predicates, peculiar to external appearance. The
312 TRANSCENDENTAL DIALECTIC.
[predicates of the internal sense, on the con- [p. 359]
trary, such as representation, thinking, etc., are by
no means contradictory to it, so that really, even if
we admit its simplicity of nature, the human soul is
by no means sufficiently distinguished from matter,
so far as its substratum is concerned, if (as it ought
to be) matter is considered as a phenomenon only.
If matter were a thing by itself, it would, as a com-
posite being, be totally different from the soul, as a
simple being. But what we call matter is an external
phenomenon only, the substratum of which cannot
possibly be known by any possible predicates. I can
therefore very well suppose that that substratum is
simple, although in the manner in which it affects our
senses it produces in us the intuition of something
extended, and therefore composite, so that the sub-
stance which, with reference to our external sense,
possesses extension, might very well by itself possess
thoughts which can be represented consciously by its
own internal sense. In such wise the same thing
which in one respect is called corporeal, would in
another respect be at the same time a thinking being,
of which though we cannot see its thoughts, we can
yet see the signs of them phenomenally. Thus the
expression that souls only (as a particular class of sub-
stances) think, would have to be dropt, and we should
return to the common expression that men [p. 360]
think, that is, that the same thing which, as an ex-
ternal phenomenon, is extended, is internally, by itself,
a subject, not composite, but simple and intelligent.
But without indulging in such hypotheses, we may
make this general remark, that if I understand by
TRANSCENDENTAL DIALECTIC. 313
[soul a being by itself, the very question would be
absurd, whether the soul be homogeneous or not with
matter which is not a thing by itself, but only a class
of representations within us ; for so much at all
events must be clear, that a thing by itself is of a
different nature from the determinations which con-
stitute its state only.
If, on the contrary, we compare the thinking I,
not with matter, but with that object of the intellect
that forms the foundation of the external phenomena
which we call matter, then it follows, as we know
nothing whatever of the matter, that we have no
right to say that the soul by itself is different from
it in any respect.
The simple consciousness is not therefore a know-
ledge of the simple nature of our subject, so that we
might thus distinguish the soul from matter, as a
composite being.
If therefore, in the only case where that con-
cept might be useful, namely in comparing myself
with objects of external experience, it is impossible
to determine the peculiar and distinguishing charac-
teristics of its nature, what is the use, if we [p. 3 61 ]
pretend to know that the thinking I, or the soul (a
name for the transcendental object of the internal
sense), is simple \ Such a proposition admits of no
application to any real object, and cannot therefore
enlarge our knowledge in the least.
Thus collapses the whole of rational psychology,
with its fundamental support, and neither here nor
elsewhere can we hope by means of mere concepts
(still less through the mere subjective form of all our
314 TRANSCENDENTAL DIALECTIC.
[concepts, that is, through our consciousness) and
without referring these concepts to a possible ex-
perience, to extend our knowledge, particularly as
even the fundamental concept of a simple nature is
such that it can never be met with in experience,
so that no chance remains of arriving at it as a
concept of objective validity.
The Third Paralogism of Personality.
Whatever is conscious of the numerical identity
of its own self at different times, is in so far a
person.
Now the Soul, &c.
Therefore the Soul is a person.
Criticism of the Third Paralogism of Transcendental
Psychology.
Whenever I want to know by experience the
numerical identity of an external object, I [p. 362]
shall have to attend to what is permanent in that
phenomenon to which, as the subject, everything else
refers as determination, and observe the identity
of the former during the time that the latter is
changing. I myself, however, am an object of the
internal sense, and all time is but the form of the
internal sense. I therefore refer each and all of my
successive determinations to the numerically identical
self ; and this in all time, that is, in the form of the
inner intuition of myself. From this point of view,
the personality of the soul should not even be con-
sidered as inferred, but as an entirely identical
proposition of self-consciousness in time, this being
TRANSCENDENTAL DIALECTIC. 315
[really the reason why it is valid a priori. For it
really says no more than this : that during the whole
time, while I am conscious of myself, I am conscious
of that time as belonging to the unity of myself; and
it comes to the same thing whether I say that this
whole time is within me as an individual unity, or
that I with numerical identity am present in all
that time.
In my own consciousness, therefore, the identity of
person is inevitably present. But if I consider my-
self from the point of view of another person (as an
object of his external intuition), then that external
observer considers me, first of all, in time, for in the
apperception time is really represented in me only.
Though he admits, therefore, the I, which at all times
accompanies all representations in my con- [p. 363]
sciousness, and with entire identity, he will not yet
infer from it the objective permanence of myself. For
as in that case the time in which the observer places
me is not the time of my own, but of his sensibility,
it follows that the identity which is connected with
my consciousness is not therefore connected with his,
that is, with the external intuition of my subject.
The identity of my consciousness at different times
is therefore a formal condition only of my thoughts
and their coherence, and proves in no way the
numerical identity of my subject, in which, in spite
of the logical identity of the I, such a change may
have passed as to make it impossible to retain its
identity, though we may still attribute to it the same
name of I, which in every other state, and even in the
change of the subject, might yet retain the thought
316 TRANSCENDENTAL DIALECTIC.
[of the preceding and hand it over to the subsequent
subject K
Although the teaching of some old schools [p. 364]
that everything is in a flux, and nothing in the world
permanent, cannot be admitted, if we admit sub-
stances, yet it must not be supposed that it can be
refuted by the unity of self-consciousness. For we
ourselves cannot judge from our own consciousness
whether, as souls, we are permanent or not, because
we reckon as belonging to our own identical self that
only of which we are conscious, being constrained to
admit that, during the whole time of which we are
conscious, we are one and the same. From the point
of view of a stranger, however, such a judgment
would not be valid, because, perceiving in the soul no
permanent phenomena, except the representation of
the I, which accompanies and connects them all, we
cannot determine whether that I (being a mere
thought) be not in the same state of flux as the other
thoughts which are chained together by the I. [p. 365]
It is curious, however, that the personality and
1 An elastic ball, which impinges on another in a straight line com-
municates to it its whole motion, and therefore (if we only consider
the places in space) its whole state. If then, in analogy with such
bodies, we admit substances of which the one communicates to the
other representations with consciousness, we could imagine a whole
series of them, in which the first communicates its state and its
consciousness to the second, the second its own state with that of
the first substance to a third, and this again all the states of the
former, together with its own, and a consciousness of them, to
another. That last substance would be conscious of all the states
of the previously changed substances, as of its own, because all of
them had been transferred to it with the consciousness of them ;
but for all that it would not have been the same person in all those
states.
TRANSCENDENTAL DIALECTIC. 317
[what is presupposed by it, namely, the permanence
and substantiality of the soul, has now to be proved
first. For if we could presuppose these, there
would follow, if not the permanence of conscious-
ness, yet the possibility of a permanent consciousness
in one and the same subject, and this is sufficient to
establish personality which does not cease at once,
because its effect is interrupted at the time. This
permanence, however, is by no means given us
before the numerical identity of ourself, which we
infer from identical apperception, but is itself inferred
from it, so that, according to rule, the concept of
substance,* which alone is empirically useful, would
have to follow first upon it. But as the identity of
person follows by no means from the identity of the
I, in the consciousness of all time in which I perceive
myself, it follows that we could not have founded
upon it the substantiality of the soul.
Like the concept of substance and of the simple,
however, the concept of personality also may remain,
so long as it is used as transcendental only, that is, as
a concept of the unity of the subject which is otherwise
unknown to us, but in the determinations of which
there is an uninterrupted connection by apperception.
In this sense such a concept is necessary for practical
purposes and sufficient, but we can never pride our-
selves on it as helping to expand our know- [p. 366]
ledge of our self by means of pure reason, which only
deceives us if we imagine that we can arrive at
an uninterrupted continuance of the subject from
the mere concept of the identical self. That concept
is only constantly turning round itself in a circle, and
318 TRANSCENDENTAL DIALECTIC.
[does not help ns as with respect to any question
which aims at synthetical knowledge. What matter
may be as a thing by itself (a transcendental object)
is entirely unknown to us ; though we may observe
its permanence as a phenomenon, since it is repre-
sented as something external. When however I
wish to observe the mere I during the change of all
representations, I have no other correlative for my
comparisons but again the I itself, with the general
conditions of my consciousness. I cannot therefore
give any but tautological answers to all questions,
because I put my concept and its unity in the
place of the qualities that belong to me as an
object, and thus really take for granted what was
wished to be known.
The Fourth Paralogism of Ideality (with regard to
external relations).
That, the existence of which can only be inferred as
a cause of given perceptions, has a doubtful existence
only : [p. 367]
All external phenomena are such that their
existence cannot be perceived immediately, but that
we can only infer them as the cause of given
perceptions :
Therefore the existence of all objects of the
external senses is doubtful. This uncertainty I
call the ideality of external phenomena, and the
doctrine of that ideality is called idealism ; in com-
parison with which the other doctrine, which main-
tains a possible certainty of the objects of the external
senses, is called dualism.
TRANSCENDENTAL DIALECTIC. 319
[Criticism of the Fourth Paralogism of Transcendental
Psychology.
We shall first have to examine the premisses.
We are perfectly justified in maintaining that that
only which is within ourselves can be perceived
immediately, and that my own existence only can
be the object of a mere perception. The existence
of a real object therefore outside me (taking this
word in its intellectual meaning) can never be given
directly in perception, but can only be added in
thought to the perception, which is a modification of
the internal sense, and thus inferred as its external
cause. Hence Cartesius was quite right in limiting
all perception, in the narrowest sense, to the propo-
sition, I (as a thinking being) am. For it must [p. 368]
be clear that, as what is without is not within me, I
cannot find it in my apperception ; nor hence in any
perception which is in reality a determination of
apperception only.
In the true sense of the word, therefore, I can
never perceive external things, but only from my
own internal perception infer their existence, taking
the perception as an effect of which something
external must be the proximate cause. An inference,
however, from a given effect to a definite cause is
always uncertain, because the effect may be due
to more than one cause. Therefore in referring a
perception to its cause, it always remains doubtful
whether that cause be internal or external ; whether
in fact all so-called external perceptions are not a
mere play of our external sense, or point to real
320 TRANSCENDENTAL DIALECTIC.
[external objects as their cause. At all events the
existence of the latter is inferential only, and liable
to all the dangers of inferences, while the object of
the internal sense (I myself with all my represen-
tations) is perceived immediately, and its existence
cannot be questioned.
It must not be supposed, therefore, that an idealist
is he who denies the existence of external objects of
the senses ; all he does is to deny that it is known
by immediate perception, and to infer that we [p. 369]
can never become perfectly certain of their reality by
any experience whatsoever.
Before I expose the deceptive illusion of our para-
logism, let me remark that we must necessarily dis-
tinguish two kinds of idealism, the transcendental
and the empirical. Transcendental idealism teaches
that all phenomena are representations only, not
things by themselves, and that space and time
therefore are only sensuous forms of our intuition,
not given determinations or conditions of objects, as
things by themselves. Opposed to this transcend-
ental idealism, is a transcendental realism, which
considers space and time as something in itself (inde-
pendent of our sensibility). Thus the transcendental
realist represents all external phenomena (admitting
their reality) as things by themselves, existing inde-
pendently of us and our sensibility, and therefore,
according to pure concepts of the understanding,
existing outside us. It is this transcendental realist
who afterwards acts the empirical idealist, and who,
after wrongly supposing that the objects of the
senses, if they are to be external, must have an
TRANSCENDENTAL DIALECTIC. 321
[existence by themselves, and without our senses, yet
from this point of view considers all our sensuous
representations insufficient to render certain the
reality of their objects.
The transcendental idealist, on the contrary, [p. 370]
may well be an empirical realist, or, as he is called, a
dualist; that is, he may admit the existence of
matter, without taking a step beyond mere self-con-
sciousness, or admitting more than the certainty of
representations within me, that is the cogito, ergo sum.
For as he considers matter, and even its internal
possibility, as a phenomenon only, which, if separated
from our sensibility, is nothing, matter with him is
only a class of representations (intuition) which are
called external, not as if they referred to objects
external by themselves, but because they refer per-
ceptions to space, in which everything is outside
everything else, while space itself is inside Us.
We have declared ourselves from the very be-
ginning in favour of this transcendental idealism.
In our system, therefore, we need not hesitate to
admit the existence of matter on the testimony of
mere self-consciousness, arid to consider it as estab-
lished in the same manner as the existence of my
self, as a thinking being. I am conscious of my
representations, and hence they exist as well as I my
self, who has these representations. External objects,
however (bodies), are phenomena only, therefore
nothing but a class of my representations, the objects
of which are something by means of these representa-
tions only, and apart from them nothing. External
things, therefore, exist by the same right [p. 37 r ]
VOL. II. Y
32 2 TRANSCENDENTAL DIALECTIC.
[as I myself, both on the immediate testimony of my
self-consciousness, with this difference only, that the
representation of myself, as a thinking subject, is re-
ferred to the internal sense only, while the repre-
sentations which indicate extended beings are referred
to the external sense also. With reference to the
reality of external objects, I need as little trust to
inference, as with reference to the reality of the ob-
ject of my internal sense (my thoughts), both being
nothing but representations, the immediate percep-
tion (consciousness) of which is at the same time a
sufficient proof of their reality.
The transcendental idealist is, therefore, an em-
pirical realist, and allows to matter, as a phenomenon,
a reality which need not be inferred, but may be
immediately perceived. The transcendental realism,
on the contrary, is necessarily left in doubt, and
obliged to give way to empirical idealism, because it
considers the objects of the external senses as some-
thing different from the senses themselves, taking-
mere phenomena as independent beings, existing
outside us. And while with the very best conscious-
ness of our representation of these things, it is far
from certain that, if a representation exists, its cor-
responding object must exist also, it is clear that in
our system external things, that is, matter in all its
shapes and changes, are nothing but mere [p. 372]
phenomena, that is, representations within us, of the
reality of which we are immediately conscious.
As, so far as I know, all psychologists who believe
in empirical idealism are transcendental realists, they
have acted no doubt quite consistently, in ascribing
TRANSCENDENTAL DIALECTIC. 323
[great importance to empirical idealism, as one of
the problems from which human reason could hardly
extricate itself. For indeed, if we consider external
phenomena as representations produced inside us by
their objects, as existing as things by themselves out-
side us, it is difficult to see how their existence could
be known otherwise but through a syllogism from
effect to cause, where it must always remain doubtful,
whether the cause be within or without us. Now
we may well admit that something which, taken
transcendentally, is outside us, may be the cause of
our external intuitions, but this can never be the
object which we mean by the representations of
matter and material things ; for these are phenomena
only, that is, certain kinds of representation existing
always within us, and the reality of which depends
on our immediate consciousness, quite as much as
the consciousness of my own thoughts. The tran-
scendental object is unknown equally in regard to
internal and external intuition.
Of this, however, we are not speaking at [p. 373]
present, but only of the empirical object, which is
called external, if represented in space, and internal,
when represented in temporal relations only, both
space and time being to be met with nowhere except
in ourselves.
The expression, outside us, involves however an
inevitable ambiguity, because it may signify either,
something which, as a thing by itself, exists apart from
us, or what belongs to outward appearance only. In
order, therefore, to remove all uncertainty from that
concept, taken in the latter meaning (which alone
Y 2
324 TRANSCENDENTAL DIALECTIC.
[affects the psychological question as to the reality
of our external intuition) we shall distinguish em-
pirically external objects from those that may be
called so in a transcendental sense, by calling the
former simply things occurring in space.,
Space and time are no doubt representations a
priori, which dwell in us as forms of our sensuous
intuition, before any real object has determined our
senses by means of sensation, enabling them to re-
present the object under those sensuous conditions.
But this something, material or real, that is to be
seen in space, presupposes necessarily perception, and
cannot be fancied or produced by means of imagina-
tion without that perception, which indicates the
reality of something in space. It is sensation, there-
fore, that indicates reality in space and time, [p. 374]
according as it is related to the one or the other mode
of sensuous intuition. If sensation is once given
(which, if referring to an object in general, and not
specialising it, is called perception), many an object
may be put together in imagination from the mani-
fold materials of perception, which has no empirical
place in space or time, but in imagination only. This
admits of no doubt, whether we take the sensations of
pain and pleasure, or the external ones of colour,
heat, &c. ; it is always perception by which the
material for thinking any objects of external intuition
must first be supplied. This perception, therefore
(to speak at present of external intuitions only), re-
presents something real in space. For, first, percep-
tion is the representation of a reality, while space is
the representation of a mere possibility of existence.
TRANSCENDENTAL DIALECTIC. 325
[Secondly, this reality is represented before the ex-
ternal sense, that is, in space. Thirdly, space itself
is nothing but mere representation, so that nothing
in it can be taken as real, except what is represented
in it 1 ; or, vice versa, whatever is given in it, [p. 375]
that is, whatever is represented by perception, is also
real in it, because, if it were not real in it, that is,
given immediately by empirical intuition, it could not
be created by fancy, the real of intuition being un-
imaginable a priori.
Thus we see that all external percej^ion proves
immediately something real in space, or rattier is that
real itself. Empirical realism is therefore perfectly
true, that is, something real in space always cor-
responds to our external intuitions. Space itself, it is
true, with all its phenomena, as representations, exists
within me only, but the real or the material of all ob-
jects of intuition is nevertheless given in that space,
independent of all fancy or imagination ; nay, it is
impossible that in that space anything outside us (in
a transcendental sense) could be given, because space
itself is nothing outside our sensibility. The strict-
est idealist, therefore, can never require that we
should prove that the object without us (in [p. 376]
its true meaning) corresponds to our perception. For
1 "We must well master this paradoxical, but quite correct pro-
position, that nothing can be in space, except what is represented
in it. For space itself is nothing but representation, and whatever
is in it must therefore be contained in that representation. There
is nothing whatever in space, except so far as it is really repre-
sented in it. That a thing can exist only in the representation of
it, may no doubt sound strange ; but will lose its strangeness if we
consider that the things with which we have to deal, are not things
by themselves, but phenomena only, that is, representations.
326 TRANSCENDENTAL DIALECTIC.
[granted there are such objects, they could never be
represented and seen, as outside us, because this pre-
supposes space, and the reality in space, as a mere
representation, is nothing but the perception itself.
It thus follows, that what is real in external pheno-
mena, is real in perception only, and cannot be given
in any other way.
From such perceptions, whether by mere play of
fancy or by experience, knowledge of objects can be
produced, and here no doubt deceptive representa-
tions may arise, without truly corresponding objects,
the deception being due, either to illusions of imagina-
tion (in dreams), or to a fault of judgment (the so-
called deceptions of the senses). In order to escape
from these false appearances, one has to follow the
rule that, whatever is connected according to empirical
laws with a perception, is real. This kind of illusion,
however, and its prevention, concerns idealism as
well as dualism, since it affects the form of experience
only. In order to refute empirical idealism and its
unfounded misgivings as to the objective reality of