become nature and objects of experience, is a law
of the understanding, which can on no account be
surrendered, and from which no single phenomenon
can be exempted ; because in doing this we should
place it outside all possible experience, separate
from all objects of possible experience, and [p. 543]
change it into a mere fiction of the mind or a cobweb
of the brain.

But although this looks merely like a chain of
causes, which in the regressus to its conditions admits
of no absolute totality, this difficulty does not detain
us in the least, because it has already been removed
in the general criticism of the antinomy of reason,



TRANSCENDENTAL DIALECTIC. 469

when, starting from the series of phenomena, it aims
at the unconditioned. Were we to yield to the
illusion of transcendental realism, we should have
neither nature nor freedom. The question therefore
is, whether, if we recognise in the whole series of
events nothing but natural necessity, we may -yet
regard the same event which on one side is an effect
of nature only, on the other side, as an effect of
freedom ; or whether there is a direct contradiction
between these two kinds of causality ?

There can certainly be nothing among phenomenal
causes that could originate a series absolutely and by
itself. Every action, as a phenomenon, so far as it
produces an event, is itself an event, presupposing
another state, in which its cause can be discovered ;
and thus everything that happens is only a continu-
ation of the series, and no beginning, happening by
itself, is possible in it. Actions of natural causes in the
succession of time are therefore themselves [p. 544]
effects, which likewise presuppose causes in the series
of time. A spontaneous action by which something
takes place, which did not exist before, cannot be
expected from the causal nexus of phenomena.

But is it really necessary that, if effects are pheno-
mena, the causality of their cause, which cause itself
is phenomenal, could be nothing but empirical ; or is
it not possible, although for every phenomenal effect
a connection with its cause, according to the laws of
empirical causality, is certainly required, that empirical
causality itself could nevertheless, without breaking
in the least its connection with the natural causes,
represent an 1 effect of a non-empirical and intelligible



47 TRANSCENDENTAL DIALECTIC.

causality, that is, of the action of a cause which, with
respect to phenomena, is original, and in so far not
phenomena], but, with respect to this faculty, intelli-
gible, although, as a link in the chain of nature, it
must be regarded as entirely belonging to the world
of sense %

We require the principle of the causality of pheno-
mena among themselves, in order to be able to look
for and to produce natural conditions, that is, pheno-
menal causes of natural events. If this is admitted
and not weakened by any exceptions, the under-
standing, which in its empirical employment re-
cognises in all events nothing but nature, and [p. 545]
is quite justified in doing so, has really all that it
can demand, and the explanations of physical pheno-
mena may proceed without let or hindrance. The
understanding would not be wronged in the least, if
we assumed, though it be a mere fiction, that some
among the natural causes have a faculty which is
intelligible only, and whose determination to activity
does not rest on empirical conditions, but on mere
grounds of the intellect, if only the 'phenomenal
activity of that cause is in accordance with all the
laws of empirical causality. For in this way the
active subject, as causa phenomenon, would be joined
with nature through the indissoluble dependence of
all its actions, and the noumenon 2 only of that subject
(with all its phenomenal causality) would contain cer-
tain conditions which, if we want to ascend from the
empirical to the transcendental object, would have to

1 Read eine, not einer.

2 It seems better to read noumenon instead of phenomenon.



TRANSCENDENTAL DIALECTIC. 47 1

be considered as intelligible only. For, if only we
follow the rule of nature in that which may be
the cause among phenomena, it is indifferent to us
what kind of ground of those phenomena, and of their
connection, may be conceived to exist in the tran-
scendental subject, which is empirically unknown to
us. This intelligible ground does not touch the
empirical questions, but concerns only, as it would
seem, the thought in the pure understanding ; and
although the effects of that thought and [p. 546]
action of the pure understanding may be discovered
in the phenomena, these have nevertheless to be
completely explained from their phenomenal cause,
according to the laws of nature, by taking their
empirical character as the highest ground of explana-
tion, and passing by the intelligible character, which
is the transcendental cause of the other, as entirely
unknown, except so far as it is indicated by the
empirical, as its sensuous sign. Let us apply this to
experience. Man is one among the phenomena of
the world of sense, and in so far one of the natural
causes the causality of which must be subject to
empirical laws. As such he must therefore have an
empirical character, like all other objects of nature.
We perceive it through the forces and faculties which
he shows in his actions and effects. In the lifeless
or merely animal nature we see no ground for admit-
ting any faculty, except as sensuously conditioned.
Man, however, who knows all the rest of nature
through his senses only, knows himself through mere
apperception also, and this in actions and internal
determinations, which he cannot ascribe to the



47 2 TRANSCENDENTAL DIALECTIC.

impressions of the senses. Man is thus to himself
partly a phenomenon, partly, however, namely with
reference to certain faculties, a purely intelligible ob-
ject, because his actions cannot be ascribed to [p. 547]
the receptivity of the senses. We call these faculties
understanding and reason. It is the latter, in parti-
cular, which is entirely distinguished from all empiri-
cally conditioned forces or faculties, because it weighs
its objects according to ideas, and determines the un-
derstanding accordingly, which then makes an empiri-
cal use of its (by themselves, however pure) concepts.

That our reason possesses causality, or that we at
least represent to ourselves such a causality in it, is
clear from the imperatives which, ' in all practical
matters, we impose as rules on our executive powers.
The ought expresses a kind of necessity and con-
nection with causes, which we do not find elsewhere
in the whole of nature. The understanding can
know in nature only what is present, past, or future.
It is impossible that anything in it ought to he dif-
ferent from what it is in reality, in all these rela-
tions of time. Nay, if we only look at the course of
nature, the ought has no meaning whatever. We
cannot ask, what ought to be in nature, as little as
we can ask, what qualities a circle ought to possess.
We can only ask what happens in it, and what
qualities that which happens has.

This ought expresses a possible action, the ground
of which cannot be anything but a mere concept ;
while in every merely natural action the [p. 548]
ground must always be a phenomenon. Now it is
quite true that the action to which the ought applies



TRANSCENDENTAL DIALECTIC. 473

must be possible under natural conditions, but these
natural conditions do not affect the determination of
the will itself, but only its effects and results among
phenomena. There may be ever so many natural
grounds which impel me to will, and ever so many
sensuous temptations, but they can never produce
the ought, but only a willing which is always con-
ditioned, but by no means necessary, and to which
the ought, pronounced by reason, opposes measure,
ay, prohibition and authority. Whether it be an
object of the senses merely (pleasure), or of pure
reason (the good), reason does not yield to the
impulse that is given empirically, and does not
follow the order of things, as they present them-
selves as phenomena, but frames for itself, with per-
fect spontaneity, a new order according to ideas to
which it adapts the empirical conditions, and accord-
ing to which it declares actions to be necessary, even
though they have not taken place, and, may be, never
will take place. Yet it is presupposed that reason
may have causality with respect to them, for other-
wise no effects in experience could be expected to
result from these ideas.

Now let us take our stand here and admit it at
least as possible, that reason really possesses [p. 549]
causality with reference to phenomena. In that case,
reason though it be, it must show nevertheless an
empirical character, because every cause presupposes
a rule according to which certain phenomena follow
as effects, and every rule requires in the effects
a homogeneousness, on which the concept of cause
(as a faculty) is founded. This, so far as it is derived



474 TRANSCENDENTAL DIALECTIC.

from mere phenomena, may be called the empirical
character, which is permanent, while the effects, ac-
cording to a diversity of concomitant, and in part,
restraining conditions, appear in changeable forms.

Every man therefore has an empirical character
of his (arbitrary) will, which is nothing but a cer-
tain causality of his reason, exhibiting in its pheno-
menal actions and effects a rule, according to which
one may infer the motives of reason and its actions,
both in kind and in degree, and judge of the sub-
jective principles of his will. As that empirical
character itself must be derived from phenomena,
as an effect, and from their rule which is supplied
by experience, all the acts of a man, so far as they
are phenomena, are determined from his empirical
character and from the other concomitant causes,
according to the order of nature ; and if we could
investigate all the manifestations of his will to the
very bottom, there would be not a single [p. 55]
human action which we could not predict with cer-
tainty and recognise from its preceding conditions as
necessary. There is no freedom therefore with refer-
ence to this empirical character, and yet it is only
with reference to it that we can consider man,
when we are merely observing, and, as is the case in
anthropology, trying to investigate the motive causes
of his actions physiologically.

If, however, we consider the same actions with
reference to reason, not with reference to speculative
reason, in order to explain their origin, but solely so
far as reason is the cause which produces them ; in
one word, if we compare actions with reason, with



TRANSCENDENTAL DIALECTIC. 475

reference to practical purposes, we find a rule and
order, totally different from the order of nature. For,
from this point of view, everything, it may be, ought
not to have happened, which according to the course
of nature has happened, and according to its empirical
grounds, was inevitable. And sometimes we find, or
believe at least that we find, that the ideas of reason
have really proved their causality with reference to
human actions as phenomena, and that these actions
have taken place, not because they were determined
by empirical causes, but by the causes of reason.

Now supposing one could say that reason [p. 551]
possesses causality in reference to phenomena, could
the action of reason be called free in that case, as it
is accurately determined by the empirical character
(the disposition) and rendered necessary by it ? That
character again is determined in the intelligible
character (way of thinking). The latter, however,
we do not know, but determine only through pheno-
mena, which in reality give us immediately a know-
ledge of the disposition (empirical character) only 1 .
An action, so far as it is to be attributed to the way
of thinking as its cause, does nevertheless not result
from it according to empirical laws, that is, it is
not preceded by the conditions of pure reason, but
only by its effects in the phenomenal form of the

1 The true morality of actions (merit or guilt), even that of our
own conduct, remains therefore entirely hidden. Our imputations
can refer to the empirical character only. How much of that may
be the pure effect of freedom, how much should be ascribed to
nature only, and to the faults of temperament, for which man is not
responsible, or its happy constitution (merito fortunse), no one can
discover, and no one can judge with perfect justice.



47^ TRANSCENDENTAL DIALECTIC.

internal sense. Pure reason, as a simple intelligible
faculty, is not subject to the form of time, or to the con-
ditions of the succession of time. The causality of rea-
son in its intelligible character does not arise or begin
at a certain time in order to produce an effect ; for
in that case it would be subject to the natural [p. 552]
law of phenomena, which determines all causal series
in time, and its causality would then be nature and
not freedom. What therefore we can say is, that
if reason can possess causality with reference to phe-
nomena, it is a faculty through which the sensuous
condition of an empirical series of effects first begins.
For the condition that lies in reason is not sensuous,
and therefore does itself not begin. Thus we get,
what we missed in all empirical series, namely that
the condition of a successive series of events should
itself be empirically unconditioned. For here the
condition is really outside the series of phenomena,
(in the intelligible), and therefore not subject to
any sensuous condition and temporal determination
through any preceding cause.

Nevertheless the same cause belongs also, in another
respect, to the series of phenomena. Man himself is
a phenomenon. His will has an empirical character,
which is the (empirical) cause of all his actions.
There is no condition, determining man according to
this character, that is not contained in the series of
natural effects and subject to their law, according to
which there can be no empirically unconditioned
causality of anything that happens in time. No
given action therefore (as it can be perceived as a
phenomenon only) can begin absolutely by [p. 553]



TRANSCENDENTAL DIALECTIC. 477

itself. Of pure reason, however, we cannot say that
the state in which it determines the will is preceded
by another in which that state itself is determined.
For as reason itself is not a phenomenon, and not
subject to any of the conditions of sensibility, there
exists in it, even in reference to its causality, no
succession of time, and the dynamical law of nature,
which determines the succession of time according to
rules, cannot be applied to it.

Reason is therefore the constant condition of all
free actions by which man takes his place in the
phenomenal world. Every one of them is determined
beforehand in Ms empirical character, before it be-
comes actual. With regard to the intelligible cha-
racter, however, of which the empirical is only the
sensuous schema, there is neither before nor after;
and every action, without regard to the temporal
relation which connects it with other phenomena,
is the immediate effect of the intelligible character of
pure reason. That reason therefore acts freely, with-
out being determined dynamically, in the chain of
natural causes, by external or internal conditions,
anterior in time. That freedom must then not only
be regarded negatively, as independence of empirical
conditions (for in that case the faculty of reason
would cease to be a cause of phenomena), but should be
determined positively also, as the faculty of beginning
spontaneously a series of events. Hence no- [p. 554]
thing begins in reason itself, and being itself the
unconditioned condition of every free action, reason
admits of no condition antecedent in time above
itself, while nevertheless its effect takes its beginning



478 TRANSCENDENTAL DIALECTIC.

in the series of phenomena, though it can never con-
stitute in that series an absolutely first beginning.

In order to illustrate the regulative principle of
reason by an example of its empirical application, not
in order to confirm it (for such arguments are useless
for transcendental propositions), let us take a voluntary
action, for example, a malicious lie, by which a man
has produced a certain confusion in society, and of
which we first try to find out the motives, and after-
wards try to determine, how far it and its consequences
may be imputed to the offender. With regard to the
first point, one has first to follow up his empirical
character to its very sources, which are to be found
in wrong education, bad society, in part also in the
viciousness of a natural disposition, and a nature in-
sensible to shame, or ascribed to frivolity and heed-
lessness, not omitting the occasioning causes at the
time. In all this the procedure is exactly the same
as in the investigation of a series of determining
causes of a given natural effect. But although one
believes that the act was thus determined, one [p. 555]
nevertheless blames the offender, and not on account
of his unhappy natural disposition, not on account of
influencing circumstances, not even on account of his
former course of fife, because one supposes one might
leave entirely out of account what that course of life
may have been, and consider the past series of con-
ditions as having never existed, and the act itself as
totally unconditioned by previous states, as if the
offender had begun with it a new series of effects,
quite by himself. This blame is founded on ajaw of
reason, reason being considered as a cause which, in-



TRANSCENDENTAL DIALECTIC. 479

dependent of all the before-mentioned empirical con-
ditions, would and should have determined the
behaviour of the man otherwise. Nay, we do not
regard the causality of reason as a concurrent agency
only, but as complete in itself, even though the sen-
suous motives did not favour, but even oppose it.
The action is imputed to a man's intelligible cha-
racter. At the moment when he tells the lie, the
guilt is entirely his ; that is, we regard reason, in
spite of all empirical conditions of the act, as com-
pletely free, and the act has to be imputed entirely
to a fault of reason.

Such an imputation clearly shows that we imagine
that reason is not affected at all by the influences of
the senses, and that it does not change (although its
manifestations, that is the mode in which it [p. 556]
shows itself by its effects, do change) : that no state
in it precedes a following state, in fact, that reason
does not belong to the series of sensuous conditions
which render phenomena necessary, according to laws
of nature. Reason, it is supposed, is present in all the
actions of man, in all circumstances of time, and always
the same ; but it is itself never in time, never in a new
state in which it was not before ; it is determining,
never determined. We cannot ask, therefore, why
reason has not determined itself differently, but only
why it has not differently determined the phenomena
by its causality. And here no answer is really
possible. For a different intelligible character would
have given a different empirical character, and if
we say that, in spite of the whole of his previous
course of life, the offender could have avoided the lie,



480 TRANSCENDENTAL DIALECTIC.

this only means that it was in the power of reason,
and that reason, in its causality, is subject to no
phenomenal and temporal conditions, and lastly, that
the difference of time, though it makes a great differ-
ence in phenomena and their relation to each other,
can, as these are neither things nor causes by them-
selves, produce no difference of action in reference to
reason.

We thus see that, in judging of voluntary [p. 557]
actions, we can, so far as their causality is concerned,
get only so far as the intelligible cause, but not
beyond. We can see that that cause is free, that it
determines as independent of sensibility, and there-
fore is capable of being the sensuously unconditioned
condition of phenomena. To explain why that in-
telligible character should, under present circum-
stances, give these phenomena and this empirical
character, and no other, transcends all the powers of
our reason, nay, all its rights of questioning, as if
we were to ask why the transcendental object of our
external sensuous intuition gives us intuition in space
only and no other. But the problem which we have
to solve does not require us to ask or to answer such
questions. Our problem was, whether freedom is
contradictory to natural necessity in one and the
same action : and this we have sufficiently answered
by showing that freedom may have relation to a
very different kind of conditions from those of nature,
so that the law of the latter does not affect the
former, and both may exist independent of, and un-
disturbed by, each other.



TRANSCENDENTAL DIALECTIC. 48 1

It should be clearly understood that, in what we
lave said, we had no intention of establishing the
'eality of freedom, as one of the faculties which [p. 558]
contain the cause of the phenomenal appearances in
3ur world of sense. For not only would this have
been no transcendental consideration at all, which is
concerned with concepts only, but it could never
have succeeded, because from experience we can never
infer anything out what must be represented in
thought according to the laws of experience. It
was not even our intention to prove the possibility
of freedom, for in this also we should not have suc-
ceeded, because from mere concepts a priori we
can never know the possibility of any real ground or
any causality. We have here treated freedom as a
transcendental idea only, which makes reason imagine
that it can absolutely begin the series of phenomenal
conditions through what is sensuously unconditioned,
but by which reason becomes involved in an antinomy
with its own laws, which it had prescribed to the
empirical use of the understanding. That this anti-
nomy rests on a mere illusion, and that nature does
not contradict the causajity of freedom, that was the
only thing which we could prove, and cared to prove.

IV. [p. 559]

Solution of the Cosmological Idea of the totality of the
dependence of phenomena, with regard to their exist-
ence in general.

In the preceding chapter we considered the changes
in the world of sense in their dynamical succession,
every one being subordinate to another as its cause.
vol. n. 1 i



482 TRANSCENDENTAL DIALECTIC.

Now, however, the succession of states is to serve
only as our guide in order to arrive at an existence
that might be the highest condition of all that is
subject to change, namely the necessary Being. We are
concerned here, not with the unconditioned causality,
but with the unconditioned existence of the substance
itself. Therefore the succession which we have before
us, is properly one of concepts and not of intui-
tions, so far as the one is the condition of the other.

It is easy to see, however, that as everything
comprehended under phenomena is changeable, and
therefore conditioned in its existence, there cannot
be, in the whole series of dependent existence, any
unconditioned link the existence of which might be
considered as absolutely necessary, and that there-"
fore, if phenomena were things by themselves, and
their condition accordingly belonged with the con-
ditioned always to one and the same series of intu-
itions, a necessary being, as the condition of [p. 5 60 ]
the existence of the phenomena of the world of sense,
could never exist.

The dynamical regressus has this peculiar distinc-
tion, as compared with the mathematical, that, as the
latter is only concerned with the composition of parts
in forming a whole, or the division of a whole into its
parts, the conditions of that series must always be
considered as parts of it, and therefore as homo-
geneous and as phenomena, while in the dynamical
regressus, where we are concerned, not with the pos-
sibility of an unconditioned whole, consisting of a
number of given parts, or of an unconditioned part
belonging to a given whole, but with the derivation



TRANSCENDENTAL DIALECTIC. 483

of a state from its cause, or of the contingent exist-
ence of the substance itself from the necessary sub-
stance, it is not required that the condition should
form one and the same empirical series with the
conditioned.

There remains therefore to us another escape from
this apparent antinomy : because both conflicting