Tải bản đầy đủ - 0trang
§3. The Rational Representation of the Origin of Evil
is disclosed in part7 in the weight we give to different kinds of reasons and
in what we do. That our choice of a fundamental character should thus
manifest itself follows from practical freedom (KR B–): namely, that
the free decisions of our practical reason are operative causes in nature.
Now, from the point of view of practical reason, there is no physical
explanation of our intelligible character (KP :); that this is so follows
from our belief in transcendental freedom. Moreover, we know about our
intelligible character only by way of our actual moral thought and conduct
(KR Bf.). From this we can discern, in rough outline anyway, the empirical character of ourselves and others. Armed with this knowledge, we can in
general foresee, or estimate, what others will do in particular circumstances.
(Kant says that could we know that character perfectly, which we cannot,
we could accurately predict our actions in any given conditions.) As thus
foreseen, Kant speaks of our conduct not as free but as necessary and subject
to the laws of nature.
But all this means is that, given the knowledge of people’s beliefs, interests, and circumstances, together with the knowledge of the weight they
give to different kinds of reasons (their empirical character), we can tell
what they will decide and do. But from a practical point of view, when we
are making mutual decisions or asking others for advice, we regard those
empirical characters as expressing the outcome of the deliberations of pure
reason. These deliberations are the upshot of the absolute spontaneity of
reason; expressing our intelligible character, they are taken as not having
a physical or other explanation.
. When in Religion I :f. (f.) Kant speaks of seeking the origin of
evil not in time but merely in rational representation, I think he means
roughly the following. Insofar as our fundamental (intelligible) character
arises from our power of free choice, we are to regard it as our responsibility
alone. Doing this is part of our view of ourselves in acting under the idea
of freedom when complemented, as it should be, by the belief in transcen7. I say in part because our empirical characters are not the work of practical reason alone.
Kant says: “The real morality of actions, their merit or guilt, even that of our own conduct, thus
remains entirely hidden from us. Our imputations can only refer to the empirical character. How
much of this character is ascribable to the pure effect of freedom, how much to mere nature, that
is, to faults of temperament for which there is no responsibility, or to its happy constitution . . . ,
can never be determined; and upon it therefore no perfectly just judgments can be passed” (KR
dental freedom. How our character can be evil—violate the moral order
of the predispositions—is indeed inscrutable (Rel I :f. [f.]): since our
nature is good, not evil but good should have arisen from it, yet long historical experience convinces us to the contrary. But that the choice of that
character has no physical or other causal explanation we must believe from
a practical point of view so as to afﬁrm the absolute spontaneity of our
reason. So to look for such an explanation, when that point of view is
appropriate, contradicts, indeed is an invitation to evade, regarding ourselves as free and responsible.
As far as possible, then, I want to understand Kant’s speaking of reason
as not subject to the form of time (KR Bf.), or not subject to the conditions of time (KP :–), and of not looking for a ﬁrst origin of character
in time (Rel I :f. [f.]), and similarly, as describing how we are to view
ourselves when we act under the idea of freedom, all the while afﬁrming
our transcendental freedom. I believe that he is describing beliefs and attitudes that we are to adopt and cultivate so as to act from the practical point
of view. Alternatively, he is characterizing the form of our self-consciousness
as possessing pure practical reason.
Yet I should say that this interpretation is not generally accepted: the
idea of the intelligible character as permanent and timeless is more often
given a metaphysical interpretation. I believe that doing this is not required
by the text and goes against the conclusions of the Dialectic of the ﬁrst
Critique, as well as Kant’s constantly repeating that what he says about freedom is to be understood from a practical point of view. (On this see the
important Sections VII–VIII of the Dialectic of the second Critique.) Thus,
to interpret as a metaphysical doctrine Kant’s speaking of reason as not
subject to the form of time, yet affecting the course of events in the world,
is not allowed by his text. Or so I think. It also leads to hopeless difﬁculties
for Kant’s view.8
8. For metaphysical interpretations, see Norman Kemp Smith, A Commentary to Kant’s Critique
of Pure Reason (London: Macmillan, ), pp. –, and Wood, “Kant’s Compatibilism,” pp. –
. To illustrate more fully: suppose one were to try to reconcile transcendental freedom with
strict causal (physical) determinism from the ﬁrst state of the world in the following way. Imagine
that there are n noumenal selves, where n is the number of persons who exist at any time in the
order of nature (past, present, and future). Imagine these n selves to make n transcendentally free
choices of an intelligible character. The choice is alleged to be atemporal once and for all, and
orders the three predispositions as the highest principles from which we act. Suppose these n choices
. To continue: in a (largely) realized realm of ends, all persons view
themselves and others as free, that is, as acting under the idea of freedom.
They also suppose that their own and others’ empirical characters as shown
in their public thought and conduct reﬂect, more or less, their intelligible
character for which each accepts responsibility. If we suppose further that
the members of such a realm of ends are lucid before themselves and speak
truthfully, so that they know the reasons from which they act and inform
one another of them as appropriate, then from within the practical point
of view they have no grounds for going behind the reasons they state to
one another when acting under the idea of freedom: those reasons everyone
accepts as the real reasons why they do what they do. Those reasons are
not, for example, viewed as simply rationalizations. There is cause for doing
that only when there is a failure of lucidity or of truthfulness, or another
such failure to act under the idea of freedom (excluded by hypothesis). In
a (largely) realized realm of ends, free public reasons—the reasons people
freely present in good faith to one another—are viewed as real reasons and
are accepted as such.
having been thus made, the divine intelligence then computes the size of the initial mass of the
universe, its shape and the distribution of particles within it, and the appropriate ﬁrst principles of
physics and chemistry, and whatnot, so that beginning with the ﬁrst state of the world, natural
events unfold in accordance with causal laws in such a way that the n intelligible characters (which
were freely selected) are reﬂected in n corresponding empirical characters of persons in various
societies in history.
Now, this fantasy might appear to reconcile transcendental freedom and causal determinism.
It allows us to say that all persons have an equal freedom to determine their intelligible character
and thus to determine their course of life in the world. Moreover, we are set above the order of
nature, and we have a hand in making the whole course of history, since our choice of an intelligible
character imposes a further constraint on the possibilities open to the divine intelligence in creating
One might think this fantasy has a certain usefulness, say in contrasting Kant’s view with
Leibniz’s. For in Leibniz’s conception, all the complete individual concepts lie from eternity in the
divine reason; Judas’ life as a possible is present and known to God in every detail before Judas
was created as part of the best of all possible worlds. Judas never makes a choice of his intelligible
character and the laws that determine his development as a spiritual substance over time—laws
that, while individually distinctive and not reducible to natural laws of science, are laws all the
same. Further, Judas doesn’t know what his own law is, as it must cohere so as to yield preestablished
harmony and the mutual reﬂections of all substances in the most perfect universe, each from its
own point of view.
Whatever illumination we imagine we gain from this comparison, Kant would regard it as
intellectually frivolous. In afﬁrming the convictions of reasonable faith we are not to apply them
theoretically in this way.
This conception of a realm of ends in which everyone is publicly recognized as acting under the idea of freedom is part of the conception of the
person that speciﬁes the conception-dependent desires in Kant’s doctrine.
I come back to this in §.
§. The Manichean Moral Psychology
. In discussing what I shall call the Manichean moral psychology,9 I do not
suppose that Kant ever held such a view or ever clearly formulated it for
himself. I think of it as his tendency to say certain things and to express
them in a manner and tone that, once we ﬁx on them, suggest a certain
moral psychology. The reason for discussing it is that it brings out a development of Kant’s thought in the Religion, or at least a sharper articulation
of his view, once he undertook to consider religion at some length. Once
he did this, then, with his Pietist background, his moral psychology had to
be Augustinian and not Manichean. The manner and tone of the Manichean
tendency are often present, but the explicit doctrine is Augustinian.
The basic idea of the Manichean moral psychology is that we have two
selves: one is the good self we have as intelligences belonging to the intelligible world; and the other is the bad self we have as natural beings belonging
to the sensible world. In speaking of the hardened criminal, Kant says (Gr
This better person he believes himself to be when he transfers himself
to the standpoint of the intelligible world. He is involuntarily constrained to do so by the idea of freedom—that is, of not being dependent on determination by causes in the sensible world; and from this
standpoint he is conscious of possessing a good will which, by his own
admission, constitutes the law for the bad will belonging to him as a
member of the sensible world—a law of whose authority he is aware
even in transgressing it. The moral “I ought” is thus an “I will” for
man as a member of the intelligible world.
9. I take the names of the two psychologies from Saint Augustine’s Confessions. For a time
before his conversion to Christianity he was a Manichee, and his many writings include an account
of the sect’s tenets. His own view is representative of Christian orthodoxy; Kant’s Augustinian moral
psychology is his more orthodox doctrine. I don’t say Kant’s view is orthodox.
The good self has just one predisposition, to use the language of the
Religion, namely, the predisposition to act from the moral law. That is why
for the good self the “I ought” is an “I will.” The only reason we fail to
act from that law as the principle of autonomy is that we are burdened
with natural desires and inclinations, and hence, as Kant says, it must be
the wish of every reasonable person to be wholly free from them (Gr II:
). They are despised by our reason, which, in the consciousness of
its own dignity, is able to achieve mastery over them (Gr II: [– ]).
Indeed, as members of the intelligible world, we do not impute our natural
desires to our proper self at all, since they are, as it were, mere incitements
and solicitations aroused in us by our needs as ﬁnite beings situated in the
order of nature (Gr III: [–]).
The bad or natural self likewise has just one predisposition, at least
insofar as it is fully rational: the predisposition to happiness, or rational selflove. This predisposition in some ways parallels the two predispositions to
animality and to humanity of the Religion, though the account of these
predispositions is not the same. As moved by our natural desires and the
principle of happiness, we must always experience injunctions of the moral
law as a frustration, as a foreign element that blocks the way to what we
want. Hence the bad or natural self lacks moral feeling; what might be
mistaken for such is simply fear and hostility, or suppressed rage aimed at
the self for safety’s sake (KP :–). Finally, the bad self is also a driven
self, since the satisfactions of natural desires are transitory and leave behind
them a greater void than before: the inclinations, even when they are goodnatured like sympathy, are blind and slavish (KP :f.).
. The Manichean moral psychology presents grave difﬁculties for Kant’s
moral doctrine: not only does it commit him to a serious heretical doctrine
at odds with the tenor of his religious thought, but also it would seem to
defeat any satisfactory account of responsibility that is acceptable to us
when we act under the idea of freedom as we assume the practical point
of view. Lacking the idea of the free power of choice, the Manichean psychology cannot provide such an account. For as members of the intelligible
world, we are to see ourselves as all equally having no choice but to act
from the moral law (we have no other predisposition), while as members
of the natural world, we are to see ourselves as all equally having no choice
but to pursue our own happiness. If we viewed ourselves that way from
the practical point of view, it would seem to be impossible to think of
ourselves as all equally responsible while some behave far better than others. We would lack suitable reasons for holding responsible those who
Of course, we could say that while our predisposition to act from the
moral law as members of the intelligible world is constant and pure, our
natural desires and inclinations can be of many different strengths and directions in view of our situation in the natural world. Our actual conduct, we
might say, is settled by the balance of forces generated by those desires,
the strength and direction of which depend on circumstances. People who
do what is right are people whose predisposition to act from the moral law
is not overridden by the direction and force of their natural desires; people
who act badly suffer the opposite fate. We get the familiar moral fatalism
of the Manichean view, which surely Kant must avoid.
. The Augustinian moral psychology overcomes these defects by attributing to the self a free power of choice and enough complexity for a satisfactory account of responsibility. There is no longer a dualism between a good
self and a bad self—two selves each with only a single predisposition, with
the predisposition of one being liable to clash with that of the other—but
one self with three predispositions, all of which are dispositions to good.
Moreover, these predispositions have an appropriate moral order suitable
for free persons capable of acting from pure practical reason. The origin of
moral evil, then, lies not in a bad self with its natural desires but solely in
the free power of choice, which may change the moral order of the dispositions and determine what we count as appropriate reasons in deciding what
As for how to understand these moral psychologies, I believe that we
should see them as ways of conceiving of our moral nature from the practical point of view, and so how to think of ourselves when we are acting
under the idea of freedom. They are not, as such, empirical accounts of
human psychology, nor subject to the usual criteria of empirical truth.
Rather, they belong to different ways of characterizing the reﬂective selfconsciousness of pure practical reason; as such they embody different ideal
conceptions of the person.
Thus, as we have said, when acting under the idea of freedom on the
Augustinian view, we cannot regard our fundamental character (the order[ ]