ponent, must prove fatal also to his own speculative
structures, if he should wish to erect such. But this
need not disturb him, because he does not wish to
shelter himself beneath them, but looks out for the
fair field of practical philosophy, where he may hope
to find firmer ground for erecting his own rational
and beneficial system.

There is, therefore, no room for real polemic in the
sphere of pure reason. Both parties beat the air
and fight with their own shadows, because they go
beyond the limits of nature, where there is nothing
that they could lay hold of with their dogmatical
grasp. They may fight to their hearts' content, the
shadows which they are cleaving grow together again
in one moment, like the heroes in Valhalla, in order
to disport themselves once more in these bloodless
contests.

Nor can we admit a sceptical use of pure reason,
which might be called the principle of neutrality in
all its disputes. Surely to stir up reason against
itself, to supply it with weapons on both sides, and
then to look on quietly and scoffingly while the fierce
battle is raging, does not look well from a dogmatical
point of view, but has the appearance of a mischievous
and malevolent disposition. If, however, we consider
the invincible obstinacy and the boasting of [p. 757]



DISCIPLINE OF PURE REASON. 649

the dogmatical sophists, who are deaf to all the warn-
ings of criticism, there really seems nothing left but
to meet the boasting on one side by an equally justi-
fied boasting on the other, in order at least to startle
reason by a display of opposition, and thus to shake
her confidence and make her willing to listen to the
voice of criticism. But to stop at this point, and
to look upon the conviction and confession of ignor-
ance, not only as a remedy against dogmatical con-
ceit, but as the best means of settling the conflict of
reason with herself, is a vain attempt that will never
give rest and peace to reason. The utmost it can do
is to rouse reason from her sweet dogmatical dreams,
and to induce her to' examine more carefully her own
position. As, however, the sceptical manner of avoid-
ing a troublesome business seems to be the shortest
way out of all difficulties, and promises to lead to a
permanent peace in philosophy, or is chosen at least
as the highroad by all who, under the pretence of a
scornful dislike of investigations of this kind, try to
give themselves the air of philosophers, it seems neces-
sary to exhibit this mode of thought in its true light.

The Impossibility of a sceptical satisfaction of [p. 758]
Pure Reason in conflict with itself.
The consciousness of my ignorance (unless we re-
cognise at the same time its necessity) ought, instead
of forming the end of my investigations, to serve, on
the contrary, as their strongest impulse. All ignorance
is either an ignorance of things, or an ignorance of
the limits of our cognition. If ignorance is accidental,
it should incite us, in the former case, to investigate



65O DISCIPLINE OF PURE REASON.

things dogmatically, in the latter to investigate the
limits of possible knowledge critically. That my ig-
norance is absolutely necessary and that I am absolved
from the duty of all further investigation, can never
be established empirically by mere observation, but
critically only, by a thorough, examination of the first
sources of our knowledge. The determination of the
true limits of our reason can be made on a priori
grounds only, while its limitation, which consists in a
general recognition of our never entirely removable
ignorance, may be realised a posteriori also, by seeing
how much remains to be known in spite of all that
can be known. The former knowledge of our ignorance,
possible only by criticism of reason, is truly scientific,
the latter is merely matter of experience, where [p. 759]
it is never possible to say, how far the inferences
drawn from it may reach. If I regard the earth, ac-
cording to the evidence of my senses, as a flat surface,
I cannot tell how far it may extend. But what experi-
ence teaches me is, that wheresoever I go, I always
see before me a space in which I can proceed further.
Thus I am conscious of the limits of my actual know-
ledge of the earth at any given moment, but not of
the limits of all possible geography. But if I have
got so far as to know that the earth is a sphere and
its surface spherical, I am able from any small portion
of it, for instance, from a degree, to know definitely
and according to principles a priori, the diameter, and
through it, the complete periphery of the earth ; and,
though I am ignorant with regard to the objects
which are contained in that surface, I am not so with
regard to its extent, its magnitude, and its limits.



DISCIPLINE OF PURE REASON- 65 1

In a similar manner the whole of the objects of our
knowledge appears to us like a level surface, with its
apparent horizon which encircles its whole extent,
and was called by us the idea of unconditioned
totality. To reach this limit empirically is impos-
sible, and all attempts have proved vain to determine
it a priori according to a certain principle. Never-
theless, all questions of pure reason refer to what lies
outside of that horizon, or, it may be, on its boundary
line. [p. 760]

The celebrated David Hume was one of those
geographers of human reason who supposed that all
those questions were sufficiently disposed of by being
relegated outside that horizon, which, however, he
was not able to determine. He was chiefly occupied
with the principle of causality, and remarked quite
rightly, that the truth of this principle (and even
the objective validity of the concept of an efficient
cause in general) was based on no knowledge, i. e. on
no cognition a priori, and that its authority rested
by no means on the necessity of such a law, but
merely on its general usefulness in experience, and
on a kind of subjective necessity arising from thence,
which he called habit. From the inability of reason
to employ this principle beyond the limits of ex-
perience he inferred the nullity of all the pretensions
of reason in her attempts to pass beyond what is
empirical.

This procedure of subjecting the facts of reason to
examination, and, if necessary, to blame, may be
termed the censorship of reason. There can be no
doubt that such a censorship must inevitably lead to



652 DISCIPLINE OF PURE REASON.

doubts against all the transcendent employ- [p. 761]
ment of such principles. But this is only the second
and by no means the last step in our inquiry. The
first step in matters of pure reason, which marks its
infancy, is dogmatism. The second, which we have
just described, is scepticism, and marks the stage of
caution on the part of reason, when rendered wiser by
experience. But a third step is necessary, that of the
maturity and manhood of judgment, based on firm
and universally applicable maxims, when not the
facts of reason, but reason itself in its whole power
and fitness for pure knowledge a priori comes to be
examined. This is not the censura merely, but the
true criticism of reason, by which not the barrier only,
but the fixed frontiers of reason, not ignorance only
on this or that point, but ignorance with reference
to all possible questions of a certain kind, must be
proved from principles, instead of being merely guessed
at. Thus scepticism is a resting-place of reason, where
it may reflect for a time on its dogmatical wanderings
and gain a survey of the region where it happens to
be, in order to choose its way with greater certainty
for the future : but it can never be its permanent
dwelling-place. That can only be found in perfect
certainty, whether of our knowledge of the objects
themselves or of the limits within which all our
knowledge of objects are enclosed. [p. 762]

Our reason is not to be considered as an inde-
finitely extended plain, the limits of which are
known in a general way only, but ought rather to
be compared to a sphere the radius of which may
be determined from the curvature of the arc of its



DISCIPLINE OF PURE REASON. 653

surface (corresponding to the nature of synthetical
propositions a priori), which enables us likewise to
fix the extent and periphery of it with perfect cer-
tainty. Outside that sphere (the field of experience)
nothing can become an object to our reason, nay,
questions even on such imaginary objects relate to the
subjective principles only for a complete determina-
tion of all the relations which may exist between the
concepts of the understanding within that sphere.

It is a fact that we are in possession of different
kinds of synthetical knowledge a priori, as shown by
the principles of the understanding which anticipate
experience. If any body finds it quite impossible to
understand the possibility of such principles, he may
at first have some doubts as to whether they really
dwell within us a priori ; but he cannot thus, by the
mere powers of the understanding, prove their im-
possibility, and declare all the steps which reason
takes under their guidance as null and void. All he
can say is that, if we could understand their origin
and genuineness, we should be able to determine the
extent and limits of our reason, and that, until that is
done, all the assertions of reason are made [p. 763]
at random. And in this way a complete scepticism
with regard to all dogmatical philosophy, which is not
guided by a criticism of reason, is well grounded,
though we could not therefore deny to reason such
further advance, after the way has once been prepared
and secured on firmer ground. For all these con-
cepts, nay, all the questions which pure reason places
before us, have their origin, not in experience, but in
reason itself, and must therefore be capable of being



654 DISCIPLINE OF PURE REASON.

solved and tested as to their validity or invalidity.
Nor are we justified, admitting that the solution of
these problems is really to be found in the nature
of things, to decline their consideration and further
investigation, under the pretext of our weakness, for
reason alone begets all these ideas by itself, and is
bound therefore to give an account of their validity
or their dialectical vanity.

All sceptical polemic should properly be directed
against the dogmatist only who, without any mis-
givings about his own fundamental objective prin-
ciples, that is, without criticism, continues his course
with undisturbed gravity, and cannot be brought
to a proper self-knowledge unless we unsettle his
brief. With regard to what we know or what we
cannot know, that polemic is of no consequence what-
ever. All the unsuccessful dogmatical attempts of
reason are facta, and it is always useful to [p. 764]
submit them to the censura of the sceptic. But this
can decide nothing as to the expectations of reason in
her hopes and claims of a better success in future
attempts ; and no mere censura can put an end to
the disputes regarding the rights of human reason.

Hume is, perhaps, the most ingenious of all sceptics,
and without doubt the most important with regard
to the influence which the sceptical method may
exercise in awakening reason to a thorough exami-
nation of its rights. It will therefore be worth our
while to make clear to ourselves the course of his
reasoning and the errors of an intelligent and es-
timable man, who at the outset of his enquiries was
certainly on the right track of truth.



DISCIPLINE OF PURE REASON. 655

Hume was probably aware, though he never made
it quite clear to himself, that in judgments of a cer-
tain kind we pass beyond our concept of the object.
I have called this class of judgments synthetical. There
is no difficulty as to how I may, by means of ex-
perience, pass beyond the concept which I have
hitherto had. Experience is itself such a synthesis
of perceptions through which a concept, which I
have by means of one perception, is increased by
means of other perceptions. But we imagine that
we are able also a priori to pass beyond [p. 765]
our concept and thus to enlarge our knowledge. This
we attempt to do either by the pure understanding,
in relation to that which can at least be an object of
experience, or even by means of pure reason, in re-
lation to such qualities of things, or even the ex-
istence of such things, as can never occur in expe-
rience. Hume in his scepticism did not distinguish
between these two kinds of judgments as he ought
to have done, but regarded more particularly this
augmentation of concepts by themselves, and, so
to say, the spontaneous generation of our under-
standing (and of our reason), without being impreg-
nated by experience, as impossible. Considering all
principles a priori as imaginary, he arrived at the
conclusion that they were nothing but a habit aris-
ing from experience and its laws ; that they were
therefore merely empirical, that is, in themselves,
contingent rules to which we wrongly ascribe ne-
cessity and universality. In order to establish
this strange proposition, he appealed to the generally
admitted principle of the relation between cause



656 DISCIPLINE OF PURE REASON.

and effect. For as no faculty of the understanding
could lead us from the concept of a thing to the
existence of something else that should follow from
it universally and necessarily, he thought himself
justified in concluding that, without experience, we
have nothing that could augment our concept and
give us a right to form a judgment that extends
itself a priori. That the light of the sun which shines
on the wax should melt the wax and at the [p. 7 66 ]
same time harden the clay, no understanding, he main-
tained, could guess from the concepts which we had
before of these things, much less infer, according to
a law, experience only being able to teach us such
a law. We have seen, on the contrary, in the tran-
scendental logic that, though we can never pass im-
mediately beyond the content of a concept that is
given us, we are nevertheless able, entirely a priori,
but yet in reference to something else, namely pos-
sible experience, to know the law of its connection
with other things. If, therefore, wax, which was
formerly hard, melts, I can know a priori that some-
thing else must have preceded (for instance the heat
of the sun) upon which this melting has followed
according to a permanent law, although without ex-
perience I could never know a priori either from
the effect the cause, or from the cause the effect.
Hume was therefore wrong in inferring from the
mere contingency of our being determined according
to the law of causality, the contingency of that law
itself, and he mistook our passing beyond the con-
cept of a thing to some possible experience (which is
entirely a priori and constitutes the objective reality of



DISCIPLINE OF PURE REASON. 657

it) for the synthesis of the objects of real experience
which, no doubt, is always empirical. He thus
changed a principle of affinity which resides in the
understanding and predicates necessary connection,
into a rule of association residing in the imitative
faculty of imagination, which can only re- [p. 7 6 7]
present contingent, but never objective connections.

The sceptical errors of that otherwise singularly
acute thinker arose chiefly from a defect, which he
shared, however, in common with all dogmatists,
namely of not having surveyed systematically all
kinds of synthesis a priori of the understanding. For
in doing this he would, without mentioning others,
have discovered, for instance, the principle of per-
manency as one which, like causality, anticipates
experience. He would thus have been able also to
fix definite limits to the understanding in its at-
tempts at expansion a priori, and to pure reason.
He only narrows the sphere of our understanding,
without definitely limiting it, and produces a general
mistrust, but no definite knowledge of that ignorance
which to us is inevitable. He only subjects certain
principles of the understanding to his censura, but
does not place the understanding, with reference
to all its faculties, on the balance of criticism. He
is not satisfied with denying to the understanding
what in reality it does not possess, but goes on to
deny to it all power of expanding a priori, though
he has never really tested all its powers. For this
reason, what always defeats scepticism has happened
to Hume also, namely, that he himself becomes subject
to scepticism, because his objections rest on facts

vol. 11. u u



658 DISCIPLINE OF PURE REASON.

only which are contingent, and not on principles
which alone can force a surrender of the [p. 768]
right of dogmatical assertion.

As, besides this, he does not sufficiently distinguish
between the well-grounded claims of the understand-
ing and the dialectical pretensions of reason, against
which, however, his attacks are chiefly directed, it
so happens that reason, the peculiar tendency of
which has not in the least been destroyed, but only
checked, does not at all consider itself shut out from
its attempts at expansion, and can never be en-
tirely turned away from them, although it may be
punished now and then. Mere attacks only provoke
counter attacks, and make us more obstinate in en-
forcing our own views. But a complete survey of
all that is really our own, and the conviction of a
certain though a small possession, make us perceive
the vanity of higher claims, and induce us, after
surrendering all disputes, to live contentedly and
peacefully within our own limited, but undisputed
domain.

These sceptical attacks are not only dangerous,
but even destructive to the uncritical dogmatist
who has not measured the sphere of his under-
standing, and has not, therefore, determined, accord-
ing to principles, the limits of his own possible
knowledge, and does not know beforehand how much
he is really able to achieve, but thinks that he is
able to find, all this out by a purely tentative method.
For if he has been found out in one single assertion
of his, which he cannot justify, or the fallacy [p. 769]
of which he cannot evolve according to principles,



DISCIPLINE OF PURE REASON. 659

suspicion falls on all his assertions, however plausible
they may appear.

And thus the sceptic is the true schoolmaster to
lead the dogmatic speculator towards a sound criti-
cism of the understanding and of reason. When he
has once been brought there, he need fear no further
attacks, for he has learnt to distinguish his own pos-
session from that which lies completely beyond it, and
on which he can lay no claim, nor become involved
in any disputes regarding it. Thus the sceptical
method, though it cannot in itself satisfy with regard
to the problems of reason, is nevertheless an excel-
lent preparation in order to awaken its circumspec-
tion, and to indicate the true means whereby the
legitimate possessions of reason may be secured
against all attacks.

DISCIPLINE OF PURE REASON.

Section III.

The Discipline of Pure Reason with regard to
Hypotheses.

As then the criticism of our reason has at last
taught us so much at least, that with its pure and
speculative use we can arrive at no knowledge at all,
would not this seem to open a wide field for hypo-
theses, as, where we cannot assert with certainty,
we are at all events at liberty to form guesses and
opinions %

If the faculty of imagination is not simply [p. 770]
to indulge in dreams, but to invent and compose

u u 2



660 DISCIPLINE OP PUKE REASON.

under the strict surveillance of reason, it is necessary
that there should always be something perfectly
certain, and not only invented or resting on opinion,
and that is the possibility of the object itself. If
that is once given, it is then allowable, so far as its
reality is concerned, to have recourse to opinion,
which opinion, however, if it is not to be utterly
groundless, must be brought in connection with what
is really given and therefore certain, as its ground of
explanation. In that case, and in that case only, can
we speak of an hypothesis.

As we cannot form the least conception of the pos-
sibility of a dynamical connection a priori, and as the
categories of the pure understanding are not intended
to invent any such connection, but only, when it is
given in experience, to understand it, we cannot by
means of these categories invent one single object as
endowed with a new quality not found in experience,
or base any permissible hypothesis on such a quality;
otherwise we should be supplying our reason with
empty chimeras, and not with concepts of things.
Thus it is not permissible to invent any new and
original powers, as, for instance, an understanding
capable of perceiving objects without the aid of the
senses ; or a force of attraction without any contact ;
a new kind of substances that should exist, for in-
stance, in space, without being impenetrable, and
consequently, also, any connection of substances, dif-
ferent from that which is supplied by expe- [p. 771]
rience ; no presence, except in space, no duration, ex-
cept in time. In one word, our reason can only use
the conditions of possible experience as the conditions



DISCIPLINE OF PURE REASON. 66 1

of the possibility of things ; it cannot invent them in-
dependently, because such concepts, although not self-
contradictory, would always be without an object.

The concepts of reason, as was said before, are
mere ideas, and it is true that they have no object
corresponding to them in experience ; but they do not,
for all that, refer to purely imaginary objects, which
are supposed to be possible. They are purely pro-
blematical, in order to supply (as heuristic fictions)
regulative principles for the systematical employment
of the understanding in the sphere of experience. If
they are not that, they would become mere fictions
the possibility of which is quite indemonstrable, and
which, therefore, can never be employed as hypotheses
for the explanation of real phenomena. It is quite
permissible to represent the soul to ourselves as
simple, in order, according to this idea, to use the
complete and necessary unity of all the faculties of
the soul, although we cannot understand it in con-
creto, as the principle of all our inquiries into its
internal phenomena. But to assume the soul as a
simple substance (which is a transcendent concept)
would be a proposition, not only indemonstrable (this
is the case with several physical hypotheses), [p. 77 2 ]
but purely arbitrary and rash : because the simple
can never occur in any experience, and if by sub-
stance we understand the permanent object of sensu-
ous intuition, the very possibility of a simple pheno-
menon is perfectly inconceivable. Reason has no right
whatever to assume, as an opinion, purely intelligible
beings, or purely intelligible qualities of the objects
of the senses ; although, on the other side, as we



662 DISCIPLINE OF PURE REASON.

have no concepts whatever, either of their possibility
or impossibility, we cannot claim any truer insight
enabling us to deny dogmatically their possibility.

In order to explain given phenomena, no other
things or reasons can be adduced but those which,
according to the already known laws of phenomena,
have been put in connection with them. A tran-
scendental hypothesis, adducing a mere idea of reason
for the explanation of natural things, would there-
fore be no explanation at all, because it would really
be an attempt at explaining what, according to known
empirical principles, we do not understand sufficiently
by something which we do not understand at all.
Nor would the principle of such an hypothesis serve
to help the understanding with regard to its objects,
but only to satisfy our reason. Order and design in na-
ture must themselves be explained on natural grounds
and according to natural laws ; and for this [p. 773]
purpose even the wildest hypotheses, if only they
are physical, are more tolerable than a hyperphysical
one, that is, the appeal to the Divine Author, who is
called in for that very purpose. This would be a
principle of ratio ignava, to pass by all causes the
objective reality of which, in their possibility at least,
may be known by continued experience, in order to
rest on a mere idea, which no doubt is very agreeable
to our reason. With regard to the absolute totality
of the explanation as applied to the series of causes,
there can be no difficulty, considering that all mun-
dane objects are nothing but phenomena, in which
we can never hope to find absolute completeness in
the synthesis of the series of conditions.