Tải bản đầy đủ - 0trang
§6. The Categorical Imperative:In What Way Synthetic A Priori?
Critique, we have to distinguish between the a priori in connection with
the understanding and its categories as well as in connection with reason
and the ideas of reason. Understanding and reason (Verstand and Vernunft)
have different roles in the overall structure of knowledge, and the categories
and the ideas of reason have their own distinctive roles. All of this must
be kept track of. And there are also the a priori intuitions of space and time.
By contrast, in the second Critique, there is only the a priori of practical
. Now, for Kant, there are two marks of a priori knowledge: necessity
and universality; and these marks apply to practical as well as to theoretical
(a) Necessity here means practical necessity: that is, what is required
by the principles of pure practical reason. So whatever is required by
the categorical imperative (via the CI-procedure) is practically necessary
(b) As for universality, this means that the requirements in question
hold for all reasonable and rational persons in virtue of their nature as
such persons, independently of any particular conditions of inclination and
circumstances that mark off one reasonable and rational person from another. Kant insists in the preface of the Groundwork (Pref:– [–] )
that “ ‘Thou shalt lie’ could not hold merely for men, other rational beings
having no obligation to abide by it—and similarly with all other genuinely
moral laws; that here consequently the ground of obligation must be looked
for, not in the nature of man nor in the circumstances of the world in which
he is placed, but solely a priori in the concepts of pure reason.”
So far, all this is quite straightforward. The categorical imperative is a
priori as grounded on pure practical reason; it is both practically necessary
and holds universally for all reasonable and rational persons; and the same
is true of particular categorical imperatives when the persons in question
are, like us, ﬁnite persons with needs and hence subject to that imperative.
We must be careful to distinguish, as we have done before, the categorical imperative from the hypothetical. Two differences are these:
First, the hypothetical imperative Kant regards as analytic and not synthetic, and it holds in virtue of empirical rather than pure practical reason.
For Kant, it is simply part of being rational that if we desire the end, we
also desire the (most effective) means. For him, there is no difﬁculty in
understanding how hypothetical imperatives can determine our will (Gr II:
). He doesn’t see it as a problem.
The second difference is that particular hypothetical imperatives are
conditional: they apply to us or not depending on our speciﬁc wants and
. As we might expect from these two differences, the difﬁculty with
understanding how the categorical imperative determines the will concerns
how we are to understand it as a synthetic a priori and practical proposition.
The problem is that in contrast with hypothetical imperatives, they are
unconditional, both in regard to the means they forbid us to adopt in pursuing our ends (these means are restricted by the duties of justice) and in the
obligatory ends they require us to give at least some weight to (these ends
are speciﬁed by the duties of virtue). Recall that to say that particular categorical imperatives are unconditional is to say that they apply to us whatever may be the ends sought by our interests and inclinations. But if that
is so, how is it possible that they can determine our will? What possible
foothold can they have in our person?
There is a parallel here, as Kant indicates in passing, with the problem
of synthetic a priori propositions in the ﬁrst Critique. For in that work Kant
sees no difﬁculty in understanding how empirical concepts apply to objects.
They are abstracted from instances of those concepts met with in experience, and so we are assured of their objective reality: that is, we are assured
that those concepts can apply to objects of experience. After all, the basic
ones have been abstracted from instances to which those concepts already
did apply. But the categories of the understanding, such as the concept of
cause, are a priori and have not been abstracted from objects in experience,
so how can we be assured that they can apply to objects? Thus arises the
problem of the transcendental deduction.
. How are we, then, to understand the categorical imperative as
synthetic a priori? We have already noted the general sense in which this
imperative is a priori as arising from practical reason. But there is also a
special sense in which it is a priori: it formulates the requirements of pure
practical reason and so it is a priori with respect to empirical practical
The thought is that just as the categories of the understanding specify
a priori conditions for the possibility of the experience of objects, the cate[ ]
gorical imperative and particular categorical imperatives to which it leads
impose a priori constraints on the permissible exercise of empirical practical
reason (for our maxims must be acceptable to the CI-procedure). These
constraints are synthetic a priori because:
(i) they are imposed unconditionally on reasonable and rational
(ii) they are imposed on such persons without being derived
from the concept of a person as reasonable and rational.
Hence they are both synthetic (as in [i]) and not analytic (as in [ii]).
This in fact is just what Kant says in Gr II:n. (): “Without presupposing a condition taken from some inclination I connect an action with
the will a priori and therefore necessarily. . . . Here we have a practical
proposition in which the willing of an action is not derived analytically from
some other willing already presupposed . . . but is on the contrary connected
immediately with the concept of the will of a reasonable being as something
which is not contained in this concept.”
I have simpliﬁed the footnote by omitting the parentheses. We can elaborate the footnote as simpliﬁed as follows: Without presupposing any particular ends wanted by natural desires, the CI-procedure connects an action
with what a reasonable and rational person ought to do. Here we have a
practical proposition in which what ought to be done is not derived analytically from other speciﬁc ends already presupposed, as in the case of hypothetical imperatives. Nor, on the other hand, is the action derived analytically from the concept of a reasonable and rational person, again as in the
case of hypothetical imperatives. Rather what such a person ought to do
in a particular case is worked out directly by the CI-procedure without the
intermediary of an end of speciﬁc natural desires.
. While this elaboration accurately describes how the CI-procedure
works, several questions remain to be settled.
(a) First, are particular categorical imperatives (particular duties of justice and duties of virtue) a priori? No, they are not: in arriving at them by
means of the CI-procedure, we rely on certain laws of nature and use various kinds of empirical knowledge about our social world. Of course, they
are unconditional in contrast with hypothetical imperatives. Thus it is no
objection to Kant’s view (as Mill mistakenly thinks it is; see Utilitarianism,
Ch. V, paragraph ) that particular categorical imperatives are not a priori.
What is synthetic a priori is the moral law and the CI-procedure as its
procedural representation as it applies to us.
(b) Second, suppose it is objected that particular categorical imperatives
are really hypothetical and conditional, since the CI-procedure in arriving
at them relies on a conception of true human needs (or another like conception) at step ().
In reply, it sufﬁces to remark (as we said in Kant V) that a conception
of these needs is a special conception of the good introduced to ensure
content for the moral law; introducing it meets a need of pure practical
reason. These considerations distinguish true human needs from particular
inclinations and wants as they arise in everyday life and prompt us to formulate (rational) maxims (hypothetical imperatives) at step (). This reply can
be strengthened by connecting these true human needs with our needs as
rational persons in the order of nature who need society, and so with the
conditions required for a viable social order and necessary for realizing our
rationality. This answer is suggested in section at the end of the preceding
(c) Third, does the CI-procedure take account of the consequences of
everyone’s acting from particular categorical imperatives? Yes, of course:
we see this from the comparison of adjusted social worlds at step (). A
moral conception that took no account of consequences at all (as it is sometimes said of Kant’s view) would be simply mad. The question is always
how to do so.
. To conclude: earlier we said that pure practical reason constructs out
of itself its a priori object, that is, the realm of ends as a commonwealth
of reasonable and rational persons that would exist in the order of nature
when all its members act from the moral law. We can now see that the
phrase “out of itself ” is an exaggeration; the metaphor is a bit out of control.
To correct this, we say instead that pure practical reason, as represented
by the three formulations of the categorical imperative, constructs its object
from the materials (the matter) presented to it by rational maxims at step ().
Alternatively, we can say that the procedure is a selection device: it
accepts some maxims (as generalized precepts) and rejects others according
to whether they meet the criteria of practical reason it incorporates. Like
a mathematical function, the CI-procedure is applied to something (as a
function is applied to a number) to give something else (a corresponding
number), and can’t, to speak accurately, produce its object out of itself.
Finally, at the end of Kant IV, we referred to the important passage in
the second Critique (KP :) where Kant says that the moral law gives
to the sensible world the form of an intelligible world by specifying through
the totality of the precepts it counts as valid the public order of a possible
realm of ends. This public order of a possible realm of ends is the a priori
necessary object of a will determined by the moral law. At KP : there
is another important passage cited already in Kant IV:
Through reason we are conscious of a law to which all our maxims
are subject as though through our will a natural order must arise.
Therefore, this law must be the idea of a supersensuous nature, a nature not empirically given yet possible through freedom; to this nature
we give objective reality, at least in a practical context, because we
regard it as the object of our will as pure rational beings.
The difference, therefore, between the laws of a system of nature
to which the will is subject and a system of nature which is subject to a
will (as far as the relation of the will to its free actions is concerned)
rests on this: in the former, the objects must be the causes of the
conceptions which determine the will, and in the latter, the will is the
cause of the objects. Hence, in the latter the causality of the objects
has its determining ground solely in the pure faculty of reason, which
therefore may be called pure practical reason.
Rather than simply say, as Kant does here, that “the will is the cause
of the objects,” we may elaborate: our will as pure practical reason constructs its own a priori object through the CI-procedure, which object is
the public moral order of a possible realm of ends. So for the members of
a realm of ends, their society as an object in the order of nature is not the
cause of the conception of their society that determines their will. Rather,
in constructing its own a priori object, their pure practical reason is free as
reason is free. It has the freedom of reason.
The Fact of Reason
. Today I discuss the fact of reason, one of the central ideas of Kant’s moral
philosophy. This idea appears for the ﬁrst time in the Critique of Practical
Reason, and focusing upon it is one way to approach that work.
I begin by stating the question that the doctrine of the fact of reason
addresses. Recall that the ﬁrst two chapters of the Groundwork are said by
Kant to be merely analytic: the ﬁrst develops an argument to the moral
law (sketched in Kant I) from our commonsense concept of moral worth
of character, and the second chapter presents the three formulations of the
categorical imperative. But at the end of chapter II (Gr II: [–]), Kant
Any one . . . who takes morality to be something, and not merely a
chimerical idea without truth, must admit the principle we have put
forward. This chapter . . . like the ﬁrst, has been merely analytic. In
order to prove that morality is not simply a phantom of the brain—
a conclusion which follows if the categorical imperative, and with it
the autonomy of the will, is true and is absolutely necessary as an a
priori principle—we require a possible synthetic use of pure practical reason. On such a use we cannot venture without prefacing it by a critique
of this power of reason itself—a critique whose main features, so far
as sufﬁcient for our purpose, we must outline in our ﬁnal chapter.
What is involved here is Kant’s view that for a concept to have objective
reality—that is, for it to be applicable to something and in that sense true—
it is not sufﬁcient that it have an analysis that shows it to be consistent and
intelligible. For all he has done in the ﬁrst two chapters of the Groundwork,
the moral law may still be only a “chimerical idea.” What Kant wants to
show is that the moral law does apply to something, and in particular, that
it applies to us; for if it does, then we can act from that law and not merely
in accordance with it.
. I recall from Kant II two of the essential conditions that we said any
account of the categorical imperative must satisfy.
First, the content condition: the categorical imperative must not be
merely formal but must have sufﬁcent structure to specify requirements
on moral deliberation so that suitably many maxims are shown to be ﬁt
or unﬁt to be made universal law. Otherwise the moral law would be empty
and without content.
Second, the freedom condition: the categorical imperative must represent the moral law as a principle of autonomy, so that from our consciousness of this law as supremely authoritative for us (as reasonable persons),
we can recognize that we can act from the principle of autonomy as a
principle of reason.
The explanation of these two conditions is as follows. The aim of the
second Critique is to show that there is pure practical reason and that it
actually shows its existence in our thought, feeling, and conduct, or in what
Kant calls “the fact of reason.” Put another way, the aim of the second
Critique is to show that pure reason can be practical and can directly determine our will (KP :). But pure reason cannot do this if it is merely formal
and lacking in content. It would be empty; anything we did would satisfy
it. Hence the ﬁrst (content) condition.
The reason for the second condition is this: only if the moral law is a
principle of autonomy in Kant’s sense can this law and our capacity to act
from it disclose our freedom to us, that is, disclose both our independence
of the natural order (negative freedom) and our capacity to act from principles of pure practical reason with a deﬁnite content (positive freedom).
Hence the second (freedom) condition.
. But plainly, since the moral law is simply an idea of reason, the ﬁrst
two conditions do not alone sufﬁce for the objective reality of pure practical
reason. It is further essential that we be conscious of the moral law as
authoritative for us, and that it is possible for us to act from that law, not
merely in accordance with it. Thus there are two further conditions as
Third, the fact of reason condition: our consciousness of the moral law
as supremely authoritative for us as reasonable and rational persons must
be found in our everyday moral thought, feeling, and judgment; and the
moral law must be at least implicitly recognized as such by ordinary human
Fourth, the motivation condition: our consciousness of the moral law
as supremely authoritative for us must be so deeply rooted in our person
as reasonable and rational that this law by itself, when fully known and
understood, can be a sufﬁcient motive for us to act from it, whatever our
Kant holds that if these four conditions are satisﬁed, then there is pure
practical reason. These conditions are both necessary and sufﬁcient. In the
second Critique, he maintains that they are indeed satisﬁed and hence that
pure practical reason exists and that we are free from a practical point of
§. The First Fact of Reason Passage
. There are altogether six fact of reason passages, as I shall call them. For
expository purposes, let’s think of them as extending over a number of
paragraphs, even though a reference to the fact of reason may occur only
once. The fact of reason is mentioned one or more times at eight places
(KP :, , , , , , , ), but some of these belong to the same
passage. Either the whole phrase “das Faktum der Vernunft” or simply “ein
Faktum” occurs; in the latter case, the context makes it clear that the fact
of reason, or a fact related to this fact, is meant. The six fact of reason
passages are these:
(a) The Preface –: the ﬁrst eight paragraphs with the phrase
“ein Faktum” occurring only in paragraph .
(b) In Analytic I –: starting with the statement of Problems
I and II in §§– and continuing to the end of the statement
of Theorem IV but not including the remarks. There are
three explicit references to the fact of reason, all in §.
In the ﬁrst Appendix to Analytic I –: the whole appendix except for the last paragraph. It may be divided into
three parts: paragraphs –, paragraphs –, and paragraphs
In the second Appendix to Analytic I f.
In the Elucidation –: paragraphs –.
In the appendix to the Elucidation –: paragraph .
. I begin with the ﬁrst passage, which opens the Preface, and, as I shall
do in each case, I note only the more important points, as time permits.
This passage introduces the main themes of the work. To appreciate the
signiﬁcance of the ﬁrst sentence, recall that in the Groundwork, Kant had
said that the only foundation of a metaphysic of morals is a critique of pure
practical reason, just as a critique of pure speculative reason is the only
foundation for a metaphysic of nature.
Why this critique is not called a Critique of Pure Practical Reason but
rather simply Critique of Practical Reason, though the parallelism
between it and the critique of speculative reason seems to demand the
latter title, will be sufﬁciently shown in the treatise itself. Its task
is merely to show that there is a pure practical reason, and, in order
to do this, it critically examines reason’s entire practical faculty. If
it succeeds in this task, there is no need to examine the pure faculty
itself to see whether it, like speculative reason, presumptuously
overreaches itself. For if pure reason is actually practical, it will show
its reality and that of its concepts in action [in der Tat], and all disputations which aim to prove its impossibility will be vain. (KP :) (Kant’s
To see what Kant means by the title without the word “pure,” distinguish two senses of critique implicit in his remarks in the Preface to the
ﬁrst Critique (nd. ed., Bxxxii–xxxvii). In one sense, a critique involves giving
an overall account of the concepts and principles of reason as one uniﬁed