Immanuel Kant's

trans. by Norman Kemp Smith

Table of Contents Previous Section Next Section Search Engine

P 436
THE ANTINOMY OF PURE REASON
Section 5
SCEPTICAL REPRESENTATION OF THE COSMOLOGICAL
QUESTIONS IN THE FOUR TRANSCENDENTAL IDEAS
We should of ourselves desist from the demand that our
questions be answered dogmatically, if from the start we
understood that whatever the dogmatic answer might turn out
to be it would only increase our ignorance, and cast us from
one inconceivability into another, from one obscurity into
another still greater, and perhaps even into contradictions. If
our question is directed simply to a yes or no, we are well
advised to leave aside the supposed grounds of the answer, and
first consider what we should gain according as the answer is
in the affirmative or in the negative. Should we then find that
in both cases the outcome is mere nonsense, there will be good
reason for instituting a critical examination of our question, to
determine whether the question does not itself rest on a groundless
presupposition, in that it plays with an idea the falsity of
which can be more easily detected through study of its application
and consequences than in its own separate representation.
This is the great utility of the sceptical mode of dealing with
the questions which pure reason puts to pure reason. By its
means we can deliver ourselves, at but a small cost, from a
great body of sterile dogmatism, and set in its place a sober
critique, which as a true cathartic will effectively guard us
against such groundless beliefs and the supposed polymathy
to which they lead.
If therefore, in dealing with a cosmological idea, I were
able to appreciate beforehand that whatever view may be
taken of the unconditioned in the successive synthesis of 
appearances, it must either be too large or too small for any 
concept of the understanding, I should be in a position to 
understand that since the cosmological idea has no bearing save
upon an object of experience which has to be in conformity
with a possible concept of the understanding, it must be
P 437
entirely empty and without meaning; for its object, view it as
we may, cannot be made to agree with it. This is in fact the
case with all cosmical concepts; and this is why reason, so
long as it holds to them, is involved in an unavoidable
antinomy. For suppose: --
First, that the world has no beginning: it is then too large
for our concept, which, consisting as it does in a successive
regress, can never reach the whole eternity that has elapsed.
Or suppose that the world has a beginning, it will then, in the
necessary empirical regress, be too small for the concept of
the understanding. For since the beginning still presupposes a
time which precedes it, it is still not unconditioned; and the law
of the empirical employment of the understanding therefore
obliges us to look for a higher temporal condition; and the
world [as limited in time] is therefore obviously too small for
this law.
This is also true of the twofold answer to the question
regarding the magnitude of the world in space. If it is infinite
and unlimited, it is too large for any possible empirical 
concept. If it is finite and limited, we have a right to ask what
determines these limits. Empty space is no self-subsistent
correlate of things, and cannot be a condition at which we
could stop; still less can it be an empirical condition, forming
part of a possible experience. (For how can there be any 
experience of the absolutely void? ) And yet to obtain absolute
totality in the empirical synthesis it is always necessary that
the unconditioned be an empirical concept. Consequently, a
limited world is too small for our concept.
Secondly, if every appearance in space (matter) consists of
infinitely many parts, the regress in the division will always
be too great for our concept; while if the division of space is
to stop at any member of the division (the simple), the regress
will be too small for the idea of the unconditioned. For this
member always still allows of a regress to further parts 
contained in it.
Thirdly, if we suppose that nothing happens in the world
save in accordance with the laws of nature, the causality of
the cause will always itself be something that happens, making
necessary a regress to a still higher cause, and thus a continuation
of the series of conditions a parte priori without end.
P 438
Nature, as working always through efficient causes, is thus
too large for any of the concepts which we can employ in the
synthesis of cosmical events.
If, in certain cases, we admit the occurrence of self-caused
events, that is, generation through freedom, then by an 
unavoidable law of nature the question 'why' still pursues us,
constraining us, in accordance with the law of causality
[which governs] experience, to pass beyond such events; and
we thus find that such totality of connection is too small for
our necessary empirical concept.
Fourthly, if we admit an absolutely necessary being
(whether it be the world itself, or something in the world, or
the cause of the world), we set it in a time infinitely remote
from any given point of time, because otherwise it would be
dependent upon another and antecedent being. But such an
existence is then too large for our empirical concept, and is
unapproachable through any regress, however far this be
carried.
 If, again, we hold that everything belonging to the world
(whether as conditioned or as condition) is contingent, any
and every given existence is too small for our concept. For
we are constrained always still to look about for some other
existence upon which it is dependent.
We have said that in all these cases the cosmical idea is
either too large or too small for the empirical regress, and
therefore for any possible concept of the understanding. We
have thus been maintaining that the fault lies with the idea, in
being too large or too small for that to which it is directed,
namely, possible experience. Why have we not expressed ourselves
in the opposite manner, saying that in the former case
the empirical concept is always too small for the idea, and in
the latter too large, and that the blame therefore attaches to
the empirical regress? The reason is this. Possible experience
is that which can alone give reality to our concepts; in its
absence a concept is a mere idea, without truth, that is, without
relation to any object. The possible empirical concept is 
therefore the standard by which we must judge whether the idea
is a mere idea and thought-entity, or whether it finds its object
in the world. For we can say of anything that it is too large
P 439
or too small relatively to something else, only if the former is
required for the sake of the latter, and has to be adapted to it.
Among the puzzles propounded in the ancient dialectical
Schools was the question, whether, if a ball cannot pass
through a hole, we should say that the ball is too large or the
hole too small. In such a case it is a matter of indifference
how we choose to express ourselves, for we do not know which
exists for the sake of the other. In the case, however, of a man
and his coat, we do not say that a man is too tall for his coat,
but that the coat is too short for the man.
We have thus been led to what is at least a well-grounded
suspicion that the cosmological ideas, and with them all the
mutually conflicting pseudo-rational assertions, may perhaps
rest on an empty and merely fictitious concept of the manner
in which the object of these ideas is given to us; and this 
suspicion may set us on the right path for laying bare the illusion
which has so long led us astray.
THE ANTINOMY OF PURE REASON
Section 6
TRANSCENDENTAL IDEALISM AS THE KEY TO THE
SOLUTION OF THE COSMOLOGICAL DIALECTIC
We have sufficiently proved in the Transcendental Aesthetic
that everything intuited in space or time, and therefore all
objects of any experience possible to us, are nothing but 
appearances, that is, mere representations, which, in the manner
in which they are represented, as extended beings, or as series of
alterations, have no independent existence outside our thoughts.
This doctrine I entitle transcendental idealism. The realist, in
the transcendental meaning of this term, treats these modifications
of our sensibility as self-subsistent things, that is, treats
mere representations as things in themselves.
++ I have also, elsewhere, sometimes entitled it formal idealism,
to distinguish it from material idealism, that is, from the usual type
of idealism which doubts or denies the existence of outer things
themselves.
P 439
It would be unjust to ascribe to us that long-decried
P 440
empirical idealism, which, while it admits the genuine reality
of space, denies the existence of the extended beings in it, or
at least considers their existence doubtful, and so does not
in this regard allow of any properly demonstrable distinction
between truth and dreams. As to the appearances of inner
sense in time, empirical idealism finds no difficulty in regarding
them as real things; indeed it even asserts that this inner
experience is the sufficient as well as the only proof of the
actual existence of its object (in itself, with all this time-
determination).
 Our transcendental idealism, on the contrary, admits the
reality of the objects of outer intuition, as intuited in space, and
of all changes in time, as represented by inner sense. For since
space is a form of that intuition which we entitle outer, and
since without objects in space there would be no empirical 
representation whatsoever, we can and must regard the extended
beings in it as real; and the same is true of time. But this space
and this time, and with them all appearances, are not in themselves
things; they are nothing but representations, and cannot
exist outside our mind. Even the inner and sensible intuition
of our mind (as object of consciousness) which is represented
as being determined by the succession of different states in
time, is not the self proper, as it exists in itself -- that is, is not
the transcendental subject -- but only an appearance that has
been given to the sensibility of this, to us unknown, being.
This inner appearance cannot be admitted to exist in any such
manner in and by itself; for it is conditioned by time, and time
cannot be a determination of a thing in itself. The empirical
truth of appearances in space and time is, however, sufficiently
secured; it is adequately distinguished from dreams, if both
dreams and genuine appearances cohere truly and completely
in one experience, in accordance with empirical laws.
 The objects of experience, then, are never given in themselves,
but only in experience, and have no existence outside it.
That there may be inhabitants in the moon, although no one
has ever perceived them, must certainly be admitted. This,
however, only means that in the possible advance of experience
we may encounter them. For everything is real which
stands in connection with a perception in accordance with the
P 441
laws of empirical advance. They are therefore real if they
stand in an empirical connection with my actual consciousness,
although they are not for that reason real in themselves, that
is, outside this advance of experience.
Nothing is really given us save perception and the empirical
advance from this to other possible perceptions. For the
appearances, as mere representations, are in themselves real
only in perception, which perception is in fact nothing but the
reality of an empirical representation, that is, appearance. To
call an appearance a real thing prior to our perceiving it, either
means that in the advance of experience we must meet with
such a perception, or it means nothing at all. For if we were
speaking of a thing in itself, we could indeed say that it exists
in itself apart from relation to our senses and possible 
experience. But we are here speaking only of an appearance in space
and time, which are not determinations of things in themselves
but only of our sensibility. Accordingly, that which is in
space and time is an appearance; it is not anything in itself
but consists merely of representations, which, if not given in
us -- that is to say, in perception -- are nowhere to be met with.
The faculty of sensible intuition is strictly only a receptivity,
a capacity of being affected in a certain manner with
representations, the relation of which to one another is a pure
intuition of space and of time (mere forms of our sensibility),
and which, in so far as they are connected in this manner in
space and time, and are determinable according to laws of the
unity of experience, are entitled objects. The non-sensible cause
of these representations is completely unknown to us, and cannot
therefore be intuited by us as object. For such an object would
have to be represented as neither in space nor in time (these
being merely conditions of sensible representation), and apart
from such conditions we cannot think any intuition. We may,
however, entitle the purely intelligible cause of appearances in
general the transcendental object, but merely in order to have
something corresponding to sensibility viewed as a receptivity.
To this transcendental object we can ascribe the whole extent
and connection of our possible perceptions, and can say that it
is given in itself prior to all experience. But the appearances,
P 442
while conforming to it, are not given in themselves, but only in
this experience, being mere representations, which as perceptions
can mark out a real object only in so far as the perception
connects with all others according to the rules of the unity of
experience. Thus we can say that the real things of past time
are given in the transcendental object of experience; but they
are objects for me and real in past time only in so far as I represent
to myself (either by the light of history or by the guiding-
clues of causes and effects) that a regressive series of possible
perceptions in accordance with empirical laws, in a word, that
the course of the world, conducts us to a past time-series as 
condition of the present time -- a series which, however, can be 
represented as actual not in itself but only in the connection of a
possible experience. Accordingly, all events which have taken
place in the immense periods that have preceded my own existence
mean really nothing but the possibility of extending the
chain of experience from the present perception back to the
conditions which determine this perception in respect of time.
If, therefore, I represent to myself all existing objects of
the senses in all time and in all places, I do not set them in
space and time [as being there] prior to experience. This
representation is nothing but the thought of a possible 
experience in its absolute completeness. Since the objects are
nothing but mere representations, only in such a possible
experience are they given. To say that they exist prior to
all my experience is only to assert that they are to be met
with if, starting from perception, I advance to that part of
experience to which they belong. The cause of the empirical
conditions of this advance (that which determines what members
I shall meet with, or how far I can meet with any such
in my regress) is transcendental, and is therefore necessarily
unknown to me. We are not, however, concerned with this
transcendental cause, but only with the rule of the advance in
the experience in which objects, that is to say, appearances,
are given to me. Moreover, in outcome it is a matter of 
indifference whether I say that in the empirical advance in
space I can meet with stars a hundred times farther removed
than the outermost now perceptible to me, or whether I say
that they are perhaps to be met with in cosmical space even
P 443
though no human being has ever perceived or ever will perceive
them. For even supposing they were given as things in
themselves, without relation to possible experience, it still
remains true that they are nothing to me, and therefore are
not objects, save in so far as they are contained in the series of
the empirical regress. Only in another sort of relation, when
these appearances would be used for the cosmological idea of
an absolute whole, and when, therefore, we are dealing with a
question which oversteps the limits of possible experience,
does distinction of the mode in which we view the reality of
those objects of the senses become of importance, as serving
to guard us against a deceptive error which is bound to arise
if we misinterpret our empirical concepts.
THE ANTINOMY OF PURE REASON
Section 7
CRITICAL SOLUTION OF THE COSMOLOGICAL CONFLICT
OF REASON WITH ITSELF
The whole antinomy of pure reason rests upon the dialectical
argument: If the conditioned is given, the entire series
of all its conditions is likewise given; objects of the senses are
given as conditioned; therefore, etc. Through this syllogism,
the major premiss of which appears so natural and evident, as
many cosmological ideas are introduced as there are differences
in the conditions (in the synthesis of appearances) that
constitute a series. The ideas postulate absolute totality of
these series; and thereby they set reason in unavoidable
conflict with itself. We shall be in a better position to detect
what is deceptive in this pseudo-rational argument, if we first
correct and define some of the concepts employed in it.
In the first place, it is evident beyond all possibility of
doubt, that if the conditioned is given, a regress in the series of
all its conditions is set us as a task. For it is involved in the
very concept of the conditioned that something is referred to a
condition, and if this condition is again itself conditioned, to a
more remote condition, and so through all the members of the
P 444
series. The above proposition is thus analytic, and has nothing
to fear from a transcendental criticism. It is a logical postulate
of reason, that through the understanding we follow up and
extend as far as possible that connection of a concept with its
conditions which directly results from the concept itself.
Further, if the conditioned as well as its condition are
things in themselves, then upon the former being given, the
regress to the latter is not only set as a task, but therewith
already really given. And since this holds of all members of
the series, the complete series of the conditions, and therefore
the unconditioned, is given therewith, or rather is presupposed
in view of the fact that the conditioned, which is only possible
through the complete series, is given. The synthesis of the
conditioned with its condition is here a synthesis of the mere
understanding, which represents things as they are, without
considering whether and how we can obtain knowledge of
them. If, however, what we are dealing with are appearances
-- as mere representations appearances cannot be given save
in so far as I attain knowledge of them, or rather attain them
in themselves, for they are nothing but empirical modes of
knowledge -- I cannot say, in the same sense of the terms, that
if the conditioned is given, all its conditions (as appearances)
are likewise given, and therefore cannot in any way infer the
absolute totality of the series of its conditions. The 
appearances are in their apprehension themselves nothing but an
empirical synthesis in space and time, and are given only in
this synthesis. It does not, therefore, follow, that if the 
conditioned, in the [field of] appearance, is given, the synthesis
which constitutes its empirical condition is given therewith
and is presupposed. This synthesis first occurs in the regress,
and never exists without it. What we can say is that a regress
to the conditions, that is, a continued empirical synthesis, on
the side of the conditions, is enjoined or set as a task, and that
in this regress there can be no lack of given conditions.
These considerations make it clear that the major premiss
of the cosmological inference takes the conditioned in the
transcendental sense of a pure category, while the minor premiss
takes it in the empirical sense of a concept of the 
understanding applied to mere appearances. The argument thus
commits that dialectical fallacy which is entitled sophisma
P 445
figurae dictionis. This fallacy is not, however, an artificial
one; a quite natural illusion of our common reason leads
us, when anything is given as conditioned, thus to assume in
the major premiss, as it were without thought or question, its
conditions and their series. This assumption is indeed simply
the logical requirement that we should have adequate premisses
for any given conclusion. Also, there is no reference to a
time-order in the connection of the conditioned with its condition;
they are presupposed as given together with it. Further,
it is no less natural, in the minor premiss, to regard appearances
both as things in themselves and as objects given to the
pure understanding, than to proceed as we have done in the
major, in which we have [similarly] abstracted from all those
conditions of intuition under which alone objects can be given.
Yet in so doing we have overlooked an important distinction
between the concepts. The synthesis of the conditioned with
its conditions (and the whole series of the latter) does not in
the major premiss carry with it any limitation through time
or any concept of succession. The empirical synthesis, on the
other hand, that is, the series of the conditions in appearance,
as subsumed in the minor premiss, is necessarily successive,
the members of the series being given only as following upon
one another in time; and I have therefore, in this case, no right
to assume the absolute totality of the synthesis and of the
series thereby represented. In the major premiss all the members
of the series are given in themselves, without any condition
of time, but in this minor premiss they are possible only
through the successive regress, which is given only in the
process in which it is actually carried out.
When this error has thus been shown to be involved in the
argument upon which both parties alike base their 
cosmo*********** 
unable to offer any sufficient title in support of their claims.
But the quarrel is not thereby ended -- as if one or both of the
parties had been proved to be wrong in the actual doctrines
they assert, that is, in the conclusions of their arguments. For
although they have failed to support their contentions by valid
grounds of proof, nothing seems to be clearer than that since
one of them asserts that the world has a beginning and the
other that it has no beginning and is from eternity, one of the
P 446
two must be in the right. But even if this be so, none the less,
since the arguments on both sides are equally clear, it is 
impossible to decide between them. The parties may be commanded
to keep the peace before the tribunal of reason; but the
controversy none the less continues. There can therefore be no
way of settling it once for all and to the satisfaction of both
sides, save by their becoming convinced that the very fact of
their being able so admirably to refute one another is evidence
that they are really quarrelling about nothing, and that a
certain transcendental illusion has mocked them with a reality
where none is to be found. This is the path which we shall now
proceed to follow in the settlement of a dispute that defies all
attempts to come to a decision.
* * *
Zeno of Elea, a subtle dialectician, was severely reprimanded
by Plato as a mischievous Sophist who, to show his
skill, would set out to prove a proposition through convincing
arguments and then immediately overthrow them by other
arguments equally strong. Zeno maintained, for example, that
God (probably conceived by him as simply the world) is
neither finite nor infinite, neither in motion nor at rest, neither
similar nor dissimilar to any other thing. To the critics of his
procedure he appeared to have the absurd intention of denying
both of two mutually contradictory propositions. But this 
accusation does not seem to me to be justified. The first of his
propositions I shall consider presently more in detail. As regards
the others, if by the word 'God' he meant the universe, he
would certainly have to say that it is neither abidingly present
in its place, that is, at rest, nor that it changes its place, that is,
is in motion; because all places are in the universe, and the
universe is not, therefore, itself in any place. Again, if the
universe comprehends in itself everything that exists, it cannot
be either similar or dissimilar to any other thing, because
there is no other thing, nothing outside it, with which it could
be compared. If two opposed judgments presuppose an inadmissible
condition, then in spite of their opposition, which does
not amount to a contradiction strictly so-called, both fall to the
ground, inasmuch as the condition, under which alone either
of them can be maintained, itself falls.
P 447
If it be said that all bodies have either a good smell or a
smell that is not good, a third case is possible, namely, that
a body has no smell at all; and both the conflicting propositions
may therefore be false. If, however, I say: all bodies are
either good-smelling or not good-smelling (vel suaveolens vel
non suaveolens), the two judgments are directly contradictory
to one another, and the former only is false, its contradictory
opposite, namely, that some bodies are not good-smelling,
comprehending those bodies also which have no smell at all.
Since, in the previous opposition (per disparata), smell, the
contingent condition of the concept of the body, was not
removed by the opposed judgment, but remained attached
to it, the two judgments were not related as contradictory
opposites.
If, therefore, we say that the world is either infinite in
extension or is not infinite (non est infinitus), and if the former
proposition is false, its contradictory opposite, that the world
is not infinite, must be true. And I should thus deny the existence
of an infinite world, without affirming in its place a finite
world. But if we had said that the world is either infinite or finite
(non-infinite), both statements might be false. For in that case
we should be regarding the world in itself as determined in its
magnitude, and in the opposed judgment we do not merely
remove the infinitude, and with it perhaps the entire separate
existence of the world, but attach a determination to the world,
regarded as a thing actually existing in itself. This assertion
may, however, likewise be false; the world may not be given
as a thing in itself, nor as being in its magnitude either infinite
or finite. I beg permission to entitle this kind of opposition
dialectical, and that of contradictories analytical. Thus of
two dialectically opposed judgments both may be false; for
the one is not a mere contradictory of the other, but says
something more than is required for a simple contradiction.
If we regard the two propositions, that the world is infinite
in magnitude and that it is finite in magnitude, as contradictory
opposites, we are assuming that the world, the complete
series of appearances, is a thing in itself that remains
even if I suspend the infinite or the finite regress in the series
of its appearances. If, however, I reject this assumption, or
P 448
rather this accompanying transcendental illusion, and deny
that the world is a thing in itself, the contradictory opposition
of the two assertions is converted into a merely dialectical
opposition. Since the world does not exist in itself, independently
of the regressive series of my representations, it exists
in itself neither as an infinite whole nor as a finite whole. It
exists only in the empirical regress of the series of appearances,
and is not to be met with as something in itself. If, then,
this series is always conditioned, and therefore can never be
given as complete, the world is not an unconditioned whole,
and does not exist as such a whole, either of infinite or of
finite magnitude.
What we have here said of the first cosmological idea,
that is, of the absolute totality of magnitude in the [field
of] appearance, applies also to all the others. The series of
conditions is only to be met with in the regressive synthesis
itself, not in the [field of] appearance viewed as a thing given
in and by itself, prior to all regress. We must therefore say that
the number of parts in a given appearance is in itself neither
finite nor infinite. For an appearance is not something existing
in itself, and its parts are first given in and through the regress
of the decomposing synthesis, a regress which is never given
in absolute completeness, either as finite or as infinite. This
also holds of the series of subordinated causes, and of the
series that proceeds from the conditioned to unconditioned
necessary existence. These series can never be regarded as
being in themselves in their totality either finite or infinite.
Being series of subordinated representations, they exist only
in the dynamical regress, and prior to this regress can have no
existence in themselves as self-subsistent series of things.
Thus the antinomy of pure reason in its cosmological ideas
vanishes when it is shown that it is merely dialectical, and
that it is a conflict due to an illusion which arises from our
applying to appearances that exist only in our representations,
and therefore, so far as they form a series, not otherwise than
in a successive regress, that idea of absolute totality which
holds only as a condition of things in themselves. From this
antinomy we can, however, obtain, not indeed a dogmatic, but
a critical and doctrinal advantage. It affords indirect proof of
P 449
the transcendental ideality of appearances -- a proof which
ought to convince any who may not be satisfied by the direct
proof given in the Transcendental Aesthetic. This proof would
consist in the following dilemma. If the world is a whole existing
in itself, it is either finite or infinite. But both alternatives
are false (as shown in the proofs of the antithesis and thesis
respectively). It is therefore also false that the world (the
sum of all appearances) is a whole existing in itself. From this
it then follows that appearances in general are nothing outside
our representations -- which is just what is meant by their
transcendental ideality.
This remark is of some importance. It enables us to see
that the proofs given in the fourfold antinomy are not merely
baseless deceptions. On the supposition that appearances, and
the sensible world which comprehends them all, are things
in themselves, these proofs are indeed well-grounded. The
conflict which results from the propositions thus obtained
shows, however, that there is a fallacy in this assumption, and
so leads us to the discovery of the true constitution of things,
as objects of the senses. While the transcendental dialectic does
not by any means favour scepticism, it certainly does favour
the sceptical method, which can point to such dialectic as an
example of its great services. For when the arguments of
reason are allowed to oppose one another in unrestricted
freedom, something advantageous, and likely to aid in the
correction of our judgments, will always accrue, though it
may not be what we set out to find.
THE ANTINOMY OF PURE REASON
Section 8
THE REGULATIVE PRINCIPLE OF PURE REASON IN ITS
APPLICATION TO THE COSMOLOGICAL IDEAS
Since no maximum of the series of conditions in a sensible
world, regarded as a thing in itself, is given through the 
cosmological principle of totality, but can only be set as a task
that calls for regress in the series of conditions, the principle
of pure reason has to be amended in these terms; and it
P 450
then preserves its validity, not indeed as the axiom that we
think the totality as actually in the object, but as a problem for
the understanding, and therefore for the subject, leading it to
undertake and to carry on, in accordance with the completeness
prescribed by the idea, the regress in the series of conditions of
any given conditioned. For in our sensibility, that is, in space
and time, every condition to which we can attain in the
exposition of given appearances is again conditioned. For
they are not objects in themselves -- were they such, the 
absolutely unconditioned might be found in them -- but simply
empirical representations which must always find in intuition
the condition that determines them in space and time.
The principle of reason is thus properly only a rule, 
prescribing a regress in the series of the conditions of given
appearances, and forbidding it to bring the regress to a close
by treating anything at which it may arrive as absolutely 
unconditioned. It is not a principle of the possibility of experience
and of empirical knowledge of objects of the senses, and therefore
not a principle of the understanding; for every experience,
in conformity with the given [forms of] intuition, is enclosed
within limits. Nor is it a constitutive principle of reason, 
enabling us to extend our concept of the sensible world beyond all
possible experience. It is rather a principle of the greatest 
possible continuation and extension of experience, allowing no 
empirical limit to hold as absolute. Thus it is a principle of reason
which serves as a rule, postulating what we ought to do in the
regress, but not anticipating what is present in the object as
it is in itself, prior to all regress. Accordingly I entitle it a
regulative principle of reason, to distinguish it from the principle
of the absolute totality of the series of conditions, viewed
as actually present in the object (that is, in the appearances),
which would be a constitutive cosmological principle. I have
tried to show by this distinction that there is no such 
constitutive principle, and so to prevent what otherwise, through
a transcendental subreption, inevitably takes place, namely,
the ascribing of objective reality to an idea that serves merely
as a rule.
In order properly to determine the meaning of this rule of
P 451
pure reason, we must observe, first, that it cannot tell us what
the object is, but only how the empirical regress is to be carried
out so as to arrive at the complete concept of the object. If it
attempted the former task, it would be a constitutive principle,
such as pure reason can never supply. It cannot be regarded
as maintaining that the series of conditions for a given conditioned
is in itself either finite or infinite. That would be to
treat a mere idea of absolute totality, which is only produced
in the idea, as equivalent to thinking an object that cannot be
given in any experience. For in terms of it we should be 
ascribing to a series of appearances an objective reality which
is independent of empirical synthesis. This idea of reason can
therefore do no more than prescribe a rule to the regressive
synthesis in the series of conditions; and in accordance with
this rule the synthesis must proceed from the conditioned,
through all subordinate conditions, up to the unconditioned.
Yet it can never reach this goal, for the absolutely 
unconditioned is not to be met with in experience.
We must therefore first of all determine what we are to
mean by the synthesis of a series, in cases in which the 
synthesis is never complete. In this connection two expressions
are commonly employed, which are intended to mark a distinction,
though without correctly assigning the ground of the
distinction. Mathematicians speak solely of a progressus in
infinitum. Philosophers, whose task it is to examine concepts,
refuse to accept this expression as legitimate, substituting for
it the phrase progressus in indefinitum. We need not stop to
examine the reasons for such a distinction, or to enlarge upon
its useful or useless employment. We need only determine
these concepts with such accuracy as is required for our 
particular purposes.
Of a straight line we may rightly say that it can be produced
to infinity. In this case the distinction between an infinite
and an indeterminately great advance (progressus in indefinitum)
would be mere subtlety. When we say, ' Draw a line',
it sounds indeed more correct to add in indefinitum than in
infinitum. Whereas the latter means that you must not cease
producing it -- which is not what is intended -- the former means
only, produce it as far as you please; and if we are referring
only to what it is in our power to do, this expression is quite
P 452
correct, for we can always make the line longer, without end.
So is it in all cases in which we speak only of the progress, that
is, of the advance from the condition to the conditioned: this
possible advance proceeds, without end, in the series of 
appearances. From a given pair of parents the descending line
of generation may proceed without end, and we can quite
well regard the line as actually so continuing in the world.
For in this case reason never requires an absolute totality
of the series, since it does not presuppose that totality as a
condition and as given (datum), but only as something 
conditioned, that allows of being given (dabile), and is added to
without end.
Quite otherwise is it with the problem: how far the regress
extends, when it ascends in a series from something given as
conditioned to its conditions. Can we say that the regress is in
infinitum, or only that it is indeterminately far extended (in
indefinitum)?  Can we, for instance, ascend from the men now
living, through the series of their ancestors, in infinitum; or
can we only say that, so far as we have gone back, we have
never met with an empirical ground for regarding the series as
limited at any point, and that we are therefore justified and at
the same time obliged, in the case of every ancestor, to search
further for progenitors, though not indeed to presuppose them?
We answer: when the whole is given in empirical intuition,
the regress in the series of its inner conditions proceeds
in infinitum; but when a member only of the series is
given, starting from which the regress has to proceed to absolute
totality, the regress is only of indeterminate character (in
indefinitum). Accordingly, the division of a body, that is, of a
portion of matter given between certain limits, must be said to
proceed in infinitum. For this matter is given as a whole, and
therefore with all its possible parts, in empirical intuition.
Since the condition of this whole is its part, and the condition
of this part is the part of the part, and so on, and since in
this regress of decomposition an unconditioned (indivisible)
member of this series of conditions is never met with, not only
is there never any empirical ground for stopping in the 
division, but the further members of any continued division are
themselves empirically given prior to the continuation of the
division. The division, that is to say, goes on in infinitum. On
P 453
the other hand, since the series of ancestors of any given man
is not given in its absolute totality in any possible experience,
the regress proceeds from every member in the series of 
generations to a higher member, and no empirical limit is 
encountered which exhibits a member as absolutely unconditioned.
And since the members, which might supply the condition, are
not contained in an empirical intuition of the whole, prior to
the regress, this regress does not proceed in infinitum, by division
of the given, but only indefinitely far, searching for further
members additional to those that are given, and which are
themselves again always given as conditioned.
In neither case, whether the regress be in infinitum or in
indefinitum, may the series of conditions be regarded as being
given as infinite in the object. The series are not things in
themselves, but only appearances, which, as conditions of one
another, are given only in the regress itself. The question,
therefore, is no longer how great this series of conditions may
be in itself, whether it be finite or infinite, for it is nothing in
itself; but how we are to carry out the empirical regress, and
how far we should continue it. Here we find an important 
distinction in regard to the rule governing such procedure. When
the whole is empirically given; it is possible to proceed back in
the series of its inner conditions in infinitum. When the whole
is not given, but has first to be given through empirical regress,
we can only say that the search for still higher conditions of the
series is possible in infinitum. In the former case we could say:
there are always more members, empirically given, than I can
reach through the regress of decomposition; in the latter case,
however, the position is this: we can always proceed still further
in the regress, because no member is empirically given as 
absolutely unconditioned; and since a higher member is therefore
always possible, the enquiry regarding it is necessary. In the
one case we necessarily find further members of the series; in
the other case, since no experience is absolutely limited, the
necessity is that we enquire for them. For either we have no
perception which sets an absolute limit to the empirical 
regress, in which case we must not regard the regress as 
completed, or we have a perception limiting our series, in which
case the perception cannot be part of the series traversed
(for that which limits must be distinct from that which is
P 454
thereby limited), and we must therefore continue our regress
to this condition also, and the regress is thus again resumed.
These observations will be set in their proper light by
their application in the following section.
THE ANTINOMY OF PURE REASON
Section 9
THE EMPIRICAL EMPLOYMENT OF THE REGULATIVE
PRINCIPLE OF REASON, IN RESPECT OF ALL 
COSMOLOGICAL IDEAS
We have already, on several occasions, shown that no 
transcendental employment can be made of the pure concepts either
of the understanding or of reason; that the [assertion of] absolute
totality of the series of conditions in the sensible world
rests on a transcendental employment of reason in which reason
demands this unconditioned completeness from what it assumes
to be a thing in itself; and that since the sensible world contains
no such completeness, we are never justified in enquiring, as
regards the absolute magnitude of the series in the sensible
world, whether it be limited or in itself unlimited, but only
how far we ought to go in the empirical regress, when we trace
experience back to its conditions, obeying the rule of reason,
and therefore resting content with no answer to its questions
save that which is in conformity with the object.
What therefore alone remains to us is the validity of the
principle of reason as a rule for the continuation and magnitude
of a possible experience; its invalidity as a constitutive 
principle of appearances [viewed as things] in themselves has been
sufficiently demonstrated. If we can keep these conclusions
steadily in view, the self-conflict of reason will be entirely at an
end. For not only will this critical solution destroy the illusion
which set reason at variance with itself, but will replace it by
teaching which, in correcting the misinterpretation that has
been the sole source of the conflict, brings reason into agreement
with itself. A principle which otherwise would be dialectical
will thus be converted into a doctrinal principle. In fact,
if this principle can be upheld as determining, in accordance
P 455
with its subjective significance, and yet also in conformity with
the objects of experience, the greatest possible empirical use of
understanding, the outcome will be much the same as if it
were -- what is impossible from pure reason -- an axiom which
determined a priori the objects in themselves. For only in 
proportion as the principle is effective in directing the widest
possible empirical employment of the understanding, can it
exercise, in respect of the objects of experience, any influence
in extending and correcting our knowledge.
1.
Solution of the Cosmological Idea of the Totality of the
Composition of the Appearances of a Cosmic Whole
Here, as in the other cosmological questions, the regulative
principle of reason is grounded on the proposition that in
the empirical regress we can have no experience of an absolute
limit, that is, no experience of any condition as being one
that empirically is absolutely unconditioned. The reason is
this: such an experience would have to contain a limitation
of appearances by nothing, or by the void, and in the 
continued regress we should have to be able to encounter this
limitation in a perception -- which is impossible.
This proposition, which virtually states that the only 
conditions which we can reach in the empirical regress are 
conditions which must themselves again be regarded as empirically
conditioned, contains the rule in terminis, that however
far we may have advanced in the ascending series, we must
always enquire for a still higher member of the series, which
may or may not become known to us through experience.
For the solution, therefore, of the first cosmological problem
we have only to decide whether in the regress to the 
unconditioned magnitude of the universe, in time and space, this
never limited ascent can be called a regress to infinity, or only
an indeterminately continued regress (in indefinitum).
The quite general representation of the series of all past
states of the world, as well as of all the things which coexist
in cosmic space, is itself merely a possible empirical regress
which I think to myself, though in an indeterminate manner.
Only in this way can the concept of such a series of conditions
P 456
for a given perception arise at all. Now we have the cosmic
whole only in concept, never, as a whole, in intuition. We
cannot, therefore, argue from the magnitude of the cosmic
whole to the magnitude of the regress, determining the
latter in accordance with the former; on the contrary, only
by reference to the magnitude of the empirical regress am I
in a position to make for myself a concept of the magnitude of
the world. But of this empirical regress the most that we can
ever know is that from every given member of the series of
conditions we have always still to advance empirically to a
higher and more remote member. The magnitude of the
whole of appearances is not thereby determined in any absolute
manner; and we cannot therefore say that this regress
proceeds to infinity. In doing so we should be anticipating
members which the regress has not yet reached, representing
their number as so great that no empirical synthesis could
attain thereto, and so should be determining the magnitude of
the world (although only negatively) prior to the regress --
which is impossible. Since the world is not given me, in its
totality, through any intuition, neither is its magnitude given
me prior to the regress. We cannot, therefore, say anything at
all in regard to the magnitude of the world, not even that there
is in it a regress in infinitum. All that we can do is to seek
for the concept of its magnitude according to the rule which
determines the empirical regress in it. This rule says no more
than that, however far we may have attained in the series of
empirical conditions, we should never assume an absolute
limit, but should subordinate every appearance, as 
conditioned, to another as its condition, and that we must
advance to this condition. This is the regressus in indefinitum,
which, as it determines no magnitude in the object,
is clearly enough distinguishable from the regressus in 
infinitum. 
++ This cosmic series can, therefore, be neither greater nor smaller
than the possible empirical regress upon which alone its concept
rests. And since this regress can yield neither a determinate infinite
nor a determinate finite (that is, anything absolutely limited), it is
evident that the magnitude of the world can be taken neither as
finite nor as infinite. The regress, through which it is represented,
allows of neither alternative.
P 457
I cannot say, therefore, that the world is infinite in space
or as regards past time. Any such concept of magnitude, as
being that of a given infinitude, is empirically impossible, and
therefore, in reference to the world as an object of the senses,
also absolutely impossible. Nor can I say that the regress from
a given perception to all that limits it in a series, whether in
space or in past time, proceeds to infinity; that would be to
presuppose that the world has infinite magnitude. I also cannot
say that the regress is finite; an absolute limit is likewise
empirically impossible. Thus I can say nothing regarding the
whole object of experience, the world of sense; I must limit
my assertions to the rule which determines how experience,
in conformity with its object, is to be obtained and further
extended.
Thus the first and negative answer to the cosmological
problem regarding the magnitude of the world is that the
world has no first beginning in time and no outermost limit
in space.
For if we suppose the opposite, the world would be limited
on the one hand by empty time and on the other by empty
space. Since, however, as appearance, it cannot in itself be
limited in either manner -- appearance not being a thing in
itself -- these limits of the world would have to be given in a
possible experience, that is to say, we should require to have
a perception of limitation by absolutely empty time or space.
But such an experience, as completely empty of content, is
impossible. Consequently, an absolute limit of the world is
impossible empirically, and therefore also absolutely.
The affirmative answer likewise directly follows, namely,
that the regress in the series of appearances, as a determination
of the magnitude of the world, proceeds in indefinitum.
++ It may be noted that this proof is presented in a very different
manner from the dogmatic proof of the antithesis of the first
antinomy. In that argument we regarded the sensible world, in
accordance with the common and dogmatic view, as a thing given
in itself, in its totality, prior to any regress; and we asserted that
unless it occupies all time and all places, it cannot have any 
determinate position whatsoever in them. The conclusion also was 
therefore different from that given above; for in the dogmatic proof we
inferred the actual infinity of the world.
P 458
This is equivalent to saying that, although the sensible world
has no absolute magnitude, the empirical regress (through
which alone it can be given on the side of its conditions) has
its own rule, namely, that it must always advance from every
member of the series, as conditioned, to one still more remote;
doing so by means either of our own experience, or of the
guiding-thread of history, or of the chain of effects and causes.
And as the rule further demands, our sole and constant aim
must be the extension of the possible empirical employment
of the understanding, this being the only proper task of reason
in the application of its principles.
This rule does not prescribe a determinate empirical regress
that must proceed without end in some one kind of appearance,
e.g. that in proceeding from a living person through a series
of progenitors we must never expect to meet with a first pair,
or that in the series of cosmic bodies we must never admit an
outermost sun. All that the rule requires is that the advance
from appearances be to appearances; for even if these latter
yield no actual perception (as is the case when for our 
consciousness they are too weak in degree to become experience),
as appearances they none the less still belong to a possible
experience.
All beginning is in time and all limits of the extended are
in space. But space and time belong only to the world of sense.
Accordingly, while appearances in the world are conditionally
limited, the world itself is neither conditionally nor 
unconditionally limited.
Similarly, since the world can never be given as complete,
and since even the series of conditions for that which is given
as conditioned cannot, as a cosmic series, be given as complete,
the concept of the magnitude of the world is given only through
the regress and not in a collective intuition prior to it. But the
regress consists only in the determining of the magnitude, and
does not give any determinate concept. It does not, therefore,
yield any concept of a magnitude which, in relation to a certain
[unit-] measure, can be described as infinite. In other words,
the regress does not proceed to the infinite, as if the infinite
could be given, but only indeterminately far, in order [by
means of the regress] to give that empirical magnitude which
first becomes actual in and through this very regress.
P 459
II
Solution of the Cosmological Idea of the Totality of
Division of a Whole given in Intuition
If we divide a whole which is given in intuition, we proceed
from something conditioned to the conditions of its possibility.
The division of the parts (subdivisio or decompositio)
is a regress in the series of these conditions. The absolute
totality of this series would be given only if the regress could
reach simple parts. But if all the parts in a continuously 
progressing decomposition are themselves again divisible, the
division, that is, the regress from the conditioned to its 
conditions, proceeds in infinitum. For the conditions (the parts)
are themselves contained in the conditioned, and since this
is given complete in an intuition that is enclosed between
limits the parts are one and all given together with the 
conditioned. The regress may not, therefore, be entitled merely
a regress in indefinitum. This was permissible in regard to the
first cosmological idea, since it required an advance from the
conditioned to its conditions, which, as outside it, were not given
through and along with it, but were first added to it in the 
empirical regress. We are not, however, entitled to say of a whole
which is divisible to infinity, that it is made up of infinitely
many parts. For although all parts are contained in the intuition
of the whole, the whole division is not so contained, but consists
only in the continuous decomposition, that is, in the regress
itself, whereby the series first becomes actual. Since this regress
is infinite, all the members or parts at which it arrives are
contained in the given whole, viewed as an aggregate. But the
whole series of the division is not so contained, for it is a
successive infinite and never whole, and cannot, therefore,
exhibit an infinite multiplicity, or any combination of an
infinite multiplicity in a whole.
This general statement is obviously applicable to space.
Every space intuited as within limits is such a whole, the parts
of which, as obtained by decomposition, are always themselves
spaces. Every limited space is therefore infinitely divisible.
From this a second application of the statement quite
naturally follows, namely, to an outer appearance enclosed
P 460
within limits, that is, to body. Its divisibility is grounded in
the divisibility of space, which constitutes the possibility of the
body as an extended whole. Body is therefore infinitely 
divisible, without consisting, however, of infinitely many parts.
It may seem, indeed, that a body, since it has to be represented
in space as substance, will, as regards the law of the
divisibility of space, differ from space. We may certainly grant
that decomposition can never remove all compositeness from
space; for that would mean that space, in which there is
nothing self-subsistent, had ceased to be space, which is impossible.
On the other hand, the assertion that if all compositeness
of matter be thought away nothing at all will remain, does not
appear to be compatible with the concept of a substance which
is meant to be the subject of all compositeness, and which
must persist in the elements of the composite, even although
the connection in space, whereby they constitute a body, be
removed. But while this is true of a thing in itself, as thought
through a pure concept of the understanding, it does not hold
of that which we entitle substance in the [field of] appearance.
For this latter is not an absolute subject, but only an abiding
image of sensibility; it is nothing at all save as an intuition,
in which unconditionedness is never to be met with.
But although this rule of progress in infinitum undoubtedly
applies to the subdivision of an appearance, viewed as a mere
filling of space, it cannot be made to apply to a whole in which
already, as given, the parts are so definitely distinguished off
from one another that they constitute a quantum discretum.
We cannot assume that every part of an organised whole is
itself again so organised that, in the analysis of the parts to
infinity, still other organised parts are always to be met with;
in a word, that the whole is organised to infinity. This is not a
thinkable hypothesis. It is true, indeed, that the parts of matter,
[as found] in their decomposition in infinitum, may be organised.
The infinitude of the division of a given appearance in
space is grounded solely on the fact that, through this infinitude,
only the divisibility (in itself, as regards the number of its
parts, absolutely indeterminate) is given -- the parts themselves
being given and determined only through the subdivision. In
a word, the whole is not in itself already divided. The number
P 461
of parts, therefore, which a division may determine in a whole,
will depend upon how far we care to advance in the regress of
the division. On the other hand, in the case of an organic body
conceived as organised in infinitum the whole is represented
as already divided into parts, and as yielding to us, prior to all
regress, a determinate and yet infinite number of parts. This,
however, is self-contradictory. This infinite involution is 
regarded as an infinite (that is, never to be completed) series,
and yet at the same time as completed in a [discrete] complex.
Infinite divisibility belongs to appearance only in so
far as it is a quantum continuum; it is inseparable from the
occupation of space, which is indeed its ground. To view 
anything as being a quantum discretum, is to take the number of
units in it as being determined, and therefore as being in every
case equal to some number. How far organisation can go in an
organised body, only experience can show; and although, so
far as our experience has gone, we may not have arrived with
certainty at any inorganic part, the possibility of experiencing
such parts must at least be recognised. When, however, we
have in mind the transcendental division of an appearance
in general, the question how far it may extend does not await
an answer from experience; it is decided by a principle of
reason which prescribes that, in the decomposition of the 
extended, the empirical regress, in conformity with the nature of
this appearance, be never regarded as absolutely completed.
Concluding Note on the Solution of the Mathematical - transcendental
Ideas, and Preliminary Observation on the Solution of
the Dynamical - transcendental Ideas.
In representing the antinomy of pure reason, through all
the transcendental ideas, in tabular form, and in showing that
the ground of this conflict and the only means of removing it
is by declaring both the opposed assertions to be false, we have
represented the conditions as, in all cases, standing to the 
conditioned in relations of space and time. This is the assumption
ordinarily made by the common understanding, and to it the
conflict is exclusively due. On this view all the dialectical
representations of totality, in the series of conditions for a
given conditioned, are throughout of the same character. The
P 462
condition is always a member of a series along with the 
conditioned, and so is homogeneous with it. In such a series
the regress was never thought as completed, or if it had to be
so thought, a member, in itself conditioned, must have been
falsely supposed to be a first member, and therefore to be
unconditioned; the object, that is, the conditioned, might not
always be considered merely according to its magnitude, but at
least the series of its conditions was so regarded. Thus arose the
difficulty -- a difficulty which could not be disposed of by any
compromise but solely by cutting the knot -- that reason made
the series either too long or too short for the understanding, so
that the understanding could never be equal to the prescribed
idea.
But in all this we have been overlooking an essential 
distinction that obtains among the objects, that is, among those
concepts of understanding which reason endeavours to raise
to ideas. According to the table of categories given above, two
of these concepts imply a mathematical, the other two a
dynamical synthesis of appearances. Hitherto it has not been
necessary to take account of this distinction; for just as in the
general representation of all transcendental ideas we have
been conforming to conditions within the [field of] appearance,
so in the two mathematical - transcendental ideas the only
object we have had in mind is object as appearance. But now
that we are proceeding to consider how far dynamical concepts
of the understanding are adequate to the idea of reason,
the distinction becomes of importance, and opens up to us an
entirely new view of the suit in which reason is implicated.
This suit, in our previous trial of it, has been dismissed as
resting, on both sides, on false presuppositions. But since in
the dynamical antinomy a presupposition compatible with the
pretensions of reason may perhaps be found, and since the
judge may perhaps make good what is lacking in the pleas
which both sides have been guilty of misstating, the suit may
be settled to the satisfaction of both parties, a procedure 
impossible in the case of the mathematical antinomies.
If we consider solely the extension of the series of conditions,
and whether the series are adequate to the idea, or the
idea too large or too small for the series, the series are indeed in
P 463
these respects all homogeneous. But the concept of the 
understanding, which underlies these ideas, may contain either a
synthesis solely of the homogeneous (which is presupposed
alike in the composition and in the division of every 
magnitude), or a synthesis of the heterogeneous. For the 
heterogeneous can be admitted as at least possible in the case of
dynamical synthesis, alike in causal connection and in the
connection of the necessary with the contingent.
Hence in the mathematical connection of the series of
appearances no other than a sensible condition is admissible,
that is to say, none that is not itself a part of the series. On the
other hand, in the dynamical series of sensible conditions, a
heterogeneous condition, not itself a part of the series, but
purely intelligible, and as such outside the series, can be
allowed. In this way reason obtains satisfaction and the
unconditioned is set prior to the appearances, while yet the
invariably conditioned character of the appearances is not
obscured, nor their series cut short, in violation of the
principles prescribed by the understanding.
Inasmuch as the dynamical ideas allow of a condition of
appearances outside the series of the appearances, that is, a
condition which is not itself appearance, we arrive at a 
conclusion altogether different from any that was possible in the
case of the mathematical antinomy. In it we were obliged
to denounce both the opposed dialectical assertions as false.
In the dynamical series, on the other hand, the completely
conditioned, which is inseparable from the series considered
as appearances, is bound up with a condition which, while
indeed empirically unconditioned, is also non-sensible. We
are thus able to obtain satisfaction for understanding on
the one hand and for reason on the other.
++ Understanding does not admit among appearances any condition
which can itself be empirically unconditioned. But if for some
conditioned in the [field of] appearance we can conceive an intelligible
condition, not belonging to the series of appearances as one of
its members, and can do so without in the least interrupting the
series of empirical conditions, such a condition may be accepted as
empirically unconditioned, without prejudice to the continuity of the
empirical regress.
P 464
The dialectical arguments, which in one or other way sought unconditioned
totality in mere appearances, fall to the ground, and the 
propositions of reason, when thus given this more correct 
interpretation, may both alike be true. This can never be the case
with those cosmological ideas which refer only to a mathematically
unconditioned unity; for in them no condition of the
series of appearances can be found that is not itself 
appearance, and as appearance one of the members of the series.
III
Solution of the Cosmological Idea of Totality in the
Derivation of Cosmical Events from their Causes
When we are dealing with what happens there are only two
kinds of causality conceivable by us; the causality is either
according to nature or arises from freedom. The former is
the connection in the sensible world of one state with a preceding
state on which it follows according to a rule. Since the
causality of appearances rests on conditions of time, and the
preceding state, if it had always existed, could not have produced
an effect which first comes into being in time, it follows
that the causality of the cause of that which happens or comes
into being must itself also have come into being, and that in
accordance with the principle of the understanding it must
in its turn itself require a cause.
 By freedom, on the other hand, in its cosmological meaning,
I understand the power of beginning a state spontaneously.
Such causality will not, therefore, itself stand under
another cause determining it in time, as required by the law of
nature. Freedom, in this sense, is a pure transcendental idea,
which, in the first place, contains nothing borrowed from 
experience, and which, secondly, refers to an object that cannot
be determined or given in any experience. That everything
which happens has a cause is a universal law, conditioning the
very possibility of all experience. Hence the causality of the
cause, which itself happens or comes to be, must itself in turn
have a cause; and thus the entire field of experience, however
far it may extend, is transformed into a sum-total of the
merely natural. But since in this way no absolute totality of
P 465
conditions determining causal relation can be obtained, reason
creates for itself the idea of a spontaneity which can begin to
act of itself, without requiring to be determined to action by
an antecedent cause in accordance with the law of causality.
It should especially be noted that the practical concept of
freedom is based on this transcendental idea, and that in the
latter lies the real source of the difficulty by which the 
question of the possibility of freedom has always been beset.
Freedom in the practical sense is the will's independence of
coercion through sensuous impulses. For a will is sensuous, in
so far as it is pathologically affected, i.e. by sensuous motives;
it is animal (arbitrium brutum), if it can be pathologically
necessitated. The human will is certainly an arbitrium sensitivum,
not, however, brutum but liberum. For sensibility does
not necessitate its action. There is in man a power of self-
determination, independently of any coercion through sensuous
impulses.
Obviously, if all causality in the sensible world were mere
nature, every event would be determined by another in time,
in accordance with necessary laws. Appearances, in determining
the will, would have in the actions of the will their natural
effects, and would render the actions necessary. The denial of
transcendental freedom must, therefore, involve the elimination
of all practical freedom. For practical freedom presupposes
that although something has not happened, it ought to
have happened, and that its cause, [as found] in the [field of]
appearance, is not therefore, so determining that it excludes a
causality of our will -- a causality which, independently of those
natural causes, and even contrary to their force and influence,
can produce something that is determined in the time-order
in accordance with empirical laws, and which can therefore
begin a series of events entirely of itself.
Here then, as always happens when reason, in venturing
beyond the limits of possible experience, comes into conflict
with itself the problem is not really physiological but 
transcendental. The question as to the possibility of freedom
does indeed concern psychology; since it rests on dialectical
arguments of pure reason, its treatment and solution belong
exclusively to transcendental philosophy. Before attempting
P 466
this solution, a task which transcendental philosophy cannot
decline, I must define somewhat more accurately the procedure
of transcendental philosophy in dealing with the problem.
If appearances were things in themselves, and space and
time forms of the existence of things in themselves, the 
conditions would always be members of the same series as the 
conditioned; and thus, in the present case, as in the other 
transcendental ideas, the antinomy would arise, that the series must be
too large or too small for the understanding. But the dynamical
concepts of reason, with which we have to deal in this and
the following section, possess this peculiarity that they are not
concerned with an object considered as a magnitude, but only
with its existence. Accordingly we can abstract from the magnitude
of the series of conditions, and consider only the dynamical
relation of the condition to the conditioned. The difficulty
which then meets us, in dealing with the question regarding
nature and freedom, is whether freedom is possible at all, and
if it be possible, whether it can exist along with the universality
of the natural law of causality. Is it a truly disjunctive 
proposition to say that every effect in the world must arise either
from nature or from freedom; or must we not rather say that
in one and the same event, in different relations, both can be
found? That all events in the sensible world stand in 
thoroughgoing connection in accordance with unchangeable laws of
nature is an established principle of the Transcendental Analytic,
and allows of no exception. The question, therefore, can
only be whether freedom is completely excluded by this inviolable
rule, or whether an effect, notwithstanding its being thus
determined in accordance with nature, may not at the same
time be grounded in freedom. The common but fallacious 
presupposition of the absolute reality of appearances here manifests
its injurious influence, to the confounding of reason. For
if appearances are things in themselves, freedom cannot be 
upheld. Nature will then be the complete and sufficient determining
cause of every event. The condition of the event will be
such as can be found only in the series of appearances; both it
and its effect will be necessary in accordance with the law of
nature. If, on the other hand, appearances are not taken for
more than they actually are; if they are viewed not as things in
themselves, but merely as representations, connected according
P 467
to empirical laws, they must themselves have grounds
which are not appearances. The effects of such an intelligible
cause appear, and accordingly can be determined through
other appearances, but its causality is not so determined.
While the effects are to be found in the series of empirical 
conditions, the intelligible cause, together with its causality, is
outside the series. Thus the effect may be regarded as free in
respect of its intelligible cause, and at the same time in respect
of appearances as resulting from them according to the necessity
of nature. This distinction, when stated in this quite general
and abstract manner, is bound to appear extremely subtle and
obscure, but will become clear in the course of its application.
My purpose has only been to point out that since the thoroughgoing
connection of all appearances, in a context of nature, is
an inexorable law, the inevitable consequence of obstinately
insisting upon the reality of appearances is to destroy all
freedom. Those who thus follow the common view have never
been able to reconcile nature and freedom.
Possibility of Causality through Freedom, in Harmony with the
Universal Law of Natural Necessity.
Whatever in an object of the senses is not itself appearance,
I entitle intelligible. If, therefore, that which in the sensible
world must be regarded as appearance has in itself a faculty
which is not an object of sensible intuition, but through which
it can be the cause of appearances, the causality of this being
can be regarded from two points of view. Regarded as the
causality of a thing in itself, it is intelligible in its action; 
regarded as the causality of an appearance in the world of sense,
it is sensible in its effects. We should therefore have to form both
an empirical and an intellectual concept of the causality of the
faculty of such a subject, and to regard both as referring to one
and the same effect. This twofold manner of conceiving the
faculty possessed by an object of the senses does not contradict
any of the concepts which we have to form of appearances and
of a possible experience. For since they are not things in 
themselves, they must rest upon a transcendental object which 
determines them as mere representations; and consequently there is
nothing to prevent us from ascribing to this transcendental
P 468
object, besides the quality in terms of which it appears, a
causality which is not appearance, although its effect is to be
met with in appearance. Every efficient cause must have a
character, that is, a law of its causality, without which it
would not be a cause. On the above supposition, we should,
therefore, in a subject belonging to the sensible world have,
first, an empirical character, whereby its actions, as 
appearances, stand in thoroughgoing connection with other 
appearances in accordance with unvarying laws of nature. And since
these actions can be derived from the other appearances, they
constitute together with them a single series in the order of
nature. Secondly, we should also have to allow the subject an
intelligible character, by which it is indeed the cause of those
same actions [in their quality] as appearances, but which does
not itself stand under any conditions of sensibility, and is not
itself appearance. We can entitle the former the character of
the thing in the [field of] appearance, and the latter its 
character as thing in itself.
Now this acting subject would not, in its intelligible
character, stand under any conditions of time; time is only a
condition of appearances, not of things in themselves. In this
subject no action would begin or cease, and it would not, therefore,
have to conform to the law of the determination of all that
is alterable in time, namely, that everything which happens
must have its cause in the appearances which precede it. In
a word, its causality, so far as it is intelligible, would not have
a place in the series of those empirical conditions through
which the event is rendered necessary in the world of sense.
This intelligible character can never, indeed, be immediately
known, for nothing can be perceived except in so far as it
appears. It would have to be thought in accordance with the
empirical character-- just as we are constrained to think a
transcendental object as underlying appearances, though we
know nothing of what it is in itself.
In its empirical character, therefore, this subject, as 
appearance, would have to conform to all the laws of causal
determination. To this extent it could be nothing more than
a part of the world of sense, and its effects, like all other
P 469
appearances, must be the inevitable outcome of nature. In
proportion as outer appearances are found to influence it, and
in proportion as its empirical character, that is, the law of its
causality, becomes known through experience, all its actions
must admit of explanation in accordance with the laws of
nature. In other words, all that is required for their complete
and necessary determination must be found in a possible
experience.
In its intelligible character (though we can only have a
general concept of that character) this same subject must be
considered to be free from all influence of sensibility and from
all determination through appearances. Inasmuch as it is
noumenon, nothing happens in it; there can be no change
requiring dynamical determination in time, and therefore no
causal dependence upon appearances. And consequently,
since natural necessity is to be met with only in the sensible
world, this active being must in its actions be independent
of, and free from all such necessity. No action begins in this
active being itself; but we may yet quite correctly say that the
active being of itself begins its effects in the sensible world. In
so doing, we should not be asserting that the effects in the
sensible world can begin of themselves; they are always 
predetermined through antecedent empirical conditions, though
solely through their empirical character (which is no more
than the appearance of the intelligible), and so are only possible
as a continuation of the series of natural causes. In this
way freedom and nature, in the full sense of these terms, can
exist together, without any conflict, in the same actions, according
as the actions are referred to their intelligible or to their
sensible cause.
Explanation of the Cosmological Idea of Freedom in its 
connection with Universal Natural Necessity.
I have thought it advisable to give this outline sketch of
the solution of our transcendental problem, so that we may
be the better enabled to survey the course which reason has
to adopt in arriving at the solution. I shall now proceed to set
forth the various factors involved in this solution, and to 
consider each in detail.
That everything which happens has a cause, is a law of
nature. Since the causality of this cause, that is, the action of
P 470
the cause, is antecedent in time to the effect which has ensued
upon it, it cannot itself have always existed, but must have
happened, and among the appearances must have a cause by
which it in turn is determined. Consequently, all events are
empirically determined in an order of nature. Only in virtue
of this law can appearances constitute a nature and become
objects of experience. This law is a law of the understanding,
from which no departure can be permitted, and from which
no appearance may be exempted. To allow such exemption
would be to set an appearance outside all possible experience,
to distinguish it from all objects of possible experience, and so
to make of it a mere thought-entity, a phantom of the brain.
This would seem to imply the existence of a chain of causes
which in the regress to their conditions allows of no absolute 
totality. But that need not trouble us. The point has already been
dealt with in the general discussion of the antinomy into which
reason falls when in the series of appearances it proceeds to the
unconditioned. Were we to yield to the illusion of transcendental
realism, neither nature nor freedom would remain. The only
question here is this: -- Admitting that in the whole series of
events there is nothing but natural necessity, is it yet possible
to regard one and the same event as being in one aspect merely
an effect of nature and in another aspect an effect due to 
freedom; or is there between these two kinds of causality a direct
contradiction?
Among the causes in the [field of] appearance there certainly
cannot be anything which could begin a series absolutely
and of itself. Every action, [viewed] as appearance, in so
far as it gives rise to an event, is itself an event or happening,
and presupposes another state wherein its cause is to be found.
Thus everything which happens is merely a continuation of
the series, and nothing that begins of itself is a possible 
member of the series. The actions of natural causes in the time-
sequence are thus themselves effects; they presuppose causes
antecedent to them in the temporal series. An original act,
such as can by itself bring about what did not exist before, is
not to be looked for in the causally connected appearances.
Now granting that effects are appearances and that their
cause is likewise appearance, is it necessary that the causality
of their cause should be exclusively empirical? May it not
P 471
rather be, that while for every effect in the [field of] 
appearance a connection with its cause in accordance with the
laws of empirical causality is indeed required, this empirical
causality, without the least violation of its connection with
natural causes, is itself an effect of a causality that is not
empirical but intelligible? This latter causality would be the
action of a cause which, in respect of appearances, is original,
and therefore, as pertaining to this faculty, not appearance but
intelligible; although it must otherwise, in so far as it is a link
in the chain of nature, be regarded as entirely belonging to
the world of sense.
The principle of the causal connection of appearances is
required in order that we may be able to look for and to
determine the natural conditions of natural events, that is to
say, their causes in the [field of] appearance. If this principle
be admitted, and be not weakened through any exception,
the requirements of the understanding, which in its empirical
employment sees in all happenings nothing but nature, and is
justified in so doing, are completely satisfied; and physical 
explanations may proceed on their own lines without interference.
These requirements are not in any way infringed, if we assume,
even though the assumption should be a mere fiction, that some
among the natural causes have a faculty which is intelligible
only, inasmuch as its determination to action never rests upon
empirical conditions, but solely on grounds of understanding.
We must, of course, at the same time be able to assume that
the action of these causes in the [field of] appearance is in 
conformity with all the laws of empirical causality. In this way
the acting subject, as causa phaenomenon, would be bound up
with nature through the indissoluble dependence of all its
actions, and only as we ascend from the empirical object to
the transcendental should we find that this subject, together
with all its causality in the [field of] appearance, has in its
noumenon certain conditions which must be regarded as
purely intelligible. For if in determining in what ways 
appearances can serve as causes we follow the rules of nature, we
need not concern ourselves what kind of ground for these
appearances and their connection may have to be thought as
existing in the transcendental subject, which is empirically
P 472
unknown to us. This intelligible ground does not have to be
considered in empirical enquiries; it concerns only thought
in the pure understanding; and although the effects of this
thought and action of the pure understanding are to be met
with in the appearances, these appearances must none the less
be capable of complete causal explanation in terms of other
appearances in accordance with natural laws. We have to take
their strictly empirical character as the supreme ground of
explanation, leaving entirely out of account their intelligible
character (that is, the transcendental cause of their empirical
character) as being completely unknown, save in so far as the
empirical serves for its sensible sign.
Let us apply this to experience. Man is one of the appearances
of the sensible world, and in so far one of the natural
causes the causality of which must stand under empirical
laws. Like all other things in nature, he must have an 
empirical character. This character we come to know through
the powers and faculties which he reveals in his actions. In
lifeless, or merely animal, nature we find no ground for
thinking that any faculty is conditioned otherwise than in a
merely sensible manner. Man, however, who knows all the
rest of nature solely through the senses, knows himself also
through pure apperception; and this, indeed, in acts and inner
determinations which he cannot regard as impressions of the
senses. He is thus to himself, on the one hand phenomenon,
and on the other hand, in respect of certain faculties the
action of which cannot be ascribed to the receptivity of
sensibility, a purely intelligible object. We entitle these
faculties understanding and reason. The latter, in particular,
we distinguish in a quite peculiar and especial way from all
empirically conditioned powers. For it views its objects 
exclusively in the light of ideas, and in accordance with them
determines the understanding, which then proceeds to make
an empirical use of its own similarly pure concepts.
That our reason has causality, or that we at least represent
it to ourselves as having causality, is evident from the 
imperatives which in all matters of conduct we impose as rules upon
our active powers. 'Ought' expresses a kind of necessity and of
connection with grounds which is found nowhere else in the
P 473
whole of nature. The understanding can know in nature only
what is, what has been, or what will be. We cannot say that
anything in nature ought to be other than what in all these
time-relations it actually is. When we have the course of
nature alone in view, 'ought' has no meaning whatsoever. It
is just as absurd to ask what ought to happen in the natural
world as to ask what properties a circle ought to have. All
that we are justified in asking is: what happens in nature?
what are the properties of the circle?
This 'ought' expresses a possible action the ground of
which cannot be anything but a mere concept; whereas in the
case of a merely natural action the ground must always be an
appearance. The action to which the 'ought' applies must 
indeed be possible under natural conditions. These conditions,
however, do not play any part in determining the will itself,
but only in determining the effect and its consequences in the
[field of] appearance. No matter how many natural grounds
or how many sensuous impulses may impel me to will, they
can never give rise to the 'ought', but only to a willing which,
while very far from being necessary, is always conditioned; and
the 'ought' pronounced by reason confronts such willing with a
limit and an end -- nay more, forbids or authorises it. Whether
what is willed be an object of mere sensibility (the pleasant) or
of pure reason (the good),reason will not give way to any ground
which is empirically given. Reason does not here follow the
order of things as they present themselves in appearance, but
frames to itself with perfect spontaneity an order of its own 
according to ideas, to which it adapts the empirical conditions,
and according to which it declares actions to be necessary,
even although they have never taken place, and perhaps never
will take place. And at the same time reason also presupposes
that it can have causality in regard to all these actions, since
otherwise no empirical effects could be expected from its ideas.
Now, in view of these considerations, let us take our
stand, and regard it as at least possible for reason to have
causality with respect to appearances. Reason though it be,
it must none the less exhibit an empirical character. For every
cause presupposes a rule according to which certain appearances
follow as effects; and every rule requires uniformity in
the effects. This uniformity is, indeed, that upon which the
P 474
concept of cause (as a faculty) is based, and so far as it must
be exhibited by mere appearances may be named the empirical
character of the cause. This character is permanent,
but its effects, according to variation in the concomitant and
in part limiting conditions, appear in changeable forms.
Thus the will of every man has an empirical character,
which is nothing but a certain causality of his reason, so far as
that causality exhibits, in its effects in the [field of] appearance,
a rule from which we may gather what, in their kind and degrees,
are the actions of reason and the grounds thereof, and so
may form an estimate concerning the subjective principles of
his will. Since this empirical character must itself be 
discovered from the appearances which are its effect and from
the rule to which experience shows them to conform, it
follows that all the actions of men in the [field of] appearance
are determined in conformity with the order of nature,
by their empirical character and by the other causes which 
comoderate with that character; and if we could exhaustively 
investigate all the appearances of men's wills, there would not
be found a single human action which we could not predict
with certainty, and recognise as proceeding necessarily from
its antecedent conditions. So far, then, as regards this empirical
character there is no freedom; and yet it is only in the
light of this character that man can be studied -- if, that is to
say, we are simply observing, and in the manner of anthropology
seeking to institute a physiological investigation into
the motive causes of his actions.
But when we consider these actions in their relation to
reason -- I do not mean speculative reason, by which we 
endeavour to explain their coming into being, but reason in so
far as it is itself the cause producing them -- if, that is to say,
we compare them with [the standards of] reason in its practical
bearing, we find a rule and order altogether different from the
order of nature. For it may be that all that has happened in the
course of nature, and in accordance with its empirical grounds
must inevitably have happened, ought not to have happened.
Sometimes, however, we find, or at least believe that we find,
that the ideas of reason have in actual fact proved their 
causality in respect of the actions of men, as appearances; and
that these actions have taken place, not because they were
P 475
determined by empirical causes, but because they were 
determined by grounds of reason.
Granted, then, that reason may be asserted to have causality
in respect of appearance, its action can still be said to
be free, even although its empirical character (as a mode of
sense) is completely and necessarily determined in all its
detail. This empirical character is itself determined in the 
intelligible character (as a mode of thought). The latter, 
however, we do not know; we can only indicate its nature by
means of appearances; and these really yield an immediate
knowledge only of the mode of sense, the empirical character.
The action, in so far as it can be ascribed to a mode
of thought as its cause, does not follow therefrom in accordance
with empirical laws; that is to say, it is not preceded
by the conditions of pure reason, but only by their effects in
the [field of] appearance of inner sense. Pure reason, as a
purely intelligible faculty, is not subject to the form of time,
nor consequently to the conditions of succession in time. The
causality of reason in its intelligible character does not, in 
producing an effect, arise or begin to be at a certain time. For in
that case it would itself be subject to the natural law of 
appearances, in accordance with which causal series are determined
in time; and its causality would then be nature, not freedom.
Thus all that we are justified in saying is that, if reason can
have causality in respect of appearances, it is a faculty through
which the sensible condition of an empirical series of effects
first begins. For the condition which lies in reason is not
sensible, and therefore does not itself begin to be. And thus
what we failed to find in any empirical series is disclosed as
being possible, namely, that the condition of a successive
series of events may itself be empirically unconditioned.
++ The real morality of actions, their merit or guilt, even that of
our own conduct, thus remains entirely hidden from us. Our 
imputations can refer only 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 (merito fortunae), can
never be determined; and upon it therefore no perfectly just 
judgments can be passed.
P 476
For here the condition is outside the series of appearances (in the
intelligible), and therefore is not subject to any sensible 
condition, and to no time-determination through an antecedent
cause.
The same cause does, indeed, in another relation, belong
to the series of appearances. Man is himself an appearance.
His will has an empirical character, which is the empirical
cause of all his actions. There is no condition determining
man in accordance with this character which is not contained
in the series of natural effects, or which is not subject to their
law -- the law according to which there can be no empirically
unconditioned causality of that which happens in time. Therefore
no given action (since it can be perceived only as appearance)
can begin absolutely of itself. But of pure reason we
cannot say that the state wherein the will is determined is
preceded and itself determined by some other state. For since
reason is not itself an appearance, and is not subject to any
conditions of sensibility, it follows that even as regards its
causality there is in it no time-sequence, and that the 
dynamical law of nature, which determines succession in time in
accordance with rules, is not applicable to it.
Reason is the abiding condition of all those actions of the
will under [the guise of] which man appears. Before ever they
have happened, they are one and all predetermined in the
empirical character. In respect of the intelligible character, of
which the empirical character is the sensible schema, there can
be no before and after; every action, irrespective of its relation
in time to other appearances, is the immediate effect of the
intelligible character of pure reason. Reason therefore acts
freely; it is not dynamically determined in the chain of natural
causes through either outer or inner grounds antecedent in
time. This freedom ought not, therefore, to be conceived only
negatively as independence of empirical conditions. The
faculty of reason, so regarded, would cease to be a cause of
5852)appearances. It must also be described in positive terms, as
the power of originating a series of events. In reason itself
nothing begins; as unconditioned condition of every voluntary
act, it admits of no conditions antecedent to itself in time. Its
effect has, indeed, a beginning in the series of appearances,
but never in this series an absolutely first beginning.
P 477
In order to illustrate this regulative principle of reason by
an example of its empirical employment -- not, however, to confirm
it, for it is useless to endeavour to prove transcendental
propositions by examples -- let us take a voluntary action, for
example, a malicious lie by which a certain confusion has been
caused in society. First of all, we endeavour to discover the
motives to which it has been due, and then, secondly, in the
light of these, we proceed to determine how far the action and
its consequences can be imputed to the offender. As regards the
first question, we trace the empirical character of the action to
its sources, finding these in defective education, bad company,
in part also in the viciousness of a natural disposition insensitive
to shame, in levity and thoughtlessness, not neglecting to take
into account also the occasional causes that may have intervened.
We proceed in this enquiry just as we should in ascertaining
for a given natural effect the series of its determining
causes. But although we believe that the action is thus determined,
we none the less blame the agent, not indeed on account
of his unhappy disposition, nor on account of the circumstances
that have influenced him, nor even on account of his
previous way of life; for we presuppose that we can leave out of
consideration what this way of life may have been, that we can
regard the past series of conditions as not having occurred and
the act as being completely unconditioned by any preceding
state, just as if the agent in and by himself began in this action
an entirely new series of consequences. Our blame is based on
a law of reason whereby we regard reason as a cause that
irrespective of all the above-mentioned empirical conditions
could have determined, and ought to have determined, the
agent to act otherwise. This causality of reason we do not 
regard as only a co-operating agency, but as complete in itself,
even when the sensuous impulses do not favour but are directly
opposed to it; the action is ascribed to the agent's intelligible
character; in the moment when he utters the lie, the guilt is
entirely his. Reason, irrespective of all empirical conditions of
the act, is completely free, and the lie is entirely due to its
default.
Such imputation clearly shows that we consider reason to
be unaffected by these sensible influences, and not liable to
alteration. Its appearances -- the modes in which it manifests
P 478
itself in its effects -- do alter; but in itself [so we consider] there
is no preceding state determining the state that follows. That
is to say, it does not belong to the series of sensible conditions
which render appearances necessary in accordance with laws
of nature. Reason is present in all the actions of men at all
times and under all circumstances, and is always the same;
but it is not itself in time, and does not fall into any new state
in which it was not before. In respect to new states, it is 
determining, not determinable. We may not, therefore, ask why
reason has not determined itself differently, but only why it
has not through its causality determined the appearances differently.
But to this question no answer is possible. For a different
intelligible character would have given a different empirical
character. When we say that in spite of his whole previous
course of life the agent could have refrained from lying, this
only means that the act is under the immediate power of reason,
and that reason in its causality is not subject to any conditions
of appearance or of time. Although difference of time makes a
fundamental difference to appearances in their relations to one
another -- for appearances are not things in themselves and
therefore not causes in themselves -- it can make no difference
to the relation in which the action stands to reason.
 Thus in our judgments in regard to the causality of free
actions, we can get as far as the intelligible cause, but not 
beyond it. We can know that it is free, that is, that it is 
determined independently of sensibility, and that in this way it may
be the sensibly unconditioned condition of appearances. But
to explain why in the given circumstances the intelligible 
character should give just these appearances and this empirical
character transcends all the powers of our reason, indeed all
its rights of questioning, just as if we were to ask why the 
transcendental object of our outer sensible intuition gives intuition
in space only and not some other mode of intuition. But the
problem which we have to solve does not require us to raise any
such questions. Our problem was this only: whether freedom
and natural necessity can exist without conflict in one and the
same action; and this we have sufficiently answered. We have
shown that since freedom may stand in relation to a quite
different kind of conditions from those of natural necessity,
the law of the latter does not affect the former, and that both
P 479
may exist, independently of one another and without 
interfering with each other.
* * *
The reader should be careful to observe that in what has
been said our intention has not been to establish the reality
of freedom as one of the faculties which contain the cause of
the appearances of our sensible world. For that enquiry, as it
does not deal with concepts alone, would not have been 
transcendental. And further, it could not have been successful,
since we can never infer from experience anything which cannot
be thought in accordance with the laws of experience. It
has not even been our intention to prove the possibility of
freedom. For in this also we should not have succeeded, since
we cannot from mere concepts a priori know the possibility
of any real ground and its causality. Freedom is here being
treated only as a transcendental idea whereby reason is led to
think that it can begin the series of conditions in the [field of]
appearance by means of the sensibly unconditioned, and so
becomes involved in an antinomy with those very laws which
it itself prescribes to the empirical employment of the 
understanding. What we have alone been able to show, and what we
have alone been concerned to show, is that this antinomy rests
on a sheer illusion, and that causality through freedom is at
least not incompatible with nature.
IV
Solution of the Cosmological Idea of the Totality of the 
Dependence of Appearances as regards their Existence in
general
In the preceding subsection we have considered the changes
of the sensible world in so far as they form a dynamical
series, each member being subordinate to another as effect to
cause. We shall now employ this series of states merely to
guide us in our search for an existence that may serve as
the supreme condition of all that is alterable, that is, in
our search for necessary being. We are concerned here, not
with unconditioned causality, but with the unconditioned
existence of substance itself. The series which we have in
P 480
view is, therefore, really a series of concepts, not a series
of intuitions in which one intuition is the condition of the
other.
But it is evident that since everything in the sum-total
of appearances is alterable, and therefore conditioned in its
existence, there cannot be in the whole series of dependent 
existence any unconditioned member the existence of which can
be regarded as absolutely necessary. Hence, if appearances
were things in themselves, and if, as would then follow, the
condition and the conditioned always belonged to one and the
same series of intuitions, by no possibility could a necessary
being exist as the condition of the existence of appearances in
the world of sense.
The dynamical regress is distinguished in an important 
respect from the mathematical. Since the mathematical regress
is concerned only with the combining of parts to form a whole,
or the division of a whole into parts, the conditions of this
series must always be regarded as parts of the series, and 
therefore as homogeneous and as appearances. In the dynamical
regress, on the other hand, we are concerned, not with the 
possibility of an unconditioned whole of given parts, or with an
unconditioned part for a given whole, but with the derivation
of a state from its cause, or of the contingent existence of 
substance itself from necessary existence. In this latter regress, it
is not, therefore, necessary that the condition should form part
of an empirical series along with the conditioned.
A way of escape from this apparent antinomy thus lies
open to us. Both of the conflicting propositions may be true,
if taken in different connections. All things in the world of
sense may be contingent, and so have only an empirically
conditioned existence, while yet there may be a non-empirical
condition of the whole series; that is, there may exist an 
unconditionally necessary being. This necessary being, as the
intelligible condition of the series, would not belong to it as a
member, not even as the highest member of it, nor would it
render any member of the series empirically unconditioned.
The whole sensible world, so far as regards the empirically
conditioned existence of all its various members, would be left
unaffected. This way of conceiving how an unconditioned
P 481
being may serve as the ground of appearance differs from that
which we followed in the preceding subsection, in dealing with
the empirically unconditioned causality of freedom. For there
the thing itself was as cause (substantia phaenomenon) 
conceived to belong to the series of conditions, and only its
causality was thought as intelligible. Here, on the other hand,
the necessary being must be thought as entirely outside the
series of the sensible world (as ens extramundanum), and as
purely intelligible. In no other way can it be secured against
the law which renders all appearances contingent and 
dependent. 
The regulative principle of reason, so far as it bears upon
our present problem, is therefore this, that everything in the
sensible world has an empirically conditioned existence, and
that in no one of its qualities can it be unconditionally necessary;
that for every member in the series of conditions we must
expect, and as far as possible seek, an empirical condition in
some possible experience; and that nothing justifies us in
deriving an existence from a condition outside the empirical
series or even in regarding it in its place within the series as
absolutely independent and self-sufficient. At the same time
this principle does not in any way debar us from recognising
that the whole series may rest upon some intelligible being
that is free from all empirical conditions and itself contains
the ground of the possibility of all appearances.
In these remarks we have no intention of proving the 
unconditionally necessary existence of such a being, or even of
establishing the possibility of a purely intelligible condition of
the existence of appearances in the sensible world. Just as, on
the one hand, we limit reason, lest in leaving the guiding-
thread of the empirical conditions it should go straying into
the transcendent, adopting grounds of explanation that are
incapable of any representation in concreto, so, on the other
hand, we limit the law of the purely empirical employment of
the understanding, lest it should presume to decide as to the
possibility of things in general, and should declare the 
intelligible to be impossible, merely on the ground that it is
not of any use in explaining appearances. Thus all that we
have shown is that the thoroughgoing contingency of all
natural things, and of all their empirical conditions, is quite
P 482
consistent with the optional assumption of a necessary, though
purely intelligible, condition; and that as there is no real 
contradiction between the two assertions, both may be true. Such
an absolutely necessary being, as conceived by the understanding,
may be in itself impossible, but this can in no wise
be inferred from the universal contingency and dependence of
everything belonging to the sensible world, nor from the principle
which interdicts us from stopping at any one of its 
contingent members and from appealing to a cause outside the
world. Reason proceeds by one path in its empirical use, and
by yet another path in its transcendental use.
The sensible world contains nothing but appearances, and
these are mere representations which are always sensibly 
conditioned; in this field things in themselves are never objects to
us. It is not therefore surprising that in dealing with a member
of the empirical series, no matter what member it may be, we
are never justified in making a leap out beyond the context
of sensibility. To do so is to treat the appearances as if they
were things in themselves which exist apart from their 
transcendental ground, and which can remain standing while we
seek an outside cause of their existence. This certainly would
ultimately be the case with contingent things, but not with
mere representations of things, the contingency of which is
itself merely phenomenon, and can lead to no other regress
than that which determines the phenomena, that is, solely to
the empirical regress. On the other hand, to think an intelligible
ground of the appearances, that is, of the sensible world,
and to think it as free from the contingency of appearances,
does not conflict either with the unlimited empirical regress in
the series of appearances nor with their thoroughgoing 
contingency. That, indeed, is all that we had to do in order to
remove the apparent antinomy; and it can be done in this way
only. If for everything conditioned in its existence the condition
is always sensible, and therefore belongs to the series,
it must itself in turn be conditioned, as we have shown in the
antithesis of the fourth antinomy. Either, therefore, reason
through its demand for the unconditioned must remain in
conflict with itself, or this unconditioned must be posited outside
the series, in the intelligible. Its necessity will not then
P 483
require, or allow of, any empirical condition; so far as 
appearances are concerned, it will be unconditionally necessary.
The empirical employment of reason, in reference to the
conditions of existence in the sensible world, is not affected by
the admission of a purely intelligible being; it proceeds, in
accordance with the principle of thoroughgoing contingency,
from empirical conditions to higher conditions which are
always again empirical. But it is no less true, when what we
have in view is the pure employment of reason, in reference
to ends, that this regulative principle does not exclude the
assumption of an intelligible cause which is not in the series.
For the intelligible cause then signifies only the purely 
transcendental and to us unknown ground of the possibility of the
sensible series in general. Its existence as independent of all
sensible conditions and as in respect of these conditions 
unconditionally necessary, is not inconsistent with the unlimited
contingency of appearances, that is to say, with the never-
ending regress in the series of empirical conditions.
Concluding Note on the whole Antinomy of Pure Reason.
So long as reason, in its concepts, has in view simply the
totality of conditions in the sensible world, and is considering
what satisfaction in this regard it can obtain for them, our
ideas are at once transcendental and cosmological. Immediately,
however, the unconditioned (and it is with this that we
are really concerned) is posited in that which lies entirely outside
the sensible world, and therefore outside all possible experience,
the ideas become transcendent. They then no longer serve
only for the completion of the empirical employment of reason
-- an idea [of completeness] which must always be pursued,
though it can never be completely achieved. On the contrary,
they detach themselves completely from experience, and make
for themselves objects for which experience supplies no
material, and whose objective reality is not based on completion
of the empirical series but on pure a priori concepts. Such
transcendent ideas have a purely intelligible object; and this
object may indeed be admitted as a transcendental object, but
only if we likewise admit that, for the rest, we have no knowledge
P 484
in regard to it, and that it cannot be thought as a determinate
thing in terms of distinctive inner predicates. As it is
independent of all empirical concepts, we are cut off from any
reasons that could establish the possibility of such an object,
and have not the least justification for assuming it. It is a mere
thought-entity. Nevertheless the cosmological idea which has
given rise to the fourth antinomy impels us to take this step. For
the existence of appearances, which is never self-grounded but
always conditioned, requires us to look around for something
different from all appearances, that is, for an intelligible object
in which this contingency may terminate. But once we have
allowed ourselves to assume a self-subsistent reality entirely
outside the field of sensibility, appearances can only be viewed
as contingent modes whereby beings that are themselves intelligences
represent intelligible objects. Consequently, the only
resource remaining to us is the use of analogy, by which we
employ the concepts of experience in order to form some
sort of concept of intelligible things -- things of which as
they are in themselves we have yet not the least knowledge.
Since the contingent is not to be known save through 
experience, and we are here concerned with things which are
not to be in any way objects of experience, we must derive
the knowledge of them from that which is in itself necessary,
that is, from pure concepts of things in general. Thus the
very first step which we take beyond the world of sense
obliges us, in seeking for such new knowledge, to begin with
an enquiry into absolutely necessary being, and to derive from
the concepts of it the concepts of all things in so far as they
are purely intelligible. This we propose to do in the next
chapter.

Table of Contents Previous Section Next Section Search Engine