationality of Moral Conduct," and in "Kant's Theory of Practical
Reason" both in his Dignity and Practical Reason in Kant's Moral
Theory, chapters 6 and 7.
8 To understand this as a law of nature, rather than as a tautology, we
must of course understand a "desire" not merely as something we as-
cribe to a person on the basis of her actions, but as a psychological
phenomenon of some sort. This view of desire is also implied by Kant's
account of desire as an incentive, which I explain below.
9 More specifically, Kant associates the will's spontaneity with the fact

https://doi.org/10.1017/CBO9781139174503.007 Published online by Cambridge University Press

 Morality as freedom 185
that it does not exist under temporal conditions and so is uninfluenced by
causality, but the important point here is just being uninfluenced - by
anything. I discuss the relation between freedom and time in Section 6.
10 "World" [Welt) is Kant's term, and it is in some respects unfortunate,
since it has lent credence to the interpretation of the distinction as an
ontological dualism. Actually these two worlds are two standpoints, or
ways we have of looking at things; as I will argue in the next section,
they represent a practical and a theoretical viewpoint. I have continued
to use the terminology of two worlds, since it is convenient and suits
Kant's own usage. I would like to thank Onora O'Neill for urging me to
be clearer on this point.
11 The remark is not italicized in Beck's translation, although it is in the
Akademie Textausgabe and the Paton and Abbott translations.
12 For a different reading than mine of the idea that the intelligible world
contains the grounds of the sensible world and its laws, and of why we
must conceive ourselves as among those grounds, see Onora O'Neill's
"Agency and Anthropology in Kant's Groundwork."
13 That our noumenal choices are in some way the ground of the laws of
nature is a possibility that remains open,- it is enough for the argument
that we do not conceive ourselves as choosing these laws.
14 In a footnote in "On the Common Saying: This may be True in Theory,
but it does not Apply in Practice' " Kant speaks directly of the moral
incentive as provided by the idea of the highest possible earthly good, as
"attainable through his [man's] collaboration [Mitwirkung]" (TP 28on).
15 For a different and perhaps more sympathetic account of the argument
of Groundwork III, see Onora O'Neill, "Agency and Anthropology in
Kant's Groundwork," especially Section 6.
16 The view that the idea of the intelligible world plays a motivational role
can also be supported by appeal to Kant's writings on moral education,
especially in the Methodologies of the second Critique and The Meta-
physical Principles of Virtue. In both, there is an emphasis on awaken-
ing the child to the sublimity of the intelligible existence which freedom
reveals.
17 Sidgwick, The Methods of Ethics, p. 516.
18 The metaphysical conception of the world also provides the regulative
principles used in the theoretical sphere - but what those do is regulate
the practice of science.
19 In Kantian ethics moral concepts are ideals of practical reason that are
imposed on the world, by the command of the moral law, and for practi-
cal and moral purposes only. When we praise and blame we are, so to
speak, applying the concept of "freedom" to another. The moral law
both commands and regulates the application of this concept. I discuss

https://doi.org/10.1017/CBO9781139174503.007 Published online by Cambridge University Press

 186 RANT'S MORAL PHILOSOPHY

this way of regarding moral concepts in "Two Arguments against Ly-

ing/ 7 Chapter 12 in this volume.
20 I give a fuller explanation of the attitude Kant thinks is required and the
moral basis for it in my "The Right to Lie: Kant on Dealing with Evil/7
Chapter 5 in this volume.
21 In these respects Kant's views stand in sharp contrast to the British senti-
mentalists whom he admired: Hutcheson, Hume, and Adam Smith. All
developed their ethical theories from the point of view of the spectator of
the moral conduct of others, and took approbation and disapprobation as
the central concepts of ethics, from which the other concepts of moral
thought are developed. Hutcheson and Hume believe that the best moral
agent is not thinking about morality at all, but acting from admirable
natural affections. Smith comes closer to an agent-centered theory, for he
takes the agent to act from specifically moral thoughts, but they are
generated from an internal spectator.
22 The mysterious-sounding parenthetical phrase is "welche Eigenschaft
desselben ihn zum Gegenstande der Erfahrung macht." I take the point
to be to equate sensibility and the need for an end.
23 In the Introduction to the Metaphysics of Morals, the faculty of desire is
"the capacity to be by means of one's representations the cause of the
objects of these representations" and the capacity to act in accordance
with representations is identified as "life;/ (MM 211).
24 It might seem to be a problem that the Highest Good is supposed to be
conceived as a divine end. How can God have an end if that is a need of
sensibility? Kant explains: "For while the divinity has no subjective
need of any external object, it cannot be conceived as closed up within
itself, but only as compelled by the very awareness of its own all-
sufficiency to produce the highest good outside itself. In the case of the
supreme being, this necessity (which corresponds to duty in man) can be
envisaged by us only as a moral need" (TP 28on).
25 The question of the relation between the two distinctions, perfect/
imperfect, and broad/strict, is a very difficult one. These have some-
times been thought to be simply alternative terms for the same distinc-
tion, but Kant explicitly asserts that all duties of virtue are of broad
obligation, while mentioning many that are perfect. He does not explain
himself, and his own use of the terms does not provide clear guidance.
Two important discussions of this problem are in Mary Gregor, The
Laws of Freedom: A Study of Kant's Method of Applying the Categori-
cal Imperative in the Metaphysik der Sitten, pp. 95-127, and in Onora
(O'Neill) Nell, Acting on Principle: An Essay on Kantian Ethics, pp. 4 3 -
58. The main justification I have to offer for the way I use these terms in
the text is that they enable me to make the explanation that follows.

https://doi.org/10.1017/CBO9781139174503.007 Published online by Cambridge University Press

 Morality as freedom 187
26 In one sense I may still claim to have her happiness as my end. I may hold
an end merely negatively, as something I will endeavor not to act against.
The Formula of Humanity says that we must never use another merely as
a means, and Kant says in the Groundwork that humanity is conceived
negatively, as "that which must never be acted against" (G 437). But Kant
makes it clear that virtue is going to require a more positive pursuit of the
end. He says: "It is not enough that he is not authorized to use either
himself or others merely as means (this latter including also the case of
his being indifferent to others)" (MPV 395).
27 See also this passage from the Religion: "we may also think of this endless
progress of our goodness towards conformity to the law, even if this prog-
ress is conceived in terms of actual deeds, or life-conduct, as being judged
by Him who knows the heart, through a purely intellectual intuition, as a
completed whole, because of the disposition, supersensible in its nature,
from which this progress itself is derived" (R 67-68), and from the Cri-
tique of Practical Reason: "Only endless progress from lower to higher
stages of moral perfection is possible to a rational but finite being. The
Infinite Being, to whom the temporal condition is nothing, sees in this
series, which is for us without end, a whole comformable to the moral
law" (C2 123). This is why Kant thinks that ethics leads to a view of the
"immortality" of the soul, which gives us a prospect of an endless prog-
ress toward the better. Only an endless progress is adequate to the achieve-
ment of freedom, and to wiping out the original evil in our nature (R 72;
C2 122-24).
28 This is not guaranteed. The Groundwork contains a well-known discus-
sion of the worth of a man who is helpful although "by temperament
cold and indifferent to the sufferings of others, perhaps because he is
provided with special gifts of patience and fortitude" (G 398), which
shows that Kant thinks moral worth may be combined with a recalci-
trant temperament. The discussion has unfortunately often been taken
to suggest that Kant thinks moral worth must be combined with a
recalcitrant temperament.
29 Kant's theory of free will is sometimes described as "compatibilist"
because both freedom and determinism are affirmed. This description
seems to me to be potentially misleading. Most compatibilists, I believe,
want to assert both freedom and determinism (or, both responsibility
and determinism) from the same point of view - a theoretical and ex-
planatory point of view. Kant does not do this, and could not do it
without something his view forbids - describing the relation between
the noumenal and phenomenal worlds.

https://doi.org/10.1017/CBO9781139174503.007 Published online by Cambridge University Press