<< . .

. 27
( : 33)

. . >>

never to act except in such a way™™) has the advantage of making clear that the moral
imperative is an imperative that the agent assigns to herself, and that there is no other
ground for the obligation than being the subject who thinks ˜˜I.™™ It also makes clear that the
role of the categorical imperative is mainly negative, and comes in after maxims have
already been adopted: it asks me to check whether the maxim I have always already set
for my actions from hypothetically formulated ends, is permissible or not (˜˜never act except in
such a way that you could also will . . . ™™). I will return to both points below.

necessary for achieving that end; if, for one reason or another, I do not
endorse or obtain those means, then I ought to renounce willing the end.
The form of the hypothetical imperative (˜˜if . . . then™™) conveys pre-
cisely the fact that the injunction expressed in the consequent (˜˜then I
ought to will the means™™) is conditioned by the preliminary position of an
end, expressed in the antecedent. Such a conditioning relation is just what
the hypothetical form of a judgment generally conveys. In a hypothetical
judgment, Kant said in the first Critique, only the Konsequenz, that is, the
relation between the antecedent (here, the statement of the end to achieve
or obtain) and the consequent (here, the statement of obligation one sets
to oneself, to will the means necessary to achieve the end), is asserted.
Indeed, in the case of the hypothetical imperative, the Konsequenz is not
only asserted, but apodictically asserted, since the relation between willing
the end (antecedent) and being thereby committed to willing the means
(consequent) is, according to Kant, analytic (see again iv, 419). This does
not mean of course that whoever wills the end does will the means
necessary to that end. Such would be the case only if our will were purely
rational. But in the case of our sensibly conditioned wills, all we can say is
that whoever wills the end is thereby committed to (i.e. ought to, or
should, or must: Kant says soll) either will the means or, if it turns out
the means are unavailable or unacceptable, renounce the end. Because
our wills are not purely rational, whether we will indeed clearly adopt one
or the other of these options is often a matter of no less struggle than
whether we will act according to what the categorical imperative com-
mands (a point considered below). This is precisely why there is room for
a hypothetical imperative.9

On the form of hypothetical judgments for Kant, see above, ch. 4, pp. 98“9 and ch. 6,
pp. 150“5. Note that Kant certainly does not mean to say that any imperative expressed in
the form of a hypothetical judgment would thereby be a hypothetical imperative. Rather,
the hypothetical form here captures a specific content: the relation between willing an end
and being committed to willing the means necessary to achieve that end. In other words,
Kant is making use of the forms of relation set up in the table of logical functions in the first
Critique (in this case, the relation between antecedent and consequent in hypothetical
judgments) to clarify the relation between ˜˜added condition™™ (willing an end) and con-
ditioned (being thereby rationally obliged to willing the means) at work in the particular
imperatives he therefore calls ˜˜hypothetical imperatives.™™ Note also that Kant sometimes
describes what one is committed to by willing the end as ˜˜willing the means™™ and sometimes
as simply ˜˜the means.™™ In other words, ˜˜If I will the end X, I ought to will Y,™™ or ˜˜If I will the
end X, I ought to Y.™™ The difference is minimal, since as we shall see below, willing is for
Kant one of the ways of ˜˜being by one™s representation the cause of the existence of an
object™™ (see v , 9n and below, pp. 249“50). Willing and acting to make the object of one™s
willing come about, are inseparable. Accordingly, in what follows I shall sometimes

Among the hypothetical imperatives, Kant distinguishes two main
kinds: the counsels of prudence and the rules of skill. In the counsels
of prudence, he claims, the antecedent expresses an end that all beings
which are both rational and sensible have: their own happiness. The
consequent expresses the injunction to will the means necessary to
obtain that end. The general expression of counsels of prudence could
thus be: ˜˜If I want happiness, then I ought to will all the means available
to me that are necessary to achieve happiness.™™10 The rules of skill, for
their part, have for their antecedent the expression of a particular end
that some, and not all, are apt to give themselves. They have as con-
sequent the injunction to employ the means to arrive at this possible end.
For example: ˜˜If I want to become a violinist, then I ought to practice
violin several hours a day.™™
Kant describes the counsels of prudence as assertoric and the rules of
skill as problematic. This is perplexing. Just what is being described
In the Critique of Pure Reason, Kant says that the antecedent and the
consequent of hypothetical judgments are both problematic (neither
of them is asserted), and only the relation (Konsequenz) is affirmed,
assertorically or apodictically (cf. A75/B100). As we just saw, when com-
menting on the nature of hypothetical imperatives Kant maintains that
the connection between willing the end and the ˜˜command™™ to will the
means is always analytic: this makes the connection apodictic (necessarily
true). Thus whether one considers the components of the hypothetical
imperative (the antecedent and the consequent), or the connection
between them, the modality seems to be the same: problematic for the

formulate the principle of hypothetical imperatives: ˜˜If I will X, I ought to will Y™™ and
sometimes: ˜˜If I will X, I ought to Y.™™
Kant defines happiness as ˜˜an absolute whole, a maximum of well-being in my present
condition and in every future condition™™ (iv, 418). As Allen Wood emphasizes, the repre-
sentation of such a goal is available only to rational beings. For it involves representing the
idea of a whole well-being, not just the satisfaction of particular desires and needs; and it
involves relating (comparing and connecting) present and future goods. See Allen Wood,
Kant™s Ethical Thought (Cambridge: Cambridge University Press, 1999), p. 66. I agree with
Wood that in Kant™s view, even though happiness is a goal of all rational and sensible
beings, one can fail to act according to the goal of happiness, for instance when one lets a
momentary pleasure take precedence over the goal of the ˜˜maximum of well-being.™™ This
only puts the agent in a situation of practical irrationality; it does not amount to denying
that happiness is an ongoing purpose of human beings. More difficult for Kant™s point is
the case of the neurotic who nurses his own misery and just does not want to be happy. But
I suggest such a case would be, for Kant, similar to losing the use of ˜˜I think™™ in cognition.
When it happens, it means that something deeply pathological has befallen a particular
human being in the (practical) use of ˜˜I™™ that essentially characterizes humanity.

components, apodictic for the connection. Why then does Kant disting-
uish the counsels of prudence and the rules of skill as being, respectively,
assertoric and problematic?
According to Kant™s explanation, the distinction in modality refers to
the status of the antecedent. In the counsels of prudence, the antecedent
is in fact true of any human being. All human beings, in fact, want to be
happy. Although the antecedent in the expression of the imperative has
the status of a problematic proposition, we must therefore assume the
implicit minor of a hypothetical syllogism that posits it as assertoric (˜˜I
[and everyone else] want to be happy™™) and thus warrants the con-
clusion: ˜˜So I (and everyone else) ought to will all means available to me
(to one) that are necessary to achieve happiness.™™ The antecedent becomes
assertorically stated in the minor premise. It is not apodictic, however.
All human beings seek happiness. But it is not impossible to renounce
this quest if a higher end compels one to do so. And it is all too often
abandoned in favor of lower, particular ends (being rich, say; or being
famous; or being powerful).
The hypothetical imperative that grounds the rules of skill is
described as a ˜˜problematically practical principle™™ because ˜˜it says
only that the action is good for some possible purpose™™ (iv, 415). We
might understand this in two ways. One way is to say that in the rules
of skill, all one needs is the statement of the hypothetical imperative itself
(where both antecedent and consequent are merely problematic) to
derive a maxim for action. This is what Kant seems to have in mind
when he writes:

Principles of action in which the latter is represented as necessary for
attaining some possible purpose to be brought about by it, are in fact
innumerable. All sciences have some practical part, consisting of prob-
lems which suppose that some end is possible for us and of imperatives as
to how it can be attained. These can be called, in general, imperatives of

There is no need here that the purpose expressed in the antecedent
should actually be posited in the minor premise of a hypothetical
syllogism for the precept expressed in the consequent to be binding.
The mere possibility that the antecedent might actually become a purpose
to be achieved is sufficient to ground the necessity of the ˜˜ought™™ expres-
sed in the consequent.
A second, and more complete, way to account for the ˜˜problematic™™
nature of a rule of skill is to actually formulate a practical inference

where the positing of the consequent (the detached ˜˜ought™™) follows
from the positing of the antecedent of the hypothetical major premise.
What one posits in the minor premise, however, is the mere possibility of
the antecedent™s obtaining. Kant expresses this most effectively when he
takes the example of the various learning processes to which parents
subject their offspring:
Since in early youth it is not known what ends might occur to us in the
course of life, parents seek above all to have their children learn a
great many things and to provide for skill in the use of means to
all sorts of discretionary ends, about none of which can they determine
whether it might in the future actually become their pupil™s
purpose, though it is always possible that he might at some time have
it. (iv, 416)

Consider the possibility that my child might have it in her to become a
talented violinist. We might formulate the corresponding practical infer-
ence in the following way: ˜˜If it is possible that my child should want to
become a professional violinist, she ought to practice her scales. It is
possible that she should want to become a professional violinist. So she
ought to practice her scales.™™ To be sure, the force of the imperative
(˜˜she ought to practice™™) depends on the degree of probability of the end
stated in the antecedent. If the end becomes assertorically affirmed (it
turns out my child does want to become a professional violinist), the
force of the imperative becomes as strong as that of the assertoric
imperative of prudence (˜˜I [and everyone else] ought to will the means
necessary to achieve happiness™™), while being much more determined:
my child knows exactly what she wants to achieve, and thus what she
ought to do. However, both the end and the means would be open to
challenge if it turned out that they jeopardized the overall goal of
happiness (say, my child™s wish to have a star™s career threatened her
mental or physical health, or her relations to loved ones). In this sense
the ˜˜oughts™™ of the rules of skill remain always problematic in the face of
the assertoric ˜˜ought™™ of the counsel of prudence (grounding its ˜˜ought™™
on an assertoric, although indeterminate goal, that of happiness).11 And

Note that a few years after writing Groundwork, in the First Introduction to the Critique of the
Power of Judgment Kant expresses doubts about the wisdom of calling the rules of skill
˜˜problematic imperatives.™™ The expression, he remarks, is self-contradictory: how could
an imperative be merely problematic (see AAxx, 200)? I should rather, he continues, have
called them ˜˜technical imperatives™™ or ˜˜imperatives of art.™™ The explanation he gives for
having called them ˜˜problematic™™ confirms the suggestions I have been making: rather

both hypothetical ˜˜oughts™™ are trumped if an imperative of apodictic
force comes to oppose them.

The categorical imperative
In contrast to hypothetical imperatives, the principle of moral obliga-
tion has the form of a categorical judgment.12 This form manifests the
fact that the obligation is not conditioned by any antecedently given
end or motive. According to Kant™s analysis of the forms of relation in
judgment, the condition or reason for the assertion in a categorical
judgment is contained in the subject of the judgment, be it the subject-
concept (in an analytic judgment) or the intuition thought under
the concept (in a synthetic judgment). The categorical form of the
imperative (˜˜I ought never to act except in such a way™™) thus expresses
the fact that by itself the subject ˜˜I™™ of the proposition ˜˜I ought to™™
provides the condition or reason for the obligation. In other
words, being the referent of ˜˜I™™ is a sufficient reason for being thus
The formulation of the moral imperative as a categorical imperative
thus invites us to ask the following question: what, in the nature of the
referent of ˜˜I™™ (the referent of the logical subject of the categorical
proposition ˜˜I ought never to act except in such a way™™) grounds the
assertion of the predicate, i.e. the assignment of obligation? This ques-
tion is inseparable from another, which concerns the predicate itself: on
what grounds does Kant claim that the content of the categorical obliga-
tion is that of ˜˜never acting except in such a way that I could also will that
my maxim should become a universal law™™? Kant offers an answer to the
second question (which is a question for moral philosophy) in the first
and second sections of Groundwork of the Metaphysics of Morals. He offers
an answer to the first question (which is a question for metaphysics) in
the third section of Groundwork, in the Critique of Practical Reason, and in

than the imperative itself, he says; it is the goal expressed in the antecedent that is ˜˜merely
possible,™™ thus ˜˜problematic.™™ In the same footnote Kant also notes that in the rules of
prudence, the goal is ˜˜actual™™ (thus the ˜˜assertoric™™ imperative), and ˜˜subjectively
It is also apodictic: the connection between the logical subject ˜˜I™™ in ˜˜I ought™™ and the
predicate ˜˜ought never to act etc.™™ is a necessary connection, and the obligation thus
grounded is unconditional (it does not depend on any previously stated end). Kant asserts
the point as soon as he introduces the categorical imperative (iv, 415) and attempts to
provide its ground by examining the nature of the referent of ˜˜I™™ in section three of
Groundwork (iv, 446“8). See also below, n. 13.

Religion within the Boundaries of Mere Reason.13 Since my concern in this
chapter is Kant™s view of moral judgment, the question that is more
relevant to me is the second: what justification does Kant offer for his
claim concerning the content of the moral imperative?
To this question, we find two answers. The first, given in section one of
Groundwork, consists in an examination of common moral conscience.
Kant maintains that such examination leads to the following conclusions:
(1) common conscience accords moral value to the will determined to act
from duty; (2) to act from duty is to be motivated by respect for the law;
(3) the expression of the law is that of the categorical imperative: ˜˜I
ought never to act except in such a way that I could also will that my
maxim become a universal law™™ (cf. iv, 393“402).
Kant gives a second, more extensive answer in section two. Like the
first, this second answer draws on the teachings of common moral
conscience, examined in section one. But it moves from there to a
philosophical interpretation by examining the notion of a will and the
different kinds of prescriptions the human will is capable of assigning to
itself. I shall briefly sketch out the first answer before discussing the
second in more detail.
In section one of Groundwork, the formulation of what is not yet called
a ˜˜categorical imperative™™ (the expression appears only in section two) is
preceded by several examples of supposedly common moral judgments
(iv, 397“400). These examples are supposed to show that neither the
action itself (for example, treating a client with equity), nor the sensible
motive (for example, benevolence or compassion), nor the empirically
determined end (for example, wanting the well-being of another) deter-
mines the moral worth of the action, and even less that of the agent. Only
the universal principle according to which the agent is determined to act,

In section three of Groundwork, Kant explains that it is insofar as it belongs to the intelligible
world, and in that capacity, is free of natural determinism, that the referent of ˜˜I™™ can
ground its own moral obligation (see iv, 452“3). In the Critique of Practical Reason, Kant
develops the thesis (briefly introduced in section three of Groundwork) of the reciprocity
between free will and a will that determines itself under the representation of the moral law
(AAv, 28“30). In Religion within the Boundaries of Mere Reason, he offers a striking presenta-
tion of the divided moral subject and the three aspects of its ˜˜disposition toward the good.™™
The third aspect is none other than the moral disposition (see Religion within the Boundaries
of Mere Reason, trans. and ed. Allen Wood and George Di Giovanni [Cambridge:
Cambridge University Press, 1998], AAvi, pp. 26“9). The question of the relation between
Kant™s notion of the referent for ˜˜I™™ and his conception of morality would deserve a
detailed study, which I am not undertaking in this book. Nevertheless, see above, ch. 5,
pp. 119, 138“40.

and the agent™s respect for this principle, as the motive for the action
(what Kant calls ˜˜respect for the law™™), are the source of the moral worth
of the action as well as of the agent. The very content of the principle in
question must therefore express this exclusion, from what determines
the moral value of the action and the agent, of any empirical end as well
as any sensible motive. But this leaves only the form of the law itself as a
possible content for the principle: the principle commands to act only
under that maxim (rule for action) that one can also will to become a
universal law.
Here one might object that to exclude from the determination (or
motivation) of a good will all ends dictated by the empirical interest of
the agent as well as all sensible motivation does not necessarily mean to
exclude all mention of an end or even of a motive from the content of the
law. Why would the moral worth of an action not be determined by the
fact that one acts out of respect for the law: ˜˜I should always endeavor to
contribute not only to my own well-being but also to the well-being of
other people™™? Or, if we wanted to formulate the same ideas in terms of
motivation rather than ends, why would the moral worth of an action not
be determined by the fact that one acts out of respect for the law: ˜˜I must
always include compassion in the determining motivations of my
action™™? The imperative would still have a categorical form (indicating
that being the referent of ˜˜I,™™ namely the subject assigning to herself an
obligation, is the reason or sufficient condition for the assignment of
obligation, without any particular end having to function as an added
condition). But it would also have a content, the obligation thus assigned
(the obligation to care about the well-being of others, or the obligation to
include compassion among the determining motives of one™s actions).
By contrast, if the obligation expressed by the categorical imperative is
only the obligation to want a law, no matter what that law is, it seems
suspiciously empty “ a familiar charge against Kant™s categorical
In response it should first be noted that Kant will eventually derive
both of the formulations just proposed from his own formulation of the
categorical imperative: both belong to what the Metaphysics of Morals will
define as duties of virtue.14 So in Kant™s mind, the formulation he
proposes for the categorical imperative provides a principle under

See The Metaphysics of Morals, xx26“30 (˜˜On the Duty of Love to Other Human Beings™™), vi,
450“3; xx45“7 (˜˜On Ethical Duties of Human Beings toward One Another with Regard to
Their Condition™™), vi, 468“72.

which it is possible to unify (and eventually give reasons for) the obliga-
tions that common conscience gives itself, even if the latter does not
explicitly formulate the principle when passing its moral judgments to
determine duties. At the very preliminary stage of Kant™s argument in
section one of Groundwork, Kant™s only claim is that moral judgments (of
the first kind mentioned at the beginning of this chapter: judgments
determining what we ought to do) can, in fact, be unified under the
principle: ˜˜I ought never to act except in such a way that I could also will
that the maxim of my action become a universal law.™™ That is all that the
argument I briefly recapitulated above is meant to show.
That common moral judgments not only can, but must be, so unified,
namely that the principle formulated at the end of section one is indeed
the principle (or one version of the principle) under which we determine
what we are morally obliged to do, is what Kant undertakes to show in
section two of Groundwork. To this end, he no longer rests his argument
merely on the analysis of examples drawn from common moral con-
science. Instead, he leads his reader, step by step, from the initial for-
mulation of the principle of morality (end of section one) to a system of
related formulations (section two), all of which find their starting point in
a philosophical analysis of what it amounts to for a human being to have
a will. Here it will help to say a few words about Kant™s notions of ˜˜will™™
and ˜˜faculty of desire.™™
The faculty of desire is the capacity ˜˜to be by means of one™s representations
the cause of the reality of the objects of these representations™™ (v, 9n, emphasis
Kant™s). Kant also calls it the ˜˜subjective condition of life™™ (ibid.). Animals
typically possess such a capacity. So for example, an animal has the
representation water and associates it with the feeling of pleasure that
would follow from quenching its thirst. The representation causes
the animal™s movement towards the object it covets: it is ˜˜the cause of
the existence of the object of representation,™™ not, of course, in that the
animal produces the water, but at least in that its movement will insure its
proximity to the water. Its faculty of desire is thus indeed the faculty to
be by its representation (here the representation of water associated with
a feeling of pleasure) the cause of the existence of the object of its
representation (here the presence of water).
What is specific to human beings is that concepts, judgments, and
inferences are among the representations that come into play in the
causal relation between the representation of an object and the existence
(or presence) of that object. So for example, in the case of human beings
the representation of water can be accompanied by concepts necessary

to the technical activity of digging a well, the geological knowledge of the
soil, and so on. Here the faculty of desire is a will in that it determines a
deliberate activity, carried out according to rules thought by concepts,
expressed in judgments and connected to each other in inferences. Thus
Kant writes: ˜˜Everything in nature works in accordance with laws. Only a
rational being has the capacity to act in accordance with the representation of
laws, that is, in accordance with principles, or has a will™™ (iv, 412). By
˜˜representation of laws,™™ Kant means two things. First, the representa-
tion of laws is the knowledge of the objective causal connections between
natural events. Our knowledge of these connections is what makes us
capable of representing to ourselves the relations between the ends we are
attempting to bring about by our actions, and the means for achieving
these ends. Second, the representation of laws can be that of laws that are
not descriptive (such as natural laws, which are descriptions of objective
correlations), but prescriptive or normative. When Kant speaks of
imperatives, he is of course talking about the second type of law.
A hypothetical imperative (of the form: ˜˜If I will the end X, then I
ought to will the means necessary to achieve X, insofar as they are
available to me™™) is one of the possible cases of prescriptive laws. But,
as we have seen above, such an imperative does not express a moral
norm. It expresses only a norm of consistency in willing. Since such an
imperative does not command unconditionally, which is what we expect
from a moral imperative, the question is: if there is a categorical impera-
tive (one that commands unconditionally), what is its formulation? Kant
answers this question by providing again, this time in the grammatical
form of the imperative singular (˜˜Act . . . ™™) the very same law that was
expressed, in section one of Groundwork, in the first person of the
indicative (˜˜I ought never to act except in such a way, that . . . [Ich soll
niemals anders verfahren als so, dass . . . ]).™™ The formulation now becomes:
˜˜Act only in accordance with that maxim through which you can at the same time
will that it become a universal law™™ (iv, 421). Kant endeavors to justify
this formulation by drawing on the contrast between hypothetical and
categorical imperative. He writes:
When I think of a hypothetical imperative in general, I do not know
beforehand what it will contain; I do not know this until I am given the
condition. But when I think of a categorical imperative, I know at once
what it contains. For, since the imperative contains, beyond the law, only
the necessity that the maxim be in conformity with this law, while the law
contains no condition to which it would be limited, nothing is left with

<< . .

. 27
( : 33)

. . >>